2002-10-12 11:37:38 +00:00
|
|
|
/**
|
|
|
|
* $Id$
|
|
|
|
*
|
2008-01-07 19:13:47 +00:00
|
|
|
* ***** BEGIN GPL LICENSE BLOCK *****
|
2002-10-12 11:37:38 +00:00
|
|
|
*
|
|
|
|
* This program is free software; you can redistribute it and/or
|
|
|
|
* modify it under the terms of the GNU General Public License
|
|
|
|
* as published by the Free Software Foundation; either version 2
|
2008-01-07 19:13:47 +00:00
|
|
|
* of the License, or (at your option) any later version.
|
2002-10-12 11:37:38 +00:00
|
|
|
*
|
|
|
|
* This program is distributed in the hope that it will be useful,
|
|
|
|
* but WITHOUT ANY WARRANTY; without even the implied warranty of
|
|
|
|
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
|
|
|
|
* GNU General Public License for more details.
|
|
|
|
*
|
|
|
|
* You should have received a copy of the GNU General Public License
|
|
|
|
* along with this program; if not, write to the Free Software Foundation,
|
2010-02-12 13:34:04 +00:00
|
|
|
* Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
|
2002-10-12 11:37:38 +00:00
|
|
|
*
|
|
|
|
* The Original Code is Copyright (C) 2001-2002 by NaN Holding BV.
|
|
|
|
* All rights reserved.
|
|
|
|
*
|
|
|
|
* The Original Code is: all of this file.
|
|
|
|
*
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
* Contributor(s): 2007, Joshua Leung, major recode
|
2002-10-12 11:37:38 +00:00
|
|
|
*
|
2008-01-07 19:13:47 +00:00
|
|
|
* ***** END GPL LICENSE BLOCK *****
|
2002-10-12 11:37:38 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
#include <stdio.h>
|
2008-01-20 23:53:13 +00:00
|
|
|
#include <stddef.h>
|
2002-10-12 11:37:38 +00:00
|
|
|
#include <string.h>
|
2003-12-24 11:44:57 +00:00
|
|
|
#include <math.h>
|
Merge of trunk into blender 2.5:
svn merge https://svn.blender.org/svnroot/bf-blender/trunk/blender -r12987:17416
Issues:
* GHOST/X11 had conflicting changes. Some code was added in 2.5, which was
later added in trunk also, but reverted partially, specifically revision
16683. I have left out this reversion in the 2.5 branch since I think it is
needed there.
http://projects.blender.org/plugins/scmsvn/viewcvs.php?view=rev&root=bf-blender&revision=16683
* Scons had various conflicting changes, I decided to go with trunk version
for everything except priorities and some library renaming.
* In creator.c, there were various fixes and fixes for fixes related to the -w
-W and -p options. In 2.5 -w and -W is not coded yet, and -p is done
differently. Since this is changed so much, and I don't think those fixes
would be needed in 2.5, I've left them out.
* Also in creator.c: there was code for a python bugfix where the screen was not
initialized when running with -P. The code that initializes the screen there
I had to disable, that can't work in 2.5 anymore but left it commented as a
reminder.
Further I had to disable some new function calls. using src/ and python/, as
was done already in this branch, disabled function calls:
* bpath.c: error reporting
* BME_conversions.c: editmesh conversion functions.
* SHD_dynamic: disabled almost completely, there is no python/.
* KX_PythonInit.cpp and Ketsji/ build files: Mathutils is not there, disabled.
* text.c: clipboard copy call.
* object.c: OB_SUPPORT_MATERIAL.
* DerivedMesh.c and subsurf_ccg, stipple_quarttone.
Still to be done:
* Go over files and functions that were moved to a different location but could
still use changes that were done in trunk.
2008-11-12 21:16:53 +00:00
|
|
|
#include <float.h>
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
#include "MEM_guardedalloc.h"
|
|
|
|
|
|
|
|
#include "BLI_blenlib.h"
|
2010-04-26 03:42:38 +00:00
|
|
|
#include "BLI_listbase.h"
|
2009-11-10 20:43:45 +00:00
|
|
|
#include "BLI_math.h"
|
2009-03-30 07:28:37 +00:00
|
|
|
#include "BLI_editVert.h"
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
#include "DNA_armature_types.h"
|
|
|
|
#include "DNA_constraint_types.h"
|
2009-04-20 15:06:46 +00:00
|
|
|
#include "DNA_modifier_types.h"
|
2002-10-12 11:37:38 +00:00
|
|
|
#include "DNA_object_types.h"
|
|
|
|
#include "DNA_action_types.h"
|
|
|
|
#include "DNA_curve_types.h"
|
2008-12-31 17:11:42 +00:00
|
|
|
#include "DNA_mesh_types.h"
|
2007-08-17 11:23:48 +00:00
|
|
|
#include "DNA_meshdata_types.h"
|
|
|
|
#include "DNA_lattice_types.h"
|
2004-01-27 06:08:37 +00:00
|
|
|
#include "DNA_scene_types.h"
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
#include "DNA_text_types.h"
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
#include "BKE_utildefines.h"
|
|
|
|
#include "BKE_action.h"
|
2007-08-28 08:43:38 +00:00
|
|
|
#include "BKE_anim.h" /* for the curve calculation part */
|
2002-10-12 11:37:38 +00:00
|
|
|
#include "BKE_armature.h"
|
|
|
|
#include "BKE_blender.h"
|
|
|
|
#include "BKE_constraint.h"
|
2005-12-21 17:49:43 +00:00
|
|
|
#include "BKE_displist.h"
|
2007-08-17 11:23:48 +00:00
|
|
|
#include "BKE_deform.h"
|
2007-08-28 08:43:38 +00:00
|
|
|
#include "BKE_DerivedMesh.h" /* for geometry targets */
|
|
|
|
#include "BKE_cdderivedmesh.h" /* for geometry targets */
|
2002-10-12 11:37:38 +00:00
|
|
|
#include "BKE_object.h"
|
|
|
|
#include "BKE_ipo.h"
|
|
|
|
#include "BKE_global.h"
|
|
|
|
#include "BKE_library.h"
|
2007-06-18 07:41:21 +00:00
|
|
|
#include "BKE_idprop.h"
|
2009-04-20 15:06:46 +00:00
|
|
|
#include "BKE_shrinkwrap.h"
|
|
|
|
#include "BKE_mesh.h"
|
2007-06-18 07:41:21 +00:00
|
|
|
|
2008-10-28 18:47:13 +00:00
|
|
|
#ifndef DISABLE_PYTHON
|
2007-06-18 07:41:21 +00:00
|
|
|
#include "BPY_extern.h"
|
2008-10-28 18:47:13 +00:00
|
|
|
#endif
|
2002-10-12 11:37:38 +00:00
|
|
|
|
|
|
|
#ifndef M_PI
|
|
|
|
#define M_PI 3.14159265358979323846
|
|
|
|
#endif
|
2005-03-09 19:45:59 +00:00
|
|
|
|
2002-10-12 11:37:38 +00:00
|
|
|
|
Result of 2 weeks of quiet coding work in Greece :)
Aim was to get a total refresh of the animation system. This
is needed because;
- we need to upgrade it with 21st century features
- current code is spaghetti/hack combo, and hides good design
- it should become lag-free with using dependency graphs
A full log, with complete code API/structure/design explanation
will follow, that's a load of work... so here below the list with
hot changes;
- The entire object update system (matrices, geometry) is now
centralized. Calls to where_is_object and makeDispList are
forbidden, instead we tag objects 'changed' and let the
depgraph code sort it out
- Removed all old "Ika" code
- Depgraph is aware of all relationships, including meta balls,
constraints, bevelcurve, and so on.
- Made depgraph aware of relation types and layers, to do smart
flushing of 'changed' events. Nothing gets calculated too often!
- Transform uses depgraph to detect changes
- On frame-advance, depgraph flushes animated changes
Armatures;
Almost all armature related code has been fully built from scratch.
It now reveils the original design much better, with a very clean
implementation, lag free without even calculating each Bone more than
once. Result is quite a speedup yes!
Important to note is;
1) Armature is data containing the 'rest position'
2) Pose is the changes of rest position, and always on object level.
That way more Objects can use same Pose. Also constraints are in Pose
3) Actions only contain the Ipos to change values in Poses.
- Bones draw unrotated now
- Drawing bones speedup enormously (10-20 times)
- Bone selecting in EditMode, selection state is saved for PoseMode,
and vice-versa
- Undo in editmode
- Bone renaming does vertexgroups, constraints, posechannels, actions,
for all users of Armature in entire file
- Added Bone renaming in NKey panel
- Nkey PoseMode shows eulers now
- EditMode and PoseMode now have 'active' bone too (last clicked)
- Parenting in EditMode' CTRL+P, ALT+P, with nice options!
- Pose is added in Outliner now, with showing that constraints are in
the Pose, not Armature
- Disconnected IK solving from constraints. It's a separate phase now,
on top of the full Pose calculations
- Pose itself has a dependency graph too, so evaluation order is lag free.
TODO NOW;
- Rotating in Posemode has incorrect inverse transform (Martin will fix)
- Python Bone/Armature/Pose API disabled... needs full recode too
(wait for my doc!)
- Game engine will need upgrade too
- Depgraph code needs revision, cleanup, can be much faster!
(But, compliments for Jean-Luc, it works like a charm!)
- IK changed, it now doesnt use previous position to advance to next
position anymore. That system looks nice (no flips) but is not well
suited for NLA and background render.
TODO LATER;
We now can do loadsa new nifty features as well; like:
- Kill PoseMode (can be option for armatures itself)
- Make B-Bones (Bezier, Bspline, like for spines)
- Move all silly button level edit to 3d window (like CTRL+I = add
IK)
- Much better & informative drawing
- Fix action/nla editors
- Put all ipos in Actions (object, mesh key, lamp color)
- Add hooks
- Null bones
- Much more advanced constraints...
Bugfixes;
- OGL render (view3d header) had wrong first frame on anim render
- Ipo 'recording' mode had wrong playback speed
- Vertex-key mode now sticks to show 'active key', until frame change
-Ton-
2005-07-03 17:35:38 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* ************************ Constraints - General Utilities *************************** */
|
|
|
|
/* These functions here don't act on any specific constraints, and are therefore should/will
|
|
|
|
* not require any of the special function-pointers afforded by the relevant constraint
|
|
|
|
* type-info structs.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* -------------- Naming -------------- */
|
|
|
|
|
|
|
|
/* Find the first available, non-duplicate name for a given constraint */
|
Result of 2 weeks of quiet coding work in Greece :)
Aim was to get a total refresh of the animation system. This
is needed because;
- we need to upgrade it with 21st century features
- current code is spaghetti/hack combo, and hides good design
- it should become lag-free with using dependency graphs
A full log, with complete code API/structure/design explanation
will follow, that's a load of work... so here below the list with
hot changes;
- The entire object update system (matrices, geometry) is now
centralized. Calls to where_is_object and makeDispList are
forbidden, instead we tag objects 'changed' and let the
depgraph code sort it out
- Removed all old "Ika" code
- Depgraph is aware of all relationships, including meta balls,
constraints, bevelcurve, and so on.
- Made depgraph aware of relation types and layers, to do smart
flushing of 'changed' events. Nothing gets calculated too often!
- Transform uses depgraph to detect changes
- On frame-advance, depgraph flushes animated changes
Armatures;
Almost all armature related code has been fully built from scratch.
It now reveils the original design much better, with a very clean
implementation, lag free without even calculating each Bone more than
once. Result is quite a speedup yes!
Important to note is;
1) Armature is data containing the 'rest position'
2) Pose is the changes of rest position, and always on object level.
That way more Objects can use same Pose. Also constraints are in Pose
3) Actions only contain the Ipos to change values in Poses.
- Bones draw unrotated now
- Drawing bones speedup enormously (10-20 times)
- Bone selecting in EditMode, selection state is saved for PoseMode,
and vice-versa
- Undo in editmode
- Bone renaming does vertexgroups, constraints, posechannels, actions,
for all users of Armature in entire file
- Added Bone renaming in NKey panel
- Nkey PoseMode shows eulers now
- EditMode and PoseMode now have 'active' bone too (last clicked)
- Parenting in EditMode' CTRL+P, ALT+P, with nice options!
- Pose is added in Outliner now, with showing that constraints are in
the Pose, not Armature
- Disconnected IK solving from constraints. It's a separate phase now,
on top of the full Pose calculations
- Pose itself has a dependency graph too, so evaluation order is lag free.
TODO NOW;
- Rotating in Posemode has incorrect inverse transform (Martin will fix)
- Python Bone/Armature/Pose API disabled... needs full recode too
(wait for my doc!)
- Game engine will need upgrade too
- Depgraph code needs revision, cleanup, can be much faster!
(But, compliments for Jean-Luc, it works like a charm!)
- IK changed, it now doesnt use previous position to advance to next
position anymore. That system looks nice (no flips) but is not well
suited for NLA and background render.
TODO LATER;
We now can do loadsa new nifty features as well; like:
- Kill PoseMode (can be option for armatures itself)
- Make B-Bones (Bezier, Bspline, like for spines)
- Move all silly button level edit to 3d window (like CTRL+I = add
IK)
- Much better & informative drawing
- Fix action/nla editors
- Put all ipos in Actions (object, mesh key, lamp color)
- Add hooks
- Null bones
- Much more advanced constraints...
Bugfixes;
- OGL render (view3d header) had wrong first frame on anim render
- Ipo 'recording' mode had wrong playback speed
- Vertex-key mode now sticks to show 'active key', until frame change
-Ton-
2005-07-03 17:35:38 +00:00
|
|
|
void unique_constraint_name (bConstraint *con, ListBase *list)
|
|
|
|
{
|
2010-02-18 10:14:49 +00:00
|
|
|
BLI_uniquename(list, con, "Const", '.', offsetof(bConstraint, name), sizeof(con->name));
|
2003-10-21 13:22:07 +00:00
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* ----------------- Evaluation Loop Preparation --------------- */
|
|
|
|
|
|
|
|
/* package an object/bone for use in constraint evaluation */
|
|
|
|
/* This function MEM_calloc's a bConstraintOb struct, that will need to be freed after evaluation */
|
2009-01-04 14:14:06 +00:00
|
|
|
bConstraintOb *constraints_make_evalob (Scene *scene, Object *ob, void *subdata, short datatype)
|
2003-10-21 13:22:07 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
bConstraintOb *cob;
|
|
|
|
|
|
|
|
/* create regardless of whether we have any data! */
|
|
|
|
cob= MEM_callocN(sizeof(bConstraintOb), "bConstraintOb");
|
2003-10-21 13:22:07 +00:00
|
|
|
|
2009-01-04 14:14:06 +00:00
|
|
|
/* for system time, part of deglobalization, code nicer later with local time (ton) */
|
|
|
|
cob->scene= scene;
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* based on type of available data */
|
|
|
|
switch (datatype) {
|
|
|
|
case CONSTRAINT_OBTYPE_OBJECT:
|
2007-06-18 07:41:21 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* disregard subdata... calloc should set other values right */
|
|
|
|
if (ob) {
|
|
|
|
cob->ob = ob;
|
|
|
|
cob->type = datatype;
|
2009-09-02 10:45:11 +00:00
|
|
|
cob->rotOrder = EULER_ORDER_DEFAULT; // TODO: when objects have rotation order too, use that
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(cob->matrix, ob->obmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
else
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(cob->matrix);
|
2007-06-18 07:41:21 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(cob->startmat, cob->matrix);
|
2007-06-18 07:41:21 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
|
|
|
case CONSTRAINT_OBTYPE_BONE:
|
2003-10-21 13:22:07 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* only set if we have valid bone, otherwise default */
|
|
|
|
if (ob && subdata) {
|
|
|
|
cob->ob = ob;
|
|
|
|
cob->pchan = (bPoseChannel *)subdata;
|
|
|
|
cob->type = datatype;
|
|
|
|
|
2009-09-02 10:45:11 +00:00
|
|
|
if (cob->pchan->rotmode > 0) {
|
|
|
|
/* should be some type of Euler order */
|
|
|
|
cob->rotOrder= cob->pchan->rotmode;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* Quats, so eulers should just use default order */
|
|
|
|
cob->rotOrder= EULER_ORDER_DEFAULT;
|
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* matrix in world-space */
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(cob->matrix, cob->pchan->pose_mat, ob->obmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
else
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(cob->startmat, cob->matrix);
|
2003-10-21 13:22:07 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
2007-05-17 06:45:16 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
default: /* other types not yet handled */
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(cob->matrix);
|
|
|
|
unit_m4(cob->startmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
return cob;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* cleanup after constraint evaluation */
|
|
|
|
void constraints_clear_evalob (bConstraintOb *cob)
|
|
|
|
{
|
|
|
|
float delta[4][4], imat[4][4];
|
|
|
|
|
|
|
|
/* prevent crashes */
|
|
|
|
if (cob == NULL)
|
|
|
|
return;
|
|
|
|
|
|
|
|
/* calculate delta of constraints evaluation */
|
2009-11-10 20:43:45 +00:00
|
|
|
invert_m4_m4(imat, cob->startmat);
|
|
|
|
mul_m4_m4m4(delta, imat, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* copy matrices back to source */
|
|
|
|
switch (cob->type) {
|
|
|
|
case CONSTRAINT_OBTYPE_OBJECT:
|
2005-09-07 00:11:39 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* cob->ob might not exist! */
|
|
|
|
if (cob->ob) {
|
|
|
|
/* copy new ob-matrix back to owner */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(cob->ob->obmat, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* copy inverse of delta back to owner */
|
2009-11-10 20:43:45 +00:00
|
|
|
invert_m4_m4(cob->ob->constinv, delta);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
case CONSTRAINT_OBTYPE_BONE:
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* cob->ob or cob->pchan might not exist */
|
|
|
|
if (cob->ob && cob->pchan) {
|
|
|
|
/* copy new pose-matrix back to owner */
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(cob->pchan->pose_mat, cob->matrix, cob->ob->imat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* copy inverse of delta back to owner */
|
2009-11-10 20:43:45 +00:00
|
|
|
invert_m4_m4(cob->pchan->constinv, delta);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* free tempolary struct */
|
|
|
|
MEM_freeN(cob);
|
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* -------------- Space-Conversion API -------------- */
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
|
|
|
/* This function is responsible for the correct transformations/conversions
|
|
|
|
* of a matrix from one space to another for constraint evaluation.
|
|
|
|
* For now, this is only implemented for Objects and PoseChannels.
|
|
|
|
*/
|
2007-08-14 11:19:35 +00:00
|
|
|
void constraint_mat_convertspace (Object *ob, bPoseChannel *pchan, float mat[][4], short from, short to)
|
2002-10-12 11:37:38 +00:00
|
|
|
{
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
float tempmat[4][4];
|
|
|
|
float diff_mat[4][4];
|
|
|
|
float imat[4][4];
|
|
|
|
|
|
|
|
/* prevent crashes in these unlikely events */
|
|
|
|
if (ob==NULL || mat==NULL) return;
|
|
|
|
/* optimise trick - check if need to do anything */
|
|
|
|
if (from == to) return;
|
|
|
|
|
|
|
|
/* are we dealing with pose-channels or objects */
|
|
|
|
if (pchan) {
|
|
|
|
/* pose channels */
|
|
|
|
switch (from) {
|
|
|
|
case CONSTRAINT_SPACE_WORLD: /* ---------- FROM WORLDSPACE ---------- */
|
|
|
|
{
|
|
|
|
/* world to pose */
|
2009-11-10 20:43:45 +00:00
|
|
|
invert_m4_m4(imat, ob->obmat);
|
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(mat, tempmat, imat);
|
2007-07-28 09:00:15 +00:00
|
|
|
|
2008-09-07 04:43:31 +00:00
|
|
|
/* use pose-space as stepping stone for other spaces... */
|
|
|
|
if (ELEM(to, CONSTRAINT_SPACE_LOCAL, CONSTRAINT_SPACE_PARLOCAL)) {
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
/* call self with slightly different values */
|
|
|
|
constraint_mat_convertspace(ob, pchan, mat, CONSTRAINT_SPACE_POSE, to);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case CONSTRAINT_SPACE_POSE: /* ---------- FROM POSESPACE ---------- */
|
|
|
|
{
|
|
|
|
/* pose to world */
|
|
|
|
if (to == CONSTRAINT_SPACE_WORLD) {
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(mat, tempmat, ob->obmat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
/* pose to local */
|
|
|
|
else if (to == CONSTRAINT_SPACE_LOCAL) {
|
|
|
|
if (pchan->bone) {
|
2007-07-28 10:44:03 +00:00
|
|
|
if (pchan->parent) {
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
float offs_bone[4][4];
|
|
|
|
|
|
|
|
/* construct offs_bone the same way it is done in armature.c */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m3(offs_bone, pchan->bone->bone_mat);
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(offs_bone[3], pchan->bone->head);
|
2007-07-28 10:44:03 +00:00
|
|
|
offs_bone[3][1]+= pchan->bone->parent->length;
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
|
|
|
if (pchan->bone->flag & BONE_HINGE) {
|
|
|
|
/* pose_mat = par_pose-space_location * chan_mat */
|
|
|
|
float tmat[4][4];
|
|
|
|
|
|
|
|
/* the rotation of the parent restposition */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tmat, pchan->bone->parent->arm_mat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
|
|
|
/* the location of actual parent transform */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(tmat[3], offs_bone[3]);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
offs_bone[3][0]= offs_bone[3][1]= offs_bone[3][2]= 0.0f;
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_v3(pchan->parent->pose_mat, tmat[3]);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(diff_mat, offs_bone, tmat);
|
|
|
|
invert_m4_m4(imat, diff_mat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* pose_mat = par_pose_mat * bone_mat * chan_mat */
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(diff_mat, offs_bone, pchan->parent->pose_mat);
|
|
|
|
invert_m4_m4(imat, diff_mat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* pose_mat = chan_mat * arm_mat */
|
2009-11-10 20:43:45 +00:00
|
|
|
invert_m4_m4(imat, pchan->bone->arm_mat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(mat, tempmat, imat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
/* pose to local with parent */
|
|
|
|
else if (to == CONSTRAINT_SPACE_PARLOCAL) {
|
|
|
|
if (pchan->bone) {
|
2009-11-10 20:43:45 +00:00
|
|
|
invert_m4_m4(imat, pchan->bone->arm_mat);
|
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(mat, tempmat, imat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case CONSTRAINT_SPACE_LOCAL: /* ------------ FROM LOCALSPACE --------- */
|
|
|
|
{
|
2008-09-07 04:43:31 +00:00
|
|
|
/* local to pose - do inverse procedure that was done for pose to local */
|
|
|
|
if (pchan->bone) {
|
|
|
|
/* we need the posespace_matrix = local_matrix + (parent_posespace_matrix + restpos) */
|
|
|
|
if (pchan->parent) {
|
|
|
|
float offs_bone[4][4];
|
|
|
|
|
|
|
|
/* construct offs_bone the same way it is done in armature.c */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m3(offs_bone, pchan->bone->bone_mat);
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(offs_bone[3], pchan->bone->head);
|
2008-09-07 04:43:31 +00:00
|
|
|
offs_bone[3][1]+= pchan->bone->parent->length;
|
|
|
|
|
|
|
|
if (pchan->bone->flag & BONE_HINGE) {
|
|
|
|
/* pose_mat = par_pose-space_location * chan_mat */
|
|
|
|
float tmat[4][4];
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
2008-09-07 04:43:31 +00:00
|
|
|
/* the rotation of the parent restposition */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tmat, pchan->bone->parent->arm_mat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
2008-09-07 04:43:31 +00:00
|
|
|
/* the location of actual parent transform */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(tmat[3], offs_bone[3]);
|
|
|
|
zero_v3(offs_bone[3]);
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_v3(pchan->parent->pose_mat, tmat[3]);
|
2008-09-07 04:43:31 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(diff_mat, offs_bone, tmat);
|
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(mat, tempmat, diff_mat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
else {
|
2008-09-07 04:43:31 +00:00
|
|
|
/* pose_mat = par_pose_mat * bone_mat * chan_mat */
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(diff_mat, offs_bone, pchan->parent->pose_mat);
|
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(mat, tempmat, diff_mat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
}
|
2008-09-07 04:43:31 +00:00
|
|
|
else {
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(diff_mat, pchan->bone->arm_mat);
|
2008-09-07 04:43:31 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(mat, tempmat, diff_mat);
|
2008-09-07 04:43:31 +00:00
|
|
|
}
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
2008-09-07 04:43:31 +00:00
|
|
|
|
|
|
|
/* use pose-space as stepping stone for other spaces */
|
|
|
|
if (ELEM(to, CONSTRAINT_SPACE_WORLD, CONSTRAINT_SPACE_PARLOCAL)) {
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
/* call self with slightly different values */
|
|
|
|
constraint_mat_convertspace(ob, pchan, mat, CONSTRAINT_SPACE_POSE, to);
|
2008-09-07 04:43:31 +00:00
|
|
|
}
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case CONSTRAINT_SPACE_PARLOCAL: /* -------------- FROM LOCAL WITH PARENT ---------- */
|
|
|
|
{
|
2008-09-07 04:43:31 +00:00
|
|
|
/* local + parent to pose */
|
|
|
|
if (pchan->bone) {
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(diff_mat, pchan->bone->arm_mat);
|
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(mat, diff_mat, tempmat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
2008-09-07 04:43:31 +00:00
|
|
|
|
|
|
|
/* use pose-space as stepping stone for other spaces */
|
|
|
|
if (ELEM(to, CONSTRAINT_SPACE_WORLD, CONSTRAINT_SPACE_LOCAL)) {
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
/* call self with slightly different values */
|
|
|
|
constraint_mat_convertspace(ob, pchan, mat, CONSTRAINT_SPACE_POSE, to);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* objects */
|
|
|
|
if (from==CONSTRAINT_SPACE_WORLD && to==CONSTRAINT_SPACE_LOCAL) {
|
|
|
|
/* check if object has a parent - otherwise this won't work */
|
|
|
|
if (ob->parent) {
|
|
|
|
/* 'subtract' parent's effects from owner */
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(diff_mat, ob->parentinv, ob->parent->obmat);
|
|
|
|
invert_m4_m4(imat, diff_mat);
|
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(mat, tempmat, imat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
else if (from==CONSTRAINT_SPACE_LOCAL && to==CONSTRAINT_SPACE_WORLD) {
|
|
|
|
/* check that object has a parent - otherwise this won't work */
|
|
|
|
if (ob->parent) {
|
|
|
|
/* 'add' parent's effect back to owner */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tempmat, mat);
|
|
|
|
mul_m4_m4m4(diff_mat, ob->parentinv, ob->parent->obmat);
|
|
|
|
mul_m4_m4m4(mat, tempmat, diff_mat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* ------------ General Target Matrix Tools ---------- */
|
2002-10-12 11:37:38 +00:00
|
|
|
|
2007-08-17 11:23:48 +00:00
|
|
|
/* function that sets the given matrix based on given vertex group in mesh */
|
2009-04-20 15:06:46 +00:00
|
|
|
static void contarget_get_mesh_mat (Scene *scene, Object *ob, char *substring, float mat[][4])
|
2007-08-17 11:23:48 +00:00
|
|
|
{
|
2010-03-01 09:42:39 +00:00
|
|
|
DerivedMesh *dm = NULL;
|
2008-12-31 17:11:42 +00:00
|
|
|
Mesh *me= ob->data;
|
2009-04-11 08:26:51 +00:00
|
|
|
EditMesh *em = BKE_mesh_get_editmesh(me);
|
2010-08-15 15:14:08 +00:00
|
|
|
float vec[3] = {0.0f, 0.0f, 0.0f};
|
2007-08-18 04:03:03 +00:00
|
|
|
float normal[3] = {0.0f, 0.0f, 0.0f}, plane[3];
|
|
|
|
float imat[3][3], tmat[3][3];
|
2007-08-17 11:23:48 +00:00
|
|
|
int dgroup;
|
2009-04-20 15:06:46 +00:00
|
|
|
short freeDM = 0;
|
2007-08-17 11:23:48 +00:00
|
|
|
|
|
|
|
/* initialize target matrix using target matrix */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(mat, ob->obmat);
|
2007-08-17 11:23:48 +00:00
|
|
|
|
|
|
|
/* get index of vertex group */
|
2010-01-26 13:50:17 +00:00
|
|
|
dgroup = defgroup_name_index(ob, substring);
|
2007-08-17 11:23:48 +00:00
|
|
|
if (dgroup < 0) return;
|
|
|
|
|
2007-08-28 08:43:38 +00:00
|
|
|
/* get DerivedMesh */
|
2009-03-30 07:28:37 +00:00
|
|
|
if (em) {
|
2008-07-26 00:43:29 +00:00
|
|
|
/* target is in editmode, so get a special derived mesh */
|
2009-03-30 07:28:37 +00:00
|
|
|
dm = CDDM_from_editmesh(em, ob->data);
|
2009-04-20 15:06:46 +00:00
|
|
|
freeDM= 1;
|
2007-08-28 08:43:38 +00:00
|
|
|
}
|
|
|
|
else {
|
2009-04-20 15:06:46 +00:00
|
|
|
/* when not in EditMode, use the 'final' derived mesh
|
|
|
|
* - check if the custom data masks for derivedFinal mean that we can just use that
|
|
|
|
* (this is more effficient + sufficient for most cases)
|
|
|
|
*/
|
2010-04-05 11:11:15 +00:00
|
|
|
if (!(ob->lastDataMask & CD_MASK_MDEFORMVERT)) {
|
|
|
|
dm = mesh_get_derived_final(scene, ob, CD_MASK_MDEFORMVERT);
|
2009-04-20 15:06:46 +00:00
|
|
|
freeDM= 1;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
dm = (DerivedMesh *)ob->derivedFinal;
|
2007-08-28 08:43:38 +00:00
|
|
|
}
|
|
|
|
|
2007-08-17 11:23:48 +00:00
|
|
|
/* only continue if there's a valid DerivedMesh */
|
|
|
|
if (dm) {
|
|
|
|
MDeformVert *dvert = dm->getVertDataArray(dm, CD_MDEFORMVERT);
|
|
|
|
int numVerts = dm->getNumVerts(dm);
|
|
|
|
int i, j, count = 0;
|
2007-08-18 04:03:03 +00:00
|
|
|
float co[3], nor[3];
|
|
|
|
|
2009-11-04 20:23:48 +00:00
|
|
|
/* check that dvert is a valid pointers (just in case) */
|
|
|
|
if (dvert) {
|
2007-09-27 23:36:17 +00:00
|
|
|
/* get the average of all verts with that are in the vertex-group */
|
2009-11-04 20:23:48 +00:00
|
|
|
for (i = 0; i < numVerts; i++) {
|
2007-09-27 23:36:17 +00:00
|
|
|
for (j = 0; j < dvert[i].totweight; j++) {
|
|
|
|
/* does this vertex belong to nominated vertex group? */
|
|
|
|
if (dvert[i].dw[j].def_nr == dgroup) {
|
|
|
|
dm->getVertCo(dm, i, co);
|
|
|
|
dm->getVertNo(dm, i, nor);
|
2010-04-21 12:27:48 +00:00
|
|
|
add_v3_v3(vec, co);
|
|
|
|
add_v3_v3(normal, nor);
|
2007-09-27 23:36:17 +00:00
|
|
|
count++;
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
|
2007-08-17 11:23:48 +00:00
|
|
|
}
|
|
|
|
}
|
2007-09-27 23:36:17 +00:00
|
|
|
|
|
|
|
|
|
|
|
/* calculate averages of normal and coordinates */
|
|
|
|
if (count > 0) {
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(vec, 1.0f / count);
|
|
|
|
mul_v3_fl(normal, 1.0f / count);
|
2007-09-27 23:36:17 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
/* derive the rotation from the average normal:
|
|
|
|
* - code taken from transform_manipulator.c,
|
|
|
|
* calc_manipulator_stats, V3D_MANIP_NORMAL case
|
|
|
|
*/
|
|
|
|
/* we need the transpose of the inverse for a normal... */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m3_m4(imat, ob->obmat);
|
2007-09-27 23:36:17 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
invert_m3_m3(tmat, imat);
|
|
|
|
transpose_m3(tmat);
|
|
|
|
mul_m3_v3(tmat, normal);
|
2007-09-27 23:36:17 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
normalize_v3(normal);
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(plane, tmat[1]);
|
2007-09-27 23:36:17 +00:00
|
|
|
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(tmat[2], normal);
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(tmat[0], normal, plane);
|
|
|
|
cross_v3_v3v3(tmat[1], tmat[2], tmat[0]);
|
2007-09-27 23:36:17 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m3(mat, tmat);
|
|
|
|
normalize_m4(mat);
|
2007-09-27 23:36:17 +00:00
|
|
|
|
|
|
|
|
|
|
|
/* apply the average coordinate as the new location */
|
2010-08-15 15:14:08 +00:00
|
|
|
mul_v3_m4v3(mat[3], ob->obmat, vec);
|
2007-08-17 11:23:48 +00:00
|
|
|
}
|
|
|
|
}
|
2007-08-28 08:43:38 +00:00
|
|
|
|
|
|
|
/* free temporary DerivedMesh created (in EditMode case) */
|
2009-04-20 15:06:46 +00:00
|
|
|
if (dm && freeDM)
|
|
|
|
dm->release(dm);
|
|
|
|
if (em)
|
2009-04-11 08:26:51 +00:00
|
|
|
BKE_mesh_end_editmesh(me, em);
|
2007-08-17 11:23:48 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* function that sets the given matrix based on given vertex group in lattice */
|
|
|
|
static void contarget_get_lattice_mat (Object *ob, char *substring, float mat[][4])
|
|
|
|
{
|
|
|
|
Lattice *lt= (Lattice *)ob->data;
|
|
|
|
|
|
|
|
DispList *dl = find_displist(&ob->disp, DL_VERTS);
|
|
|
|
float *co = dl?dl->verts:NULL;
|
|
|
|
BPoint *bp = lt->def;
|
|
|
|
|
|
|
|
MDeformVert *dvert = lt->dvert;
|
|
|
|
int tot_verts= lt->pntsu*lt->pntsv*lt->pntsw;
|
|
|
|
float vec[3]= {0.0f, 0.0f, 0.0f}, tvec[3];
|
|
|
|
int dgroup=0, grouped=0;
|
|
|
|
int i, n;
|
|
|
|
|
|
|
|
/* initialize target matrix using target matrix */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(mat, ob->obmat);
|
2007-08-17 11:23:48 +00:00
|
|
|
|
|
|
|
/* get index of vertex group */
|
2010-01-26 13:50:17 +00:00
|
|
|
dgroup = defgroup_name_index(ob, substring);
|
2007-08-17 11:23:48 +00:00
|
|
|
if (dgroup < 0) return;
|
2007-09-27 23:36:17 +00:00
|
|
|
if (dvert == NULL) return;
|
2007-08-17 11:23:48 +00:00
|
|
|
|
|
|
|
/* 1. Loop through control-points checking if in nominated vertex-group.
|
|
|
|
* 2. If it is, add it to vec to find the average point.
|
|
|
|
*/
|
|
|
|
for (i=0; i < tot_verts; i++, dvert++) {
|
|
|
|
for (n= 0; n < dvert->totweight; n++) {
|
|
|
|
/* found match - vert is in vgroup */
|
|
|
|
if (dvert->dw[n].def_nr == dgroup) {
|
|
|
|
/* copy coordinates of point to temporary vector, then add to find average */
|
|
|
|
if (co)
|
|
|
|
memcpy(tvec, co, 3*sizeof(float));
|
|
|
|
else
|
|
|
|
memcpy(tvec, bp->vec, 3*sizeof(float));
|
|
|
|
|
2010-04-21 12:27:48 +00:00
|
|
|
add_v3_v3(vec, tvec);
|
2007-08-17 11:23:48 +00:00
|
|
|
grouped++;
|
|
|
|
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* advance pointer to coordinate data */
|
|
|
|
if (co) co+= 3;
|
|
|
|
else bp++;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* find average location, then multiply by ob->obmat to find world-space location */
|
|
|
|
if (grouped)
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(vec, 1.0f / grouped);
|
|
|
|
mul_v3_m4v3(tvec, ob->obmat, vec);
|
2007-08-17 11:23:48 +00:00
|
|
|
|
|
|
|
/* copy new location to matrix */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(mat[3], tvec);
|
2007-08-17 11:23:48 +00:00
|
|
|
}
|
|
|
|
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
/* generic function to get the appropriate matrix for most target cases */
|
|
|
|
/* The cases where the target can be object data have not been implemented */
|
2009-04-20 15:06:46 +00:00
|
|
|
static void constraint_target_to_mat4 (Scene *scene, Object *ob, char *substring, float mat[][4], short from, short to, float headtail)
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
{
|
2002-10-12 11:37:38 +00:00
|
|
|
/* Case OBJECT */
|
Result of 2 weeks of quiet coding work in Greece :)
Aim was to get a total refresh of the animation system. This
is needed because;
- we need to upgrade it with 21st century features
- current code is spaghetti/hack combo, and hides good design
- it should become lag-free with using dependency graphs
A full log, with complete code API/structure/design explanation
will follow, that's a load of work... so here below the list with
hot changes;
- The entire object update system (matrices, geometry) is now
centralized. Calls to where_is_object and makeDispList are
forbidden, instead we tag objects 'changed' and let the
depgraph code sort it out
- Removed all old "Ika" code
- Depgraph is aware of all relationships, including meta balls,
constraints, bevelcurve, and so on.
- Made depgraph aware of relation types and layers, to do smart
flushing of 'changed' events. Nothing gets calculated too often!
- Transform uses depgraph to detect changes
- On frame-advance, depgraph flushes animated changes
Armatures;
Almost all armature related code has been fully built from scratch.
It now reveils the original design much better, with a very clean
implementation, lag free without even calculating each Bone more than
once. Result is quite a speedup yes!
Important to note is;
1) Armature is data containing the 'rest position'
2) Pose is the changes of rest position, and always on object level.
That way more Objects can use same Pose. Also constraints are in Pose
3) Actions only contain the Ipos to change values in Poses.
- Bones draw unrotated now
- Drawing bones speedup enormously (10-20 times)
- Bone selecting in EditMode, selection state is saved for PoseMode,
and vice-versa
- Undo in editmode
- Bone renaming does vertexgroups, constraints, posechannels, actions,
for all users of Armature in entire file
- Added Bone renaming in NKey panel
- Nkey PoseMode shows eulers now
- EditMode and PoseMode now have 'active' bone too (last clicked)
- Parenting in EditMode' CTRL+P, ALT+P, with nice options!
- Pose is added in Outliner now, with showing that constraints are in
the Pose, not Armature
- Disconnected IK solving from constraints. It's a separate phase now,
on top of the full Pose calculations
- Pose itself has a dependency graph too, so evaluation order is lag free.
TODO NOW;
- Rotating in Posemode has incorrect inverse transform (Martin will fix)
- Python Bone/Armature/Pose API disabled... needs full recode too
(wait for my doc!)
- Game engine will need upgrade too
- Depgraph code needs revision, cleanup, can be much faster!
(But, compliments for Jean-Luc, it works like a charm!)
- IK changed, it now doesnt use previous position to advance to next
position anymore. That system looks nice (no flips) but is not well
suited for NLA and background render.
TODO LATER;
We now can do loadsa new nifty features as well; like:
- Kill PoseMode (can be option for armatures itself)
- Make B-Bones (Bezier, Bspline, like for spines)
- Move all silly button level edit to 3d window (like CTRL+I = add
IK)
- Much better & informative drawing
- Fix action/nla editors
- Put all ipos in Actions (object, mesh key, lamp color)
- Add hooks
- Null bones
- Much more advanced constraints...
Bugfixes;
- OGL render (view3d header) had wrong first frame on anim render
- Ipo 'recording' mode had wrong playback speed
- Vertex-key mode now sticks to show 'active key', until frame change
-Ton-
2005-07-03 17:35:38 +00:00
|
|
|
if (!strlen(substring)) {
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(mat, ob->obmat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
constraint_mat_convertspace(ob, NULL, mat, from, to);
|
|
|
|
}
|
|
|
|
/* Case VERTEXGROUP */
|
2007-08-17 11:23:48 +00:00
|
|
|
/* Current method just takes the average location of all the points in the
|
2007-08-18 04:03:03 +00:00
|
|
|
* VertexGroup, and uses that as the location value of the targets. Where
|
|
|
|
* possible, the orientation will also be calculated, by calculating an
|
|
|
|
* 'average' vertex normal, and deriving the rotaation from that.
|
2007-08-17 11:23:48 +00:00
|
|
|
*
|
2007-08-18 04:03:03 +00:00
|
|
|
* NOTE: EditMode is not currently supported, and will most likely remain that
|
|
|
|
* way as constraints can only really affect things on object/bone level.
|
2007-08-17 11:23:48 +00:00
|
|
|
*/
|
|
|
|
else if (ob->type == OB_MESH) {
|
2009-04-20 15:06:46 +00:00
|
|
|
contarget_get_mesh_mat(scene, ob, substring, mat);
|
2007-08-17 11:23:48 +00:00
|
|
|
constraint_mat_convertspace(ob, NULL, mat, from, to);
|
|
|
|
}
|
|
|
|
else if (ob->type == OB_LATTICE) {
|
|
|
|
contarget_get_lattice_mat(ob, substring, mat);
|
|
|
|
constraint_mat_convertspace(ob, NULL, mat, from, to);
|
2002-10-12 11:37:38 +00:00
|
|
|
}
|
|
|
|
/* Case BONE */
|
|
|
|
else {
|
Result of 2 weeks of quiet coding work in Greece :)
Aim was to get a total refresh of the animation system. This
is needed because;
- we need to upgrade it with 21st century features
- current code is spaghetti/hack combo, and hides good design
- it should become lag-free with using dependency graphs
A full log, with complete code API/structure/design explanation
will follow, that's a load of work... so here below the list with
hot changes;
- The entire object update system (matrices, geometry) is now
centralized. Calls to where_is_object and makeDispList are
forbidden, instead we tag objects 'changed' and let the
depgraph code sort it out
- Removed all old "Ika" code
- Depgraph is aware of all relationships, including meta balls,
constraints, bevelcurve, and so on.
- Made depgraph aware of relation types and layers, to do smart
flushing of 'changed' events. Nothing gets calculated too often!
- Transform uses depgraph to detect changes
- On frame-advance, depgraph flushes animated changes
Armatures;
Almost all armature related code has been fully built from scratch.
It now reveils the original design much better, with a very clean
implementation, lag free without even calculating each Bone more than
once. Result is quite a speedup yes!
Important to note is;
1) Armature is data containing the 'rest position'
2) Pose is the changes of rest position, and always on object level.
That way more Objects can use same Pose. Also constraints are in Pose
3) Actions only contain the Ipos to change values in Poses.
- Bones draw unrotated now
- Drawing bones speedup enormously (10-20 times)
- Bone selecting in EditMode, selection state is saved for PoseMode,
and vice-versa
- Undo in editmode
- Bone renaming does vertexgroups, constraints, posechannels, actions,
for all users of Armature in entire file
- Added Bone renaming in NKey panel
- Nkey PoseMode shows eulers now
- EditMode and PoseMode now have 'active' bone too (last clicked)
- Parenting in EditMode' CTRL+P, ALT+P, with nice options!
- Pose is added in Outliner now, with showing that constraints are in
the Pose, not Armature
- Disconnected IK solving from constraints. It's a separate phase now,
on top of the full Pose calculations
- Pose itself has a dependency graph too, so evaluation order is lag free.
TODO NOW;
- Rotating in Posemode has incorrect inverse transform (Martin will fix)
- Python Bone/Armature/Pose API disabled... needs full recode too
(wait for my doc!)
- Game engine will need upgrade too
- Depgraph code needs revision, cleanup, can be much faster!
(But, compliments for Jean-Luc, it works like a charm!)
- IK changed, it now doesnt use previous position to advance to next
position anymore. That system looks nice (no flips) but is not well
suited for NLA and background render.
TODO LATER;
We now can do loadsa new nifty features as well; like:
- Kill PoseMode (can be option for armatures itself)
- Make B-Bones (Bezier, Bspline, like for spines)
- Move all silly button level edit to 3d window (like CTRL+I = add
IK)
- Much better & informative drawing
- Fix action/nla editors
- Put all ipos in Actions (object, mesh key, lamp color)
- Add hooks
- Null bones
- Much more advanced constraints...
Bugfixes;
- OGL render (view3d header) had wrong first frame on anim render
- Ipo 'recording' mode had wrong playback speed
- Vertex-key mode now sticks to show 'active key', until frame change
-Ton-
2005-07-03 17:35:38 +00:00
|
|
|
bPoseChannel *pchan;
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
Result of 2 weeks of quiet coding work in Greece :)
Aim was to get a total refresh of the animation system. This
is needed because;
- we need to upgrade it with 21st century features
- current code is spaghetti/hack combo, and hides good design
- it should become lag-free with using dependency graphs
A full log, with complete code API/structure/design explanation
will follow, that's a load of work... so here below the list with
hot changes;
- The entire object update system (matrices, geometry) is now
centralized. Calls to where_is_object and makeDispList are
forbidden, instead we tag objects 'changed' and let the
depgraph code sort it out
- Removed all old "Ika" code
- Depgraph is aware of all relationships, including meta balls,
constraints, bevelcurve, and so on.
- Made depgraph aware of relation types and layers, to do smart
flushing of 'changed' events. Nothing gets calculated too often!
- Transform uses depgraph to detect changes
- On frame-advance, depgraph flushes animated changes
Armatures;
Almost all armature related code has been fully built from scratch.
It now reveils the original design much better, with a very clean
implementation, lag free without even calculating each Bone more than
once. Result is quite a speedup yes!
Important to note is;
1) Armature is data containing the 'rest position'
2) Pose is the changes of rest position, and always on object level.
That way more Objects can use same Pose. Also constraints are in Pose
3) Actions only contain the Ipos to change values in Poses.
- Bones draw unrotated now
- Drawing bones speedup enormously (10-20 times)
- Bone selecting in EditMode, selection state is saved for PoseMode,
and vice-versa
- Undo in editmode
- Bone renaming does vertexgroups, constraints, posechannels, actions,
for all users of Armature in entire file
- Added Bone renaming in NKey panel
- Nkey PoseMode shows eulers now
- EditMode and PoseMode now have 'active' bone too (last clicked)
- Parenting in EditMode' CTRL+P, ALT+P, with nice options!
- Pose is added in Outliner now, with showing that constraints are in
the Pose, not Armature
- Disconnected IK solving from constraints. It's a separate phase now,
on top of the full Pose calculations
- Pose itself has a dependency graph too, so evaluation order is lag free.
TODO NOW;
- Rotating in Posemode has incorrect inverse transform (Martin will fix)
- Python Bone/Armature/Pose API disabled... needs full recode too
(wait for my doc!)
- Game engine will need upgrade too
- Depgraph code needs revision, cleanup, can be much faster!
(But, compliments for Jean-Luc, it works like a charm!)
- IK changed, it now doesnt use previous position to advance to next
position anymore. That system looks nice (no flips) but is not well
suited for NLA and background render.
TODO LATER;
We now can do loadsa new nifty features as well; like:
- Kill PoseMode (can be option for armatures itself)
- Make B-Bones (Bezier, Bspline, like for spines)
- Move all silly button level edit to 3d window (like CTRL+I = add
IK)
- Much better & informative drawing
- Fix action/nla editors
- Put all ipos in Actions (object, mesh key, lamp color)
- Add hooks
- Null bones
- Much more advanced constraints...
Bugfixes;
- OGL render (view3d header) had wrong first frame on anim render
- Ipo 'recording' mode had wrong playback speed
- Vertex-key mode now sticks to show 'active key', until frame change
-Ton-
2005-07-03 17:35:38 +00:00
|
|
|
pchan = get_pose_channel(ob->pose, substring);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
if (pchan) {
|
|
|
|
/* Multiply the PoseSpace accumulation/final matrix for this
|
|
|
|
* PoseChannel by the Armature Object's Matrix to get a worldspace
|
|
|
|
* matrix.
|
Result of 2 weeks of quiet coding work in Greece :)
Aim was to get a total refresh of the animation system. This
is needed because;
- we need to upgrade it with 21st century features
- current code is spaghetti/hack combo, and hides good design
- it should become lag-free with using dependency graphs
A full log, with complete code API/structure/design explanation
will follow, that's a load of work... so here below the list with
hot changes;
- The entire object update system (matrices, geometry) is now
centralized. Calls to where_is_object and makeDispList are
forbidden, instead we tag objects 'changed' and let the
depgraph code sort it out
- Removed all old "Ika" code
- Depgraph is aware of all relationships, including meta balls,
constraints, bevelcurve, and so on.
- Made depgraph aware of relation types and layers, to do smart
flushing of 'changed' events. Nothing gets calculated too often!
- Transform uses depgraph to detect changes
- On frame-advance, depgraph flushes animated changes
Armatures;
Almost all armature related code has been fully built from scratch.
It now reveils the original design much better, with a very clean
implementation, lag free without even calculating each Bone more than
once. Result is quite a speedup yes!
Important to note is;
1) Armature is data containing the 'rest position'
2) Pose is the changes of rest position, and always on object level.
That way more Objects can use same Pose. Also constraints are in Pose
3) Actions only contain the Ipos to change values in Poses.
- Bones draw unrotated now
- Drawing bones speedup enormously (10-20 times)
- Bone selecting in EditMode, selection state is saved for PoseMode,
and vice-versa
- Undo in editmode
- Bone renaming does vertexgroups, constraints, posechannels, actions,
for all users of Armature in entire file
- Added Bone renaming in NKey panel
- Nkey PoseMode shows eulers now
- EditMode and PoseMode now have 'active' bone too (last clicked)
- Parenting in EditMode' CTRL+P, ALT+P, with nice options!
- Pose is added in Outliner now, with showing that constraints are in
the Pose, not Armature
- Disconnected IK solving from constraints. It's a separate phase now,
on top of the full Pose calculations
- Pose itself has a dependency graph too, so evaluation order is lag free.
TODO NOW;
- Rotating in Posemode has incorrect inverse transform (Martin will fix)
- Python Bone/Armature/Pose API disabled... needs full recode too
(wait for my doc!)
- Game engine will need upgrade too
- Depgraph code needs revision, cleanup, can be much faster!
(But, compliments for Jean-Luc, it works like a charm!)
- IK changed, it now doesnt use previous position to advance to next
position anymore. That system looks nice (no flips) but is not well
suited for NLA and background render.
TODO LATER;
We now can do loadsa new nifty features as well; like:
- Kill PoseMode (can be option for armatures itself)
- Make B-Bones (Bezier, Bspline, like for spines)
- Move all silly button level edit to 3d window (like CTRL+I = add
IK)
- Much better & informative drawing
- Fix action/nla editors
- Put all ipos in Actions (object, mesh key, lamp color)
- Add hooks
- Null bones
- Much more advanced constraints...
Bugfixes;
- OGL render (view3d header) had wrong first frame on anim render
- Ipo 'recording' mode had wrong playback speed
- Vertex-key mode now sticks to show 'active key', until frame change
-Ton-
2005-07-03 17:35:38 +00:00
|
|
|
*/
|
Patch #7767: Constraint Subtargets can now target anywhere on a bone, not just the head or tail
Patch by: Roland Hess (harkyman)
For example, a constraint can be sub-targeted at the 50% (or 31.2% or 85% etc.) point of its target bone, giving you enormous rigging flexibility and removing the need for complex contraptions to do such things as:
- A bone whose base slides only between to points on a rig (CopyLoc with a variable, animated subtarget point)
- Bones that attach to multiple points along another bone (CopyLocs, each with a different head/tail percentage)
- Bones that need to stretch to a point midway between specific spots on two other bones (old way: too crazy to mention; new way: stretch bone between points on end bones, then another stretch to the midpoint of the first stretch)
It is only used for the constraint types for which it is relevant: CopyLoc, TrackTo, StretchTo and MinMax, TrackTo, and Floor.
Notes:
- This is accessed by the Head/Tail number-slider.
- This value can be animated per constraint
- The old "Copy Bone Tail" option for the CopyLoc constraint has been automatically converted to 1.0 Head/Bone values for the affected constraints
- In the code, this value is in the bConstraint struct, so it is available for all constraints, even though only a few implement it.
2007-11-12 04:17:03 +00:00
|
|
|
if (headtail < 0.000001) {
|
|
|
|
/* skip length interpolation if set to head */
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(mat, pchan->pose_mat, ob->obmat);
|
Patch #7767: Constraint Subtargets can now target anywhere on a bone, not just the head or tail
Patch by: Roland Hess (harkyman)
For example, a constraint can be sub-targeted at the 50% (or 31.2% or 85% etc.) point of its target bone, giving you enormous rigging flexibility and removing the need for complex contraptions to do such things as:
- A bone whose base slides only between to points on a rig (CopyLoc with a variable, animated subtarget point)
- Bones that attach to multiple points along another bone (CopyLocs, each with a different head/tail percentage)
- Bones that need to stretch to a point midway between specific spots on two other bones (old way: too crazy to mention; new way: stretch bone between points on end bones, then another stretch to the midpoint of the first stretch)
It is only used for the constraint types for which it is relevant: CopyLoc, TrackTo, StretchTo and MinMax, TrackTo, and Floor.
Notes:
- This is accessed by the Head/Tail number-slider.
- This value can be animated per constraint
- The old "Copy Bone Tail" option for the CopyLoc constraint has been automatically converted to 1.0 Head/Bone values for the affected constraints
- In the code, this value is in the bConstraint struct, so it is available for all constraints, even though only a few implement it.
2007-11-12 04:17:03 +00:00
|
|
|
}
|
|
|
|
else {
|
|
|
|
float tempmat[4][4], loc[3];
|
|
|
|
|
|
|
|
/* interpolate along length of bone */
|
2009-11-10 20:43:45 +00:00
|
|
|
interp_v3_v3v3(loc, pchan->pose_head, pchan->pose_tail, headtail);
|
Patch #7767: Constraint Subtargets can now target anywhere on a bone, not just the head or tail
Patch by: Roland Hess (harkyman)
For example, a constraint can be sub-targeted at the 50% (or 31.2% or 85% etc.) point of its target bone, giving you enormous rigging flexibility and removing the need for complex contraptions to do such things as:
- A bone whose base slides only between to points on a rig (CopyLoc with a variable, animated subtarget point)
- Bones that attach to multiple points along another bone (CopyLocs, each with a different head/tail percentage)
- Bones that need to stretch to a point midway between specific spots on two other bones (old way: too crazy to mention; new way: stretch bone between points on end bones, then another stretch to the midpoint of the first stretch)
It is only used for the constraint types for which it is relevant: CopyLoc, TrackTo, StretchTo and MinMax, TrackTo, and Floor.
Notes:
- This is accessed by the Head/Tail number-slider.
- This value can be animated per constraint
- The old "Copy Bone Tail" option for the CopyLoc constraint has been automatically converted to 1.0 Head/Bone values for the affected constraints
- In the code, this value is in the bConstraint struct, so it is available for all constraints, even though only a few implement it.
2007-11-12 04:17:03 +00:00
|
|
|
|
|
|
|
/* use interpolated distance for subtarget */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tempmat, pchan->pose_mat);
|
|
|
|
copy_v3_v3(tempmat[3], loc);
|
Patch #7767: Constraint Subtargets can now target anywhere on a bone, not just the head or tail
Patch by: Roland Hess (harkyman)
For example, a constraint can be sub-targeted at the 50% (or 31.2% or 85% etc.) point of its target bone, giving you enormous rigging flexibility and removing the need for complex contraptions to do such things as:
- A bone whose base slides only between to points on a rig (CopyLoc with a variable, animated subtarget point)
- Bones that attach to multiple points along another bone (CopyLocs, each with a different head/tail percentage)
- Bones that need to stretch to a point midway between specific spots on two other bones (old way: too crazy to mention; new way: stretch bone between points on end bones, then another stretch to the midpoint of the first stretch)
It is only used for the constraint types for which it is relevant: CopyLoc, TrackTo, StretchTo and MinMax, TrackTo, and Floor.
Notes:
- This is accessed by the Head/Tail number-slider.
- This value can be animated per constraint
- The old "Copy Bone Tail" option for the CopyLoc constraint has been automatically converted to 1.0 Head/Bone values for the affected constraints
- In the code, this value is in the bConstraint struct, so it is available for all constraints, even though only a few implement it.
2007-11-12 04:17:03 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(mat, tempmat, ob->obmat);
|
Patch #7767: Constraint Subtargets can now target anywhere on a bone, not just the head or tail
Patch by: Roland Hess (harkyman)
For example, a constraint can be sub-targeted at the 50% (or 31.2% or 85% etc.) point of its target bone, giving you enormous rigging flexibility and removing the need for complex contraptions to do such things as:
- A bone whose base slides only between to points on a rig (CopyLoc with a variable, animated subtarget point)
- Bones that attach to multiple points along another bone (CopyLocs, each with a different head/tail percentage)
- Bones that need to stretch to a point midway between specific spots on two other bones (old way: too crazy to mention; new way: stretch bone between points on end bones, then another stretch to the midpoint of the first stretch)
It is only used for the constraint types for which it is relevant: CopyLoc, TrackTo, StretchTo and MinMax, TrackTo, and Floor.
Notes:
- This is accessed by the Head/Tail number-slider.
- This value can be animated per constraint
- The old "Copy Bone Tail" option for the CopyLoc constraint has been automatically converted to 1.0 Head/Bone values for the affected constraints
- In the code, this value is in the bConstraint struct, so it is available for all constraints, even though only a few implement it.
2007-11-12 04:17:03 +00:00
|
|
|
}
|
2002-10-12 11:37:38 +00:00
|
|
|
}
|
|
|
|
else
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(mat, ob->obmat);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
|
|
|
/* convert matrix space as required */
|
|
|
|
constraint_mat_convertspace(ob, pchan, mat, from, to);
|
2004-01-27 06:08:37 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* ************************* Specific Constraints ***************************** */
|
|
|
|
/* Each constraint defines a set of functions, which will be called at the appropriate
|
|
|
|
* times. In addition to this, each constraint should have a type-info struct, where
|
|
|
|
* its functions are attached for use.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* Template for type-info data:
|
|
|
|
* - make a copy of this when creating new constraints, and just change the functions
|
|
|
|
* pointed to as necessary
|
|
|
|
* - although the naming of functions doesn't matter, it would help for code
|
|
|
|
* readability, to follow the same naming convention as is presented here
|
|
|
|
* - any functions that a constraint doesn't need to define, don't define
|
|
|
|
* for such cases, just use NULL
|
|
|
|
* - these should be defined after all the functions have been defined, so that
|
|
|
|
* forward-definitions/prototypes don't need to be used!
|
|
|
|
* - keep this copy #if-def'd so that future constraints can get based off this
|
|
|
|
*/
|
|
|
|
#if 0
|
|
|
|
static bConstraintTypeInfo CTI_CONSTRNAME = {
|
|
|
|
CONSTRAINT_TYPE_CONSTRNAME, /* type */
|
|
|
|
sizeof(bConstrNameConstraint), /* size */
|
|
|
|
"ConstrName", /* name */
|
|
|
|
"bConstrNameConstraint", /* struct name */
|
|
|
|
constrname_free, /* free data */
|
|
|
|
constrname_relink, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
constrname_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
constrname_copy, /* copy data */
|
|
|
|
constrname_new_data, /* new data */
|
|
|
|
constrname_get_tars, /* get constraint targets */
|
|
|
|
constrname_flush_tars, /* flush constraint targets */
|
|
|
|
constrname_get_tarmat, /* get target matrix */
|
|
|
|
constrname_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/* This function should be used for the get_target_matrix member of all
|
|
|
|
* constraints that are not picky about what happens to their target matrix.
|
|
|
|
*/
|
2010-10-16 14:32:17 +00:00
|
|
|
static void default_get_tarmat (bConstraint *con, bConstraintOb *cob, bConstraintTarget *ct, float UNUSED(ctime))
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (VALID_CONS_TARGET(ct))
|
2009-04-20 15:06:46 +00:00
|
|
|
constraint_target_to_mat4(cob->scene, ct->tar, ct->subtarget, ct->matrix, CONSTRAINT_SPACE_WORLD, ct->space, con->headtail);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
else if (ct)
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* This following macro should be used for all standard single-target *_get_tars functions
|
|
|
|
* to save typing and reduce maintainance woes.
|
|
|
|
* (Hopefully all compilers will be happy with the lines with just a space on them. Those are
|
|
|
|
* really just to help this code easier to read)
|
|
|
|
*/
|
2009-09-02 10:45:11 +00:00
|
|
|
// TODO: cope with getting rotation order...
|
2007-10-24 14:58:31 +00:00
|
|
|
#define SINGLETARGET_GET_TARS(con, datatar, datasubtarget, ct, list) \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{ \
|
|
|
|
ct= MEM_callocN(sizeof(bConstraintTarget), "tempConstraintTarget"); \
|
|
|
|
\
|
2007-10-24 14:58:31 +00:00
|
|
|
ct->tar= datatar; \
|
|
|
|
strcpy(ct->subtarget, datasubtarget); \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
ct->space= con->tarspace; \
|
|
|
|
ct->flag= CONSTRAINT_TAR_TEMP; \
|
|
|
|
\
|
|
|
|
if (ct->tar) { \
|
2009-09-20 11:21:44 +00:00
|
|
|
if ((ct->tar->type==OB_ARMATURE) && (ct->subtarget[0])) { \
|
|
|
|
bPoseChannel *pchan= get_pose_channel(ct->tar->pose, ct->subtarget); \
|
|
|
|
ct->type = CONSTRAINT_OBTYPE_BONE; \
|
2009-09-25 04:51:04 +00:00
|
|
|
ct->rotOrder= (pchan) ? (pchan->rotmode) : EULER_ORDER_DEFAULT; \
|
2009-09-20 11:21:44 +00:00
|
|
|
}\
|
|
|
|
else if (ELEM(ct->tar->type, OB_MESH, OB_LATTICE) && (ct->subtarget[0])) { \
|
|
|
|
ct->type = CONSTRAINT_OBTYPE_VERT; \
|
2009-09-28 10:19:20 +00:00
|
|
|
ct->rotOrder = EULER_ORDER_DEFAULT; \
|
2009-09-20 11:21:44 +00:00
|
|
|
} \
|
|
|
|
else {\
|
|
|
|
ct->type = CONSTRAINT_OBTYPE_OBJECT; \
|
2009-09-28 10:19:20 +00:00
|
|
|
ct->rotOrder= ct->tar->rotmode; \
|
2009-09-20 11:21:44 +00:00
|
|
|
} \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
} \
|
|
|
|
\
|
|
|
|
BLI_addtail(list, ct); \
|
|
|
|
}
|
|
|
|
|
|
|
|
/* This following macro should be used for all standard single-target *_get_tars functions
|
|
|
|
* to save typing and reduce maintainance woes. It does not do the subtarget related operations
|
|
|
|
* (Hopefully all compilers will be happy with the lines with just a space on them. Those are
|
|
|
|
* really just to help this code easier to read)
|
|
|
|
*/
|
2009-09-02 10:45:11 +00:00
|
|
|
// TODO: cope with getting rotation order...
|
2007-10-24 14:58:31 +00:00
|
|
|
#define SINGLETARGETNS_GET_TARS(con, datatar, ct, list) \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{ \
|
|
|
|
ct= MEM_callocN(sizeof(bConstraintTarget), "tempConstraintTarget"); \
|
|
|
|
\
|
2007-10-24 14:58:31 +00:00
|
|
|
ct->tar= datatar; \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
ct->space= con->tarspace; \
|
|
|
|
ct->flag= CONSTRAINT_TAR_TEMP; \
|
|
|
|
\
|
|
|
|
if (ct->tar) ct->type = CONSTRAINT_OBTYPE_OBJECT; \
|
|
|
|
\
|
|
|
|
BLI_addtail(list, ct); \
|
|
|
|
}
|
|
|
|
|
|
|
|
/* This following macro should be used for all standard single-target *_flush_tars functions
|
|
|
|
* to save typing and reduce maintainance woes.
|
2007-12-06 10:20:03 +00:00
|
|
|
* Note: the pointer to ct will be changed to point to the next in the list (as it gets removed)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
* (Hopefully all compilers will be happy with the lines with just a space on them. Those are
|
|
|
|
* really just to help this code easier to read)
|
|
|
|
*/
|
2007-10-24 14:58:31 +00:00
|
|
|
#define SINGLETARGET_FLUSH_TARS(con, datatar, datasubtarget, ct, list, nocopy) \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{ \
|
|
|
|
if (ct) { \
|
2007-12-06 10:20:03 +00:00
|
|
|
bConstraintTarget *ctn = ct->next; \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (nocopy == 0) { \
|
2007-10-24 14:58:31 +00:00
|
|
|
datatar= ct->tar; \
|
|
|
|
strcpy(datasubtarget, ct->subtarget); \
|
2008-10-20 06:39:08 +00:00
|
|
|
con->tarspace= (char)ct->space; \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
} \
|
|
|
|
\
|
2007-10-24 14:58:31 +00:00
|
|
|
BLI_freelinkN(list, ct); \
|
2007-12-06 10:20:03 +00:00
|
|
|
ct= ctn; \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
} \
|
|
|
|
}
|
|
|
|
|
|
|
|
/* This following macro should be used for all standard single-target *_flush_tars functions
|
2007-12-06 10:20:03 +00:00
|
|
|
* to save typing and reduce maintainance woes. It does not do the subtarget related operations.
|
|
|
|
* Note: the pointer to ct will be changed to point to the next in the list (as it gets removed)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
* (Hopefully all compilers will be happy with the lines with just a space on them. Those are
|
|
|
|
* really just to help this code easier to read)
|
|
|
|
*/
|
2007-10-24 14:58:31 +00:00
|
|
|
#define SINGLETARGETNS_FLUSH_TARS(con, datatar, ct, list, nocopy) \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{ \
|
|
|
|
if (ct) { \
|
2007-12-06 10:20:03 +00:00
|
|
|
bConstraintTarget *ctn = ct->next; \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (nocopy == 0) { \
|
2007-10-24 14:58:31 +00:00
|
|
|
datatar= ct->tar; \
|
2008-10-20 06:39:08 +00:00
|
|
|
con->tarspace= (char)ct->space; \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
} \
|
|
|
|
\
|
2007-10-24 14:58:31 +00:00
|
|
|
BLI_freelinkN(list, ct); \
|
2007-12-06 10:20:03 +00:00
|
|
|
ct= ctn; \
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
} \
|
|
|
|
}
|
|
|
|
|
|
|
|
/* --------- ChildOf Constraint ------------ */
|
|
|
|
|
|
|
|
static void childof_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bChildOfConstraint *data= (bChildOfConstraint *)cdata;
|
|
|
|
|
|
|
|
data->flag = (CHILDOF_LOCX | CHILDOF_LOCY | CHILDOF_LOCZ |
|
|
|
|
CHILDOF_ROTX |CHILDOF_ROTY | CHILDOF_ROTZ |
|
|
|
|
CHILDOF_SIZEX | CHILDOF_SIZEY | CHILDOF_SIZEZ);
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(data->invmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void childof_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bChildOfConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int childof_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bChildOfConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void childof_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bChildOfConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void childof_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bChildOfConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
/* only evaluate if there is a target */
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float parmat[4][4], invmat[4][4], tempmat[4][4];
|
|
|
|
float loc[3], eul[3], size[3];
|
|
|
|
float loco[3], eulo[3], sizo[3];
|
|
|
|
|
|
|
|
/* get offset (parent-inverse) matrix */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(invmat, data->invmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* extract components of both matrices */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(loc, ct->matrix[3]);
|
2010-05-27 11:56:31 +00:00
|
|
|
mat4_to_eulO(eul, ct->rotOrder, ct->matrix);
|
|
|
|
mat4_to_size(size, ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(loco, invmat[3]);
|
2010-05-27 11:56:31 +00:00
|
|
|
mat4_to_eulO(eulo, cob->rotOrder, invmat);
|
|
|
|
mat4_to_size(sizo, invmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* disable channels not enabled */
|
|
|
|
if (!(data->flag & CHILDOF_LOCX)) loc[0]= loco[0]= 0.0f;
|
|
|
|
if (!(data->flag & CHILDOF_LOCY)) loc[1]= loco[1]= 0.0f;
|
|
|
|
if (!(data->flag & CHILDOF_LOCZ)) loc[2]= loco[2]= 0.0f;
|
|
|
|
if (!(data->flag & CHILDOF_ROTX)) eul[0]= eulo[0]= 0.0f;
|
|
|
|
if (!(data->flag & CHILDOF_ROTY)) eul[1]= eulo[1]= 0.0f;
|
|
|
|
if (!(data->flag & CHILDOF_ROTZ)) eul[2]= eulo[2]= 0.0f;
|
|
|
|
if (!(data->flag & CHILDOF_SIZEX)) size[0]= sizo[0]= 1.0f;
|
|
|
|
if (!(data->flag & CHILDOF_SIZEY)) size[1]= sizo[1]= 1.0f;
|
|
|
|
if (!(data->flag & CHILDOF_SIZEZ)) size[2]= sizo[2]= 1.0f;
|
|
|
|
|
|
|
|
/* make new target mat and offset mat */
|
2009-11-10 20:43:45 +00:00
|
|
|
loc_eulO_size_to_mat4(ct->matrix, loc, eul, size, ct->rotOrder);
|
|
|
|
loc_eulO_size_to_mat4(invmat, loco, eulo, sizo, cob->rotOrder);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* multiply target (parent matrix) by offset (parent inverse) to get
|
|
|
|
* the effect of the parent that will be exherted on the owner
|
|
|
|
*/
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(parmat, invmat, ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* now multiply the parent matrix by the owner matrix to get the
|
|
|
|
* the effect of this constraint (i.e. owner is 'parented' to parent)
|
|
|
|
*/
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tempmat, cob->matrix);
|
2010-09-20 06:47:34 +00:00
|
|
|
mul_m4_m4m4(cob->matrix, tempmat, parmat);
|
|
|
|
|
|
|
|
/* without this, changes to scale and rotation can change location
|
|
|
|
* of a parentless bone or a disconnected bone. Even though its set
|
|
|
|
* to zero above. */
|
|
|
|
if (!(data->flag & CHILDOF_LOCX)) cob->matrix[3][0]= tempmat[3][0];
|
|
|
|
if (!(data->flag & CHILDOF_LOCY)) cob->matrix[3][1]= tempmat[3][1];
|
|
|
|
if (!(data->flag & CHILDOF_LOCZ)) cob->matrix[3][2]= tempmat[3][2];
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_CHILDOF = {
|
|
|
|
CONSTRAINT_TYPE_CHILDOF, /* type */
|
|
|
|
sizeof(bChildOfConstraint), /* size */
|
|
|
|
"ChildOf", /* name */
|
|
|
|
"bChildOfConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
childof_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
childof_new_data, /* new data */
|
|
|
|
childof_get_tars, /* get constraint targets */
|
|
|
|
childof_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get a target matrix */
|
|
|
|
childof_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* -------- TrackTo Constraint ------- */
|
|
|
|
|
|
|
|
static void trackto_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bTrackToConstraint *data= (bTrackToConstraint *)cdata;
|
|
|
|
|
|
|
|
data->reserved1 = TRACK_Y;
|
|
|
|
data->reserved2 = UP_Z;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void trackto_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bTrackToConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int trackto_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bTrackToConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void trackto_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bTrackToConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2006-12-03 18:38:51 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static int basis_cross (int n, int m)
|
2006-12-03 18:38:51 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
switch (n-m) {
|
|
|
|
case 1:
|
|
|
|
case -2:
|
|
|
|
return 1;
|
|
|
|
|
|
|
|
case -1:
|
|
|
|
case 2:
|
|
|
|
return -1;
|
|
|
|
|
|
|
|
default:
|
|
|
|
return 0;
|
|
|
|
}
|
2006-12-03 18:38:51 +00:00
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static void vectomat (float *vec, float *target_up, short axis, short upflag, short flags, float m[][3])
|
2006-12-03 18:38:51 +00:00
|
|
|
{
|
|
|
|
float n[3];
|
|
|
|
float u[3]; /* vector specifying the up axis */
|
|
|
|
float proj[3];
|
|
|
|
float right[3];
|
|
|
|
float neg = -1;
|
2006-12-05 02:06:51 +00:00
|
|
|
int right_index;
|
2010-08-15 15:14:08 +00:00
|
|
|
|
|
|
|
if (normalize_v3_v3(n, vec) == 0.0) {
|
2006-12-03 18:38:51 +00:00
|
|
|
n[0] = 0.0;
|
|
|
|
n[1] = 0.0;
|
|
|
|
n[2] = 1.0;
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (axis > 2) axis -= 3;
|
2009-11-10 20:43:45 +00:00
|
|
|
else negate_v3(n);
|
2006-12-03 18:38:51 +00:00
|
|
|
|
|
|
|
/* n specifies the transformation of the track axis */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (flags & TARGET_Z_UP) {
|
2006-12-03 18:38:51 +00:00
|
|
|
/* target Z axis is the global up axis */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(u, target_up);
|
2006-12-03 18:38:51 +00:00
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* world Z axis is the global up axis */
|
|
|
|
u[0] = 0;
|
|
|
|
u[1] = 0;
|
|
|
|
u[2] = 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* project the up vector onto the plane specified by n */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(proj, u, n); /* first u onto n... */
|
|
|
|
sub_v3_v3v3(proj, u, proj); /* then onto the plane */
|
2006-12-03 18:38:51 +00:00
|
|
|
/* proj specifies the transformation of the up axis */
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
if (normalize_v3(proj) == 0.0) { /* degenerate projection */
|
2006-12-03 18:38:51 +00:00
|
|
|
proj[0] = 0.0;
|
|
|
|
proj[1] = 1.0;
|
|
|
|
proj[2] = 0.0;
|
|
|
|
}
|
|
|
|
|
2007-04-04 13:18:41 +00:00
|
|
|
/* Normalized cross product of n and proj specifies transformation of the right axis */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(right, proj, n);
|
|
|
|
normalize_v3(right);
|
2006-12-03 18:38:51 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (axis != upflag) {
|
2006-12-03 18:38:51 +00:00
|
|
|
right_index = 3 - axis - upflag;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
neg = (float)basis_cross(axis, upflag);
|
|
|
|
|
2006-12-03 18:38:51 +00:00
|
|
|
/* account for up direction, track direction */
|
|
|
|
m[right_index][0] = neg * right[0];
|
|
|
|
m[right_index][1] = neg * right[1];
|
|
|
|
m[right_index][2] = neg * right[2];
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(m[upflag], proj);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(m[axis], n);
|
2006-12-03 18:38:51 +00:00
|
|
|
}
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
/* identity matrix - don't do anything if the two axes are the same */
|
2006-12-03 18:38:51 +00:00
|
|
|
else {
|
2010-08-15 15:14:08 +00:00
|
|
|
unit_m3(m);
|
2006-12-03 18:38:51 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static void trackto_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
2002-10-12 11:37:38 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
bTrackToConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float size[3], vec[3];
|
|
|
|
float totmat[3][3];
|
|
|
|
float tmat[4][4];
|
|
|
|
|
|
|
|
/* Get size property, since ob->size is only the object's own relative size, not its global one */
|
2010-05-27 11:56:31 +00:00
|
|
|
mat4_to_size(size, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* Clear the object's rotation */
|
|
|
|
cob->matrix[0][0]=size[0];
|
|
|
|
cob->matrix[0][1]=0;
|
|
|
|
cob->matrix[0][2]=0;
|
|
|
|
cob->matrix[1][0]=0;
|
|
|
|
cob->matrix[1][1]=size[1];
|
|
|
|
cob->matrix[1][2]=0;
|
|
|
|
cob->matrix[2][0]=0;
|
|
|
|
cob->matrix[2][1]=0;
|
|
|
|
cob->matrix[2][2]=size[2];
|
|
|
|
|
|
|
|
/* targetmat[2] instead of ownermat[2] is passed to vectomat
|
|
|
|
* for backwards compatability it seems... (Aligorith)
|
|
|
|
*/
|
2009-11-10 20:43:45 +00:00
|
|
|
sub_v3_v3v3(vec, cob->matrix[3], ct->matrix[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
vectomat(vec, ct->matrix[2],
|
|
|
|
(short)data->reserved1, (short)data->reserved2,
|
|
|
|
data->flags, totmat);
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tmat, cob->matrix);
|
|
|
|
mul_m4_m3m4(cob->matrix, totmat, tmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
2002-10-12 11:37:38 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static bConstraintTypeInfo CTI_TRACKTO = {
|
|
|
|
CONSTRAINT_TYPE_TRACKTO, /* type */
|
|
|
|
sizeof(bTrackToConstraint), /* size */
|
|
|
|
"TrackTo", /* name */
|
|
|
|
"bTrackToConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
trackto_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
trackto_new_data, /* new data */
|
|
|
|
trackto_get_tars, /* get constraint targets */
|
|
|
|
trackto_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
trackto_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* --------- Inverse-Kinemetics --------- */
|
|
|
|
|
|
|
|
static void kinematic_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bKinematicConstraint *data= (bKinematicConstraint *)cdata;
|
|
|
|
|
|
|
|
data->weight= (float)1.0;
|
|
|
|
data->orientweight= (float)1.0;
|
|
|
|
data->iterations = 500;
|
2009-09-24 21:22:24 +00:00
|
|
|
data->dist= (float)1.0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
data->flag= CONSTRAINT_IK_TIP|CONSTRAINT_IK_STRETCH|CONSTRAINT_IK_POS;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void kinematic_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bKinematicConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* chain target */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
|
|
|
|
/* poletarget */
|
|
|
|
func(con, (ID**)&data->poletar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int kinematic_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bKinematicConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
/* standard target-getting macro for single-target constraints is used twice here */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
|
|
|
SINGLETARGET_GET_TARS(con, data->poletar, data->polesubtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 2;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void kinematic_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bKinematicConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->poletar, data->polesubtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void kinematic_get_tarmat (bConstraint *con, bConstraintOb *cob, bConstraintTarget *ct, float UNUSED(ctime))
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
bKinematicConstraint *data= con->data;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct))
|
2009-04-20 15:06:46 +00:00
|
|
|
constraint_target_to_mat4(cob->scene, ct->tar, ct->subtarget, ct->matrix, CONSTRAINT_SPACE_WORLD, ct->space, con->headtail);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
else if (ct) {
|
|
|
|
if (data->flag & CONSTRAINT_IK_AUTO) {
|
|
|
|
Object *ob= cob->ob;
|
|
|
|
|
|
|
|
if (ob == NULL) {
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
2006-02-19 17:42:20 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
else {
|
|
|
|
float vec[3];
|
|
|
|
/* move grabtarget into world space */
|
2010-07-26 06:34:56 +00:00
|
|
|
mul_v3_m4v3(vec, ob->obmat, data->grabtarget);
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(ct->matrix, ob->obmat);
|
2010-07-26 06:34:56 +00:00
|
|
|
copy_v3_v3(ct->matrix[3], vec);
|
2002-10-12 11:37:38 +00:00
|
|
|
}
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
else
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_KINEMATIC = {
|
|
|
|
CONSTRAINT_TYPE_KINEMATIC, /* type */
|
|
|
|
sizeof(bKinematicConstraint), /* size */
|
|
|
|
"IK", /* name */
|
|
|
|
"bKinematicConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
kinematic_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
kinematic_new_data, /* new data */
|
|
|
|
kinematic_get_tars, /* get constraint targets */
|
|
|
|
kinematic_flush_tars, /* flush constraint targets */
|
|
|
|
kinematic_get_tarmat, /* get target matrix */
|
|
|
|
NULL /* evaluate - solved as separate loop */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* -------- Follow-Path Constraint ---------- */
|
|
|
|
|
|
|
|
static void followpath_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bFollowPathConstraint *data= (bFollowPathConstraint *)cdata;
|
|
|
|
|
|
|
|
data->trackflag = TRACK_Y;
|
|
|
|
data->upflag = UP_Z;
|
|
|
|
data->offset = 0;
|
|
|
|
data->followflag = 0;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void followpath_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bFollowPathConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int followpath_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bFollowPathConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints without subtargets */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGETNS_GET_TARS(con, data->tar, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void followpath_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bFollowPathConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGETNS_FLUSH_TARS(con, data->tar, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void followpath_get_tarmat (bConstraint *con, bConstraintOb *cob, bConstraintTarget *ct, float UNUSED(ctime))
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
bFollowPathConstraint *data= con->data;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
Curve *cu= ct->tar->data;
|
2010-10-08 07:29:08 +00:00
|
|
|
float vec[4], dir[3], radius;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
float totmat[4][4];
|
|
|
|
float curvetime;
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(totmat);
|
|
|
|
unit_m4(ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* note: when creating constraints that follow path, the curve gets the CU_PATH set now,
|
|
|
|
* currently for paths to work it needs to go through the bevlist/displist system (ton)
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* only happens on reload file, but violates depsgraph still... fix! */
|
2009-09-12 14:12:37 +00:00
|
|
|
if (cu->path==NULL || cu->path->data==NULL)
|
2009-01-04 14:14:06 +00:00
|
|
|
makeDispListCurveTypes(cob->scene, ct->tar, 0);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
if (cu->path && cu->path->data) {
|
2010-10-08 07:29:08 +00:00
|
|
|
float quat[4];
|
|
|
|
if ((data->followflag & FOLLOWPATH_STATIC) == 0) {
|
2009-09-08 09:41:15 +00:00
|
|
|
/* animated position along curve depending on time */
|
2009-09-21 12:09:00 +00:00
|
|
|
if (cob->scene)
|
2009-12-14 06:25:42 +00:00
|
|
|
curvetime= bsystem_time(cob->scene, ct->tar, cu->ctime, 0.0) - data->offset;
|
2009-09-21 12:09:00 +00:00
|
|
|
else
|
2009-12-14 06:25:42 +00:00
|
|
|
curvetime= cu->ctime - data->offset;
|
2009-09-08 09:41:15 +00:00
|
|
|
|
|
|
|
/* ctime is now a proper var setting of Curve which gets set by Animato like any other var that's animated,
|
|
|
|
* but this will only work if it actually is animated...
|
|
|
|
*
|
2009-12-14 06:25:42 +00:00
|
|
|
* we divide the curvetime calculated in the previous step by the length of the path, to get a time
|
|
|
|
* factor, which then gets clamped to lie within 0.0 - 1.0 range
|
2009-09-08 09:41:15 +00:00
|
|
|
*/
|
2009-12-14 06:25:42 +00:00
|
|
|
curvetime /= cu->pathlen;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
CLAMP(curvetime, 0.0, 1.0);
|
2003-10-21 13:22:07 +00:00
|
|
|
}
|
2009-09-08 09:41:15 +00:00
|
|
|
else {
|
|
|
|
/* fixed position along curve */
|
2010-02-03 09:05:31 +00:00
|
|
|
curvetime= data->offset_fac;
|
2009-09-08 09:41:15 +00:00
|
|
|
}
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
2010-10-08 07:29:08 +00:00
|
|
|
if ( where_on_path(ct->tar, curvetime, vec, dir, (data->followflag & FOLLOWPATH_FOLLOW) ? quat : NULL, &radius, NULL) ) { /* quat_pt is quat or NULL*/
|
2009-09-08 09:41:15 +00:00
|
|
|
if (data->followflag & FOLLOWPATH_FOLLOW) {
|
2010-10-08 07:29:08 +00:00
|
|
|
#if 0
|
|
|
|
float x1, q[4];
|
2009-12-14 06:25:42 +00:00
|
|
|
vec_to_quat(quat, dir, (short)data->trackflag, (short)data->upflag);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
normalize_v3(dir);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
q[0]= (float)cos(0.5*vec[3]);
|
|
|
|
x1= (float)sin(0.5*vec[3]);
|
|
|
|
q[1]= -x1*dir[0];
|
|
|
|
q[2]= -x1*dir[1];
|
|
|
|
q[3]= -x1*dir[2];
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_qt_qtqt(quat, q, quat);
|
2010-10-08 07:29:08 +00:00
|
|
|
#else
|
|
|
|
quat_apply_track(quat, data->trackflag, data->upflag);
|
|
|
|
#endif
|
|
|
|
|
2009-12-14 06:25:42 +00:00
|
|
|
quat_to_mat4(totmat, quat);
|
2003-10-21 13:22:07 +00:00
|
|
|
}
|
2010-10-08 07:29:08 +00:00
|
|
|
|
2009-09-12 16:25:49 +00:00
|
|
|
if (data->followflag & FOLLOWPATH_RADIUS) {
|
|
|
|
float tmat[4][4], rmat[4][4];
|
2009-11-10 20:43:45 +00:00
|
|
|
scale_m4_fl(tmat, radius);
|
|
|
|
mul_m4_m4m4(rmat, totmat, tmat);
|
|
|
|
copy_m4_m4(totmat, rmat);
|
2009-09-12 16:25:49 +00:00
|
|
|
}
|
2009-11-01 11:29:40 +00:00
|
|
|
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(totmat[3], vec);
|
2007-06-18 07:41:21 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_serie_m4(ct->matrix, ct->tar->obmat, totmat, NULL, NULL, NULL, NULL, NULL, NULL);
|
2007-06-18 07:41:21 +00:00
|
|
|
}
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
else if (ct)
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void followpath_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
/* only evaluate if there is a target */
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float obmat[4][4];
|
2009-09-12 16:25:49 +00:00
|
|
|
float size[3];
|
|
|
|
bFollowPathConstraint *data= con->data;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2009-12-14 06:25:42 +00:00
|
|
|
/* get Object transform (loc/rot/size) to determine transformation from path */
|
|
|
|
// TODO: this used to be local at one point, but is probably more useful as-is
|
|
|
|
copy_m4_m4(obmat, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* get scaling of object before applying constraint */
|
2009-12-14 06:25:42 +00:00
|
|
|
mat4_to_size(size, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* apply targetmat - containing location on path, and rotation */
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_serie_m4(cob->matrix, ct->matrix, obmat, NULL, NULL, NULL, NULL, NULL, NULL);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* un-apply scaling caused by path */
|
2009-09-12 16:25:49 +00:00
|
|
|
if ((data->followflag & FOLLOWPATH_RADIUS)==0) { /* XXX - assume that scale correction means that radius will have some scale error in it - Campbell */
|
|
|
|
float obsize[3];
|
2009-11-01 11:29:40 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
mat4_to_size( obsize,cob->matrix);
|
2009-09-12 16:25:49 +00:00
|
|
|
if (obsize[0])
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[0], size[0] / obsize[0]);
|
2009-09-12 16:25:49 +00:00
|
|
|
if (obsize[1])
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[1], size[1] / obsize[1]);
|
2009-09-12 16:25:49 +00:00
|
|
|
if (obsize[2])
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[2], size[2] / obsize[2]);
|
2009-09-12 16:25:49 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_FOLLOWPATH = {
|
|
|
|
CONSTRAINT_TYPE_FOLLOWPATH, /* type */
|
|
|
|
sizeof(bFollowPathConstraint), /* size */
|
|
|
|
"Follow Path", /* name */
|
|
|
|
"bFollowPathConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
followpath_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
followpath_new_data, /* new data */
|
|
|
|
followpath_get_tars, /* get constraint targets */
|
|
|
|
followpath_flush_tars, /* flush constraint targets */
|
|
|
|
followpath_get_tarmat, /* get target matrix */
|
|
|
|
followpath_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* --------- Limit Location --------- */
|
|
|
|
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void loclimit_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *UNUSED(targets))
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
bLocLimitConstraint *data = con->data;
|
|
|
|
|
|
|
|
if (data->flag & LIMIT_XMIN) {
|
|
|
|
if (cob->matrix[3][0] < data->xmin)
|
|
|
|
cob->matrix[3][0] = data->xmin;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_XMAX) {
|
|
|
|
if (cob->matrix[3][0] > data->xmax)
|
|
|
|
cob->matrix[3][0] = data->xmax;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_YMIN) {
|
|
|
|
if (cob->matrix[3][1] < data->ymin)
|
|
|
|
cob->matrix[3][1] = data->ymin;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_YMAX) {
|
|
|
|
if (cob->matrix[3][1] > data->ymax)
|
|
|
|
cob->matrix[3][1] = data->ymax;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_ZMIN) {
|
|
|
|
if (cob->matrix[3][2] < data->zmin)
|
|
|
|
cob->matrix[3][2] = data->zmin;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_ZMAX) {
|
|
|
|
if (cob->matrix[3][2] > data->zmax)
|
|
|
|
cob->matrix[3][2] = data->zmax;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_LOCLIMIT = {
|
|
|
|
CONSTRAINT_TYPE_LOCLIMIT, /* type */
|
|
|
|
sizeof(bLocLimitConstraint), /* size */
|
|
|
|
"Limit Location", /* name */
|
|
|
|
"bLocLimitConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
NULL, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
NULL, /* new data */
|
|
|
|
NULL, /* get constraint targets */
|
|
|
|
NULL, /* flush constraint targets */
|
|
|
|
NULL, /* get target matrix */
|
|
|
|
loclimit_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* -------- Limit Rotation --------- */
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void rotlimit_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *UNUSED(targets))
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
bRotLimitConstraint *data = con->data;
|
2010-10-26 16:55:38 +00:00
|
|
|
float eul_zero[3]= {0.0f, 0.0f, 0.0f};
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
float loc[3];
|
|
|
|
float eul[3];
|
|
|
|
float size[3];
|
|
|
|
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(loc, cob->matrix[3]);
|
2010-05-27 11:56:31 +00:00
|
|
|
mat4_to_size(size, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2010-10-26 16:55:38 +00:00
|
|
|
/* use compat function because it uses the rotation without axis flipping [#24002] */
|
|
|
|
mat4_to_compatible_eulO(eul, eul_zero, cob->rotOrder, cob->matrix);
|
|
|
|
|
2010-01-25 06:24:05 +00:00
|
|
|
/* constraint data uses radians internally */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* limiting of euler values... */
|
|
|
|
if (data->flag & LIMIT_XROT) {
|
|
|
|
if (eul[0] < data->xmin)
|
|
|
|
eul[0] = data->xmin;
|
2007-04-07 03:32:57 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (eul[0] > data->xmax)
|
|
|
|
eul[0] = data->xmax;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_YROT) {
|
|
|
|
if (eul[1] < data->ymin)
|
|
|
|
eul[1] = data->ymin;
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (eul[1] > data->ymax)
|
|
|
|
eul[1] = data->ymax;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_ZROT) {
|
|
|
|
if (eul[2] < data->zmin)
|
|
|
|
eul[2] = data->zmin;
|
2007-07-21 07:26:15 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (eul[2] > data->zmax)
|
|
|
|
eul[2] = data->zmax;
|
|
|
|
}
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
loc_eulO_size_to_mat4(cob->matrix, loc, eul, size, cob->rotOrder);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2004-09-05 20:21:16 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static bConstraintTypeInfo CTI_ROTLIMIT = {
|
|
|
|
CONSTRAINT_TYPE_ROTLIMIT, /* type */
|
|
|
|
sizeof(bRotLimitConstraint), /* size */
|
|
|
|
"Limit Rotation", /* name */
|
|
|
|
"bRotLimitConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
NULL, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
NULL, /* new data */
|
|
|
|
NULL, /* get constraint targets */
|
|
|
|
NULL, /* flush constraint targets */
|
|
|
|
NULL, /* get target matrix */
|
|
|
|
rotlimit_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* --------- Limit Scaling --------- */
|
|
|
|
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void sizelimit_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *UNUSED(targets))
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
bSizeLimitConstraint *data = con->data;
|
|
|
|
float obsize[3], size[3];
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
mat4_to_size( size,cob->matrix);
|
|
|
|
mat4_to_size( obsize,cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
if (data->flag & LIMIT_XMIN) {
|
|
|
|
if (size[0] < data->xmin)
|
|
|
|
size[0] = data->xmin;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_XMAX) {
|
|
|
|
if (size[0] > data->xmax)
|
|
|
|
size[0] = data->xmax;
|
2002-10-12 11:37:38 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (data->flag & LIMIT_YMIN) {
|
|
|
|
if (size[1] < data->ymin)
|
|
|
|
size[1] = data->ymin;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_YMAX) {
|
|
|
|
if (size[1] > data->ymax)
|
|
|
|
size[1] = data->ymax;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_ZMIN) {
|
|
|
|
if (size[2] < data->zmin)
|
|
|
|
size[2] = data->zmin;
|
|
|
|
}
|
|
|
|
if (data->flag & LIMIT_ZMAX) {
|
|
|
|
if (size[2] > data->zmax)
|
|
|
|
size[2] = data->zmax;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (obsize[0])
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[0], size[0]/obsize[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (obsize[1])
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[1], size[1]/obsize[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (obsize[2])
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[2], size[2]/obsize[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2002-10-12 11:37:38 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static bConstraintTypeInfo CTI_SIZELIMIT = {
|
|
|
|
CONSTRAINT_TYPE_SIZELIMIT, /* type */
|
|
|
|
sizeof(bSizeLimitConstraint), /* size */
|
|
|
|
"Limit Scaling", /* name */
|
|
|
|
"bSizeLimitConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
NULL, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
NULL, /* new data */
|
|
|
|
NULL, /* get constraint targets */
|
|
|
|
NULL, /* flush constraint targets */
|
|
|
|
NULL, /* get target matrix */
|
|
|
|
sizelimit_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* ----------- Copy Location ------------- */
|
|
|
|
|
|
|
|
static void loclike_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bLocateLikeConstraint *data= (bLocateLikeConstraint *)cdata;
|
|
|
|
|
|
|
|
data->flag = LOCLIKE_X|LOCLIKE_Y|LOCLIKE_Z;
|
2002-10-12 11:37:38 +00:00
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void loclike_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bLocateLikeConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int loclike_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bLocateLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static void loclike_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
2002-10-12 11:37:38 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (con && list) {
|
|
|
|
bLocateLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
2002-10-12 11:37:38 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static void loclike_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bLocateLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float offset[3] = {0.0f, 0.0f, 0.0f};
|
|
|
|
|
|
|
|
if (data->flag & LOCLIKE_OFFSET)
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(offset, cob->matrix[3]);
|
2006-09-22 15:57:58 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (data->flag & LOCLIKE_X) {
|
|
|
|
cob->matrix[3][0] = ct->matrix[3][0];
|
2006-09-22 15:57:58 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (data->flag & LOCLIKE_X_INVERT) cob->matrix[3][0] *= -1;
|
|
|
|
cob->matrix[3][0] += offset[0];
|
|
|
|
}
|
|
|
|
if (data->flag & LOCLIKE_Y) {
|
|
|
|
cob->matrix[3][1] = ct->matrix[3][1];
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (data->flag & LOCLIKE_Y_INVERT) cob->matrix[3][1] *= -1;
|
|
|
|
cob->matrix[3][1] += offset[1];
|
|
|
|
}
|
|
|
|
if (data->flag & LOCLIKE_Z) {
|
|
|
|
cob->matrix[3][2] = ct->matrix[3][2];
|
2005-09-07 00:11:39 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (data->flag & LOCLIKE_Z_INVERT) cob->matrix[3][2] *= -1;
|
|
|
|
cob->matrix[3][2] += offset[2];
|
2005-09-07 00:11:39 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
2002-10-12 11:37:38 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static bConstraintTypeInfo CTI_LOCLIKE = {
|
|
|
|
CONSTRAINT_TYPE_LOCLIKE, /* type */
|
|
|
|
sizeof(bLocateLikeConstraint), /* size */
|
|
|
|
"Copy Location", /* name */
|
|
|
|
"bLocateLikeConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
loclike_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
loclike_new_data, /* new data */
|
|
|
|
loclike_get_tars, /* get constraint targets */
|
|
|
|
loclike_flush_tars, /* flush constraint targets */
|
Patch #7767: Constraint Subtargets can now target anywhere on a bone, not just the head or tail
Patch by: Roland Hess (harkyman)
For example, a constraint can be sub-targeted at the 50% (or 31.2% or 85% etc.) point of its target bone, giving you enormous rigging flexibility and removing the need for complex contraptions to do such things as:
- A bone whose base slides only between to points on a rig (CopyLoc with a variable, animated subtarget point)
- Bones that attach to multiple points along another bone (CopyLocs, each with a different head/tail percentage)
- Bones that need to stretch to a point midway between specific spots on two other bones (old way: too crazy to mention; new way: stretch bone between points on end bones, then another stretch to the midpoint of the first stretch)
It is only used for the constraint types for which it is relevant: CopyLoc, TrackTo, StretchTo and MinMax, TrackTo, and Floor.
Notes:
- This is accessed by the Head/Tail number-slider.
- This value can be animated per constraint
- The old "Copy Bone Tail" option for the CopyLoc constraint has been automatically converted to 1.0 Head/Bone values for the affected constraints
- In the code, this value is in the bConstraint struct, so it is available for all constraints, even though only a few implement it.
2007-11-12 04:17:03 +00:00
|
|
|
default_get_tarmat, /* get target matrix */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
loclike_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* ----------- Copy Rotation ------------- */
|
|
|
|
|
|
|
|
static void rotlike_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bRotateLikeConstraint *data= (bRotateLikeConstraint *)cdata;
|
|
|
|
|
|
|
|
data->flag = ROTLIKE_X|ROTLIKE_Y|ROTLIKE_Z;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void rotlike_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bChildOfConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int rotlike_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bRotateLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void rotlike_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bRotateLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void rotlike_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bRotateLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float loc[3];
|
|
|
|
float eul[3], obeul[3];
|
|
|
|
float size[3];
|
|
|
|
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(loc, cob->matrix[3]);
|
2010-05-27 11:56:31 +00:00
|
|
|
mat4_to_size(size, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2009-09-20 12:54:30 +00:00
|
|
|
/* to allow compatible rotations, must get both rotations in the order of the owner... */
|
2010-05-27 11:56:31 +00:00
|
|
|
mat4_to_eulO(obeul, cob->rotOrder, cob->matrix);
|
2010-09-24 17:49:33 +00:00
|
|
|
/* we must get compatible eulers from the beginning because some of them can be modified below (see bug #21875) */
|
|
|
|
mat4_to_compatible_eulO(eul, obeul, cob->rotOrder, ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2007-10-30 11:18:50 +00:00
|
|
|
if ((data->flag & ROTLIKE_X)==0)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
eul[0] = obeul[0];
|
2007-10-30 11:18:50 +00:00
|
|
|
else {
|
|
|
|
if (data->flag & ROTLIKE_OFFSET)
|
2010-04-25 03:34:16 +00:00
|
|
|
rotate_eulO(eul, cob->rotOrder, 'X', obeul[0]);
|
2007-10-30 11:18:50 +00:00
|
|
|
|
|
|
|
if (data->flag & ROTLIKE_X_INVERT)
|
|
|
|
eul[0] *= -1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
2007-10-30 11:18:50 +00:00
|
|
|
if ((data->flag & ROTLIKE_Y)==0)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
eul[1] = obeul[1];
|
2007-10-30 11:18:50 +00:00
|
|
|
else {
|
|
|
|
if (data->flag & ROTLIKE_OFFSET)
|
2010-04-25 03:34:16 +00:00
|
|
|
rotate_eulO(eul, cob->rotOrder, 'Y', obeul[1]);
|
2007-10-30 11:18:50 +00:00
|
|
|
|
|
|
|
if (data->flag & ROTLIKE_Y_INVERT)
|
|
|
|
eul[1] *= -1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
2007-10-30 11:18:50 +00:00
|
|
|
if ((data->flag & ROTLIKE_Z)==0)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
eul[2] = obeul[2];
|
2007-10-30 11:18:50 +00:00
|
|
|
else {
|
|
|
|
if (data->flag & ROTLIKE_OFFSET)
|
2010-04-25 03:34:16 +00:00
|
|
|
rotate_eulO(eul, cob->rotOrder, 'Z', obeul[2]);
|
2007-10-30 11:18:50 +00:00
|
|
|
|
|
|
|
if (data->flag & ROTLIKE_Z_INVERT)
|
|
|
|
eul[2] *= -1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
2010-09-24 17:49:33 +00:00
|
|
|
/* good to make eulers compatible again, since we don't know how much they were changed above */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
compatible_eul(eul, obeul);
|
2009-11-10 20:43:45 +00:00
|
|
|
loc_eulO_size_to_mat4(cob->matrix, loc, eul, size, cob->rotOrder);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_ROTLIKE = {
|
|
|
|
CONSTRAINT_TYPE_ROTLIKE, /* type */
|
|
|
|
sizeof(bRotateLikeConstraint), /* size */
|
|
|
|
"Copy Rotation", /* name */
|
|
|
|
"bRotateLikeConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
rotlike_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
rotlike_new_data, /* new data */
|
|
|
|
rotlike_get_tars, /* get constraint targets */
|
|
|
|
rotlike_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
rotlike_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* ---------- Copy Scaling ---------- */
|
|
|
|
|
|
|
|
static void sizelike_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bSizeLikeConstraint *data= (bSizeLikeConstraint *)cdata;
|
|
|
|
|
|
|
|
data->flag = SIZELIKE_X|SIZELIKE_Y|SIZELIKE_Z;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void sizelike_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bSizeLikeConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int sizelike_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bSizeLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void sizelike_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bSizeLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void sizelike_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bSizeLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float obsize[3], size[3];
|
|
|
|
|
2010-01-02 04:14:17 +00:00
|
|
|
mat4_to_size(size, ct->matrix);
|
|
|
|
mat4_to_size(obsize, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2007-10-31 10:04:57 +00:00
|
|
|
if ((data->flag & SIZELIKE_X) && (obsize[0] != 0)) {
|
|
|
|
if (data->flag & SIZELIKE_OFFSET) {
|
|
|
|
size[0] += (obsize[0] - 1.0f);
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[0], size[0] / obsize[0]);
|
2007-10-31 10:04:57 +00:00
|
|
|
}
|
|
|
|
else
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[0], size[0] / obsize[0]);
|
2007-10-31 10:04:57 +00:00
|
|
|
}
|
|
|
|
if ((data->flag & SIZELIKE_Y) && (obsize[1] != 0)) {
|
|
|
|
if (data->flag & SIZELIKE_OFFSET) {
|
|
|
|
size[1] += (obsize[1] - 1.0f);
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[1], size[1] / obsize[1]);
|
2007-10-31 10:04:57 +00:00
|
|
|
}
|
|
|
|
else
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[1], size[1] / obsize[1]);
|
2007-10-31 10:04:57 +00:00
|
|
|
}
|
|
|
|
if ((data->flag & SIZELIKE_Z) && (obsize[2] != 0)) {
|
|
|
|
if (data->flag & SIZELIKE_OFFSET) {
|
|
|
|
size[2] += (obsize[2] - 1.0f);
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[2], size[2] / obsize[2]);
|
2007-10-31 10:04:57 +00:00
|
|
|
}
|
|
|
|
else
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_v3_fl(cob->matrix[2], size[2] / obsize[2]);
|
2007-10-31 10:04:57 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_SIZELIKE = {
|
|
|
|
CONSTRAINT_TYPE_SIZELIKE, /* type */
|
|
|
|
sizeof(bSizeLikeConstraint), /* size */
|
|
|
|
"Copy Scale", /* name */
|
|
|
|
"bSizeLikeConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
sizelike_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
sizelike_new_data, /* new data */
|
|
|
|
sizelike_get_tars, /* get constraint targets */
|
|
|
|
sizelike_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
sizelike_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
2010-01-02 04:14:17 +00:00
|
|
|
/* ----------- Copy Transforms ------------- */
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void translike_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bTransLikeConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2010-01-02 04:14:17 +00:00
|
|
|
static int translike_get_tars (bConstraint *con, ListBase *list)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bTransLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void translike_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bTransLikeConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void translike_evaluate (bConstraint *UNUSED(con), bConstraintOb *cob, ListBase *targets)
|
2010-01-02 04:14:17 +00:00
|
|
|
{
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
/* just copy the entire transform matrix of the target */
|
|
|
|
copy_m4_m4(cob->matrix, ct->matrix);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_TRANSLIKE = {
|
|
|
|
CONSTRAINT_TYPE_TRANSLIKE, /* type */
|
|
|
|
sizeof(bTransLikeConstraint), /* size */
|
|
|
|
"Copy Transforms", /* name */
|
|
|
|
"bTransLikeConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
translike_id_looper, /* id looper */
|
2010-01-02 04:14:17 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
NULL, /* new data */
|
|
|
|
translike_get_tars, /* get constraint targets */
|
|
|
|
translike_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
translike_evaluate /* evaluate */
|
|
|
|
};
|
2008-10-28 18:47:13 +00:00
|
|
|
|
2010-03-16 12:55:56 +00:00
|
|
|
/* ---------- Maintain Volume ---------- */
|
|
|
|
|
|
|
|
static void samevolume_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bSameVolumeConstraint *data= (bSameVolumeConstraint *)cdata;
|
|
|
|
|
|
|
|
data->flag = SAMEVOL_Y;
|
|
|
|
data->volume = 1.0f;
|
|
|
|
}
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void samevolume_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *UNUSED(targets))
|
2010-03-16 12:55:56 +00:00
|
|
|
{
|
|
|
|
bSameVolumeConstraint *data= con->data;
|
|
|
|
|
2010-05-27 11:56:31 +00:00
|
|
|
float volume = data->volume;
|
|
|
|
float fac = 1.0f;
|
2010-03-16 12:55:56 +00:00
|
|
|
float obsize[3];
|
|
|
|
|
|
|
|
mat4_to_size(obsize, cob->matrix);
|
2010-05-27 11:56:31 +00:00
|
|
|
|
|
|
|
/* calculate normalising scale factor for non-essential values */
|
|
|
|
if (obsize[data->flag] != 0)
|
|
|
|
fac = sqrt(volume / obsize[data->flag]) / obsize[data->flag];
|
|
|
|
|
|
|
|
/* apply scaling factor to the channels not being kept */
|
2010-03-16 12:55:56 +00:00
|
|
|
switch (data->flag) {
|
|
|
|
case SAMEVOL_X:
|
2010-05-27 11:56:31 +00:00
|
|
|
mul_v3_fl(cob->matrix[1], fac);
|
|
|
|
mul_v3_fl(cob->matrix[2], fac);
|
2010-03-16 12:55:56 +00:00
|
|
|
break;
|
|
|
|
case SAMEVOL_Y:
|
2010-05-27 11:56:31 +00:00
|
|
|
mul_v3_fl(cob->matrix[0], fac);
|
|
|
|
mul_v3_fl(cob->matrix[2], fac);
|
2010-03-16 12:55:56 +00:00
|
|
|
break;
|
|
|
|
case SAMEVOL_Z:
|
2010-05-27 11:56:31 +00:00
|
|
|
mul_v3_fl(cob->matrix[0], fac);
|
|
|
|
mul_v3_fl(cob->matrix[1], fac);
|
2010-03-16 12:55:56 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_SAMEVOL = {
|
|
|
|
CONSTRAINT_TYPE_SAMEVOL, /* type */
|
|
|
|
sizeof(bSameVolumeConstraint), /* size */
|
|
|
|
"Maintain Volume", /* name */
|
|
|
|
"bSameVolumeConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
|
|
|
NULL, /* id looper */
|
|
|
|
NULL, /* copy data */
|
|
|
|
samevolume_new_data, /* new data */
|
|
|
|
NULL, /* get constraint targets */
|
|
|
|
NULL, /* flush constraint targets */
|
|
|
|
NULL, /* get target matrix */
|
|
|
|
samevolume_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* ----------- Python Constraint -------------- */
|
|
|
|
|
|
|
|
static void pycon_free (bConstraint *con)
|
|
|
|
{
|
|
|
|
bPythonConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* id-properties */
|
|
|
|
IDP_FreeProperty(data->prop);
|
|
|
|
MEM_freeN(data->prop);
|
|
|
|
|
|
|
|
/* multiple targets */
|
2007-10-23 08:23:57 +00:00
|
|
|
BLI_freelistN(&data->targets);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void pycon_relink (bConstraint *con)
|
|
|
|
{
|
|
|
|
bPythonConstraint *data= con->data;
|
|
|
|
|
|
|
|
ID_NEW(data->text);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void pycon_copy (bConstraint *con, bConstraint *srccon)
|
|
|
|
{
|
|
|
|
bPythonConstraint *pycon = (bPythonConstraint *)con->data;
|
|
|
|
bPythonConstraint *opycon = (bPythonConstraint *)srccon->data;
|
|
|
|
|
|
|
|
pycon->prop = IDP_CopyProperty(opycon->prop);
|
2008-01-01 18:29:19 +00:00
|
|
|
BLI_duplicatelist(&pycon->targets, &opycon->targets);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void pycon_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bPythonConstraint *data= (bPythonConstraint *)cdata;
|
|
|
|
|
|
|
|
/* everything should be set correctly by calloc, except for the prop->type constant.*/
|
|
|
|
data->prop = MEM_callocN(sizeof(IDProperty), "PyConstraintProps");
|
|
|
|
data->prop->type = IDP_GROUP;
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int pycon_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bPythonConstraint *data= con->data;
|
|
|
|
|
|
|
|
list->first = data->targets.first;
|
|
|
|
list->last = data->targets.last;
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return data->tarnum;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void pycon_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bPythonConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* targets */
|
|
|
|
for (ct= data->targets.first; ct; ct= ct->next)
|
|
|
|
func(con, (ID**)&ct->tar, userdata);
|
|
|
|
|
|
|
|
/* script */
|
|
|
|
func(con, (ID**)&data->text, userdata);
|
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* Whether this approach is maintained remains to be seen (aligorith) */
|
2010-10-16 14:32:17 +00:00
|
|
|
static void pycon_get_tarmat (bConstraint *con, bConstraintOb *cob, bConstraintTarget *ct, float UNUSED(ctime))
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
2010-04-12 00:36:50 +00:00
|
|
|
#ifndef DISABLE_PYTHON
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
bPythonConstraint *data= con->data;
|
2010-04-12 00:36:50 +00:00
|
|
|
#endif
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
/* special exception for curves - depsgraph issues */
|
|
|
|
if (ct->tar->type == OB_CURVE) {
|
|
|
|
Curve *cu= ct->tar->data;
|
2002-10-12 11:37:38 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* this check is to make sure curve objects get updated on file load correctly.*/
|
|
|
|
if (cu->path==NULL || cu->path->data==NULL) /* only happens on reload file, but violates depsgraph still... fix! */
|
2009-01-04 14:14:06 +00:00
|
|
|
makeDispListCurveTypes(cob->scene, ct->tar, 0);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* firstly calculate the matrix the normal way, then let the py-function override
|
|
|
|
* this matrix if it needs to do so
|
|
|
|
*/
|
2009-04-20 15:06:46 +00:00
|
|
|
constraint_target_to_mat4(cob->scene, ct->tar, ct->subtarget, ct->matrix, CONSTRAINT_SPACE_WORLD, ct->space, con->headtail);
|
2008-09-19 12:43:21 +00:00
|
|
|
|
|
|
|
/* only execute target calculation if allowed */
|
2008-10-28 18:47:13 +00:00
|
|
|
#ifndef DISABLE_PYTHON
|
2010-02-27 01:47:46 +00:00
|
|
|
if (G.f & G_SCRIPT_AUTOEXEC)
|
2008-09-19 12:43:21 +00:00
|
|
|
BPY_pyconstraint_target(data, ct);
|
2008-10-28 18:47:13 +00:00
|
|
|
#endif
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
else if (ct)
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void pycon_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
2008-10-28 18:47:13 +00:00
|
|
|
#ifdef DISABLE_PYTHON
|
2010-10-23 16:03:31 +00:00
|
|
|
(void)con; (void)cob; (void)targets; /* unused */
|
2008-10-28 18:47:13 +00:00
|
|
|
return;
|
|
|
|
#else
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
bPythonConstraint *data= con->data;
|
|
|
|
|
2008-09-19 12:43:21 +00:00
|
|
|
/* only evaluate in python if we're allowed to do so */
|
2010-02-27 01:47:46 +00:00
|
|
|
if ((G.f & G_SCRIPT_AUTOEXEC)==0) return;
|
2008-06-01 16:13:04 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* currently removed, until I this can be re-implemented for multiple targets */
|
|
|
|
#if 0
|
|
|
|
/* Firstly, run the 'driver' function which has direct access to the objects involved
|
|
|
|
* Technically, this is potentially dangerous as users may abuse this and cause dependency-problems,
|
|
|
|
* but it also allows certain 'clever' rigging hacks to work.
|
|
|
|
*/
|
|
|
|
BPY_pyconstraint_driver(data, cob, targets);
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/* Now, run the actual 'constraint' function, which should only access the matrices */
|
|
|
|
BPY_pyconstraint_eval(data, cob, targets);
|
2008-10-28 18:47:13 +00:00
|
|
|
#endif /* DISABLE_PYTHON */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_PYTHON = {
|
|
|
|
CONSTRAINT_TYPE_PYTHON, /* type */
|
|
|
|
sizeof(bPythonConstraint), /* size */
|
|
|
|
"Script", /* name */
|
|
|
|
"bPythonConstraint", /* struct name */
|
|
|
|
pycon_free, /* free data */
|
|
|
|
pycon_relink, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
pycon_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
pycon_copy, /* copy data */
|
|
|
|
pycon_new_data, /* new data */
|
|
|
|
pycon_get_tars, /* get constraint targets */
|
|
|
|
NULL, /* flush constraint targets */
|
|
|
|
pycon_get_tarmat, /* get target matrix */
|
|
|
|
pycon_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* -------- Action Constraint ----------- */
|
|
|
|
|
|
|
|
static void actcon_relink (bConstraint *con)
|
|
|
|
{
|
|
|
|
bActionConstraint *data= con->data;
|
|
|
|
ID_NEW(data->act);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void actcon_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bActionConstraint *data= (bActionConstraint *)cdata;
|
|
|
|
|
|
|
|
/* set type to 20 (Loc X), as 0 is Rot X for backwards compatability */
|
|
|
|
data->type = 20;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void actcon_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bActionConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
|
|
|
|
/* action */
|
|
|
|
func(con, (ID**)&data->act, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int actcon_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bActionConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void actcon_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bActionConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void actcon_get_tarmat (bConstraint *con, bConstraintOb *cob, bConstraintTarget *ct, float UNUSED(ctime))
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
extern void chan_calc_mat(bPoseChannel *chan);
|
|
|
|
bActionConstraint *data = con->data;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float tempmat[4][4], vec[3];
|
|
|
|
float s, t;
|
|
|
|
short axis;
|
|
|
|
|
|
|
|
/* initialise return matrix */
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* get the transform matrix of the target */
|
2009-04-20 15:06:46 +00:00
|
|
|
constraint_target_to_mat4(cob->scene, ct->tar, ct->subtarget, tempmat, CONSTRAINT_SPACE_WORLD, ct->space, con->headtail);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* determine where in transform range target is */
|
|
|
|
/* data->type is mapped as follows for backwards compatability:
|
|
|
|
* 00,01,02 - rotation (it used to be like this)
|
|
|
|
* 10,11,12 - scaling
|
|
|
|
* 20,21,22 - location
|
|
|
|
*/
|
|
|
|
if (data->type < 10) {
|
|
|
|
/* extract rotation (is in whatever space target should be in) */
|
2009-11-14 02:50:45 +00:00
|
|
|
mat4_to_eul(vec, tempmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
vec[0] *= (float)(180.0/M_PI);
|
|
|
|
vec[1] *= (float)(180.0/M_PI);
|
|
|
|
vec[2] *= (float)(180.0/M_PI);
|
|
|
|
axis= data->type;
|
|
|
|
}
|
|
|
|
else if (data->type < 20) {
|
|
|
|
/* extract scaling (is in whatever space target should be in) */
|
2009-11-14 02:50:45 +00:00
|
|
|
mat4_to_size(vec, tempmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
axis= data->type - 10;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* extract location */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(vec, tempmat[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
axis= data->type - 20;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Target defines the animation */
|
|
|
|
s = (vec[axis]-data->min) / (data->max-data->min);
|
|
|
|
CLAMP(s, 0, 1);
|
2009-11-14 02:50:45 +00:00
|
|
|
t = (s * (data->end-data->start)) + data->start;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2009-01-29 23:27:24 +00:00
|
|
|
if (G.f & G_DEBUG)
|
|
|
|
printf("do Action Constraint %s - Ob %s Pchan %s \n", con->name, cob->ob->id.name+2, (cob->pchan)?cob->pchan->name:NULL);
|
2009-01-29 11:22:34 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* Get the appropriate information from the action */
|
2007-10-22 02:43:07 +00:00
|
|
|
if (cob->type == CONSTRAINT_OBTYPE_BONE) {
|
2009-01-29 11:22:34 +00:00
|
|
|
Object workob;
|
2007-10-22 02:43:07 +00:00
|
|
|
bPose *pose;
|
|
|
|
bPoseChannel *pchan, *tchan;
|
|
|
|
|
|
|
|
/* make a temporary pose and evaluate using that */
|
|
|
|
pose = MEM_callocN(sizeof(bPose), "pose");
|
|
|
|
|
2009-11-14 02:50:45 +00:00
|
|
|
/* make a copy of the bone of interest in the temp pose before evaluating action, so that it can get set
|
|
|
|
* - we need to manually copy over a few settings, including rotation order, otherwise this fails
|
|
|
|
*/
|
2007-10-22 02:43:07 +00:00
|
|
|
pchan = cob->pchan;
|
2009-11-14 02:50:45 +00:00
|
|
|
|
2007-10-22 02:43:07 +00:00
|
|
|
tchan= verify_pose_channel(pose, pchan->name);
|
2009-11-14 02:50:45 +00:00
|
|
|
tchan->rotmode= pchan->rotmode;
|
2007-10-22 02:43:07 +00:00
|
|
|
|
2009-01-29 11:22:34 +00:00
|
|
|
/* evaluate action using workob (it will only set the PoseChannel in question) */
|
2009-04-15 12:59:49 +00:00
|
|
|
what_does_obaction(cob->scene, cob->ob, &workob, pose, data->act, pchan->name, t);
|
2007-10-22 02:43:07 +00:00
|
|
|
|
2009-01-29 11:22:34 +00:00
|
|
|
/* convert animation to matrices for use here */
|
|
|
|
chan_calc_mat(tchan);
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(ct->matrix, tchan->chan_mat);
|
2007-10-22 02:43:07 +00:00
|
|
|
|
|
|
|
/* Clean up */
|
== Bone Groups ==
I'm committing some work-in-progress code for "bone groups" now, as I there have been are some major bugs caused by the timeoffset stuff (some of my test files were not loading, and other files were showing all sorts of weird problems).
Anyway, in this commit, the following things for "bone groups" have been done:
* Bone groups are stored per armature (internally, this is per bPose block)
* Added controls for editing bone-groups per armature - "add", "remove", "rename". These can be found in the "Links and Materials" panel in PoseMode, beside the settings for PoseLib.
* Reorganised buttons for editing selected bones in PoseMode. I've replaced the "dist" and "weight" buttons (they existed in EditMode anyway) with a menu to choose the bone-group and the custom-shape-ob field. In the place of the old custom-shape-ob field, I've restored the "Hide" button. This might break muscle-memory a bit, but there isn't a lot of space to play with there.
Some stuff I'd been originally planning to do before committing:
* When adding keyframes for bones, an action-group with the same name as the bone's group will be added to the action, and the action-channel will be made a member of that.
* New action/bone groups have unique names (renaming/adding new should check if name exists before assigning it)
* There's a setting under Bone-Groups stuff which sets which custom-colour set is used to colour that group's bones. Currently, this is non-functional, as the necessary drawing code for armatures is not in place yet.
2008-01-20 02:55:35 +00:00
|
|
|
free_pose(pose);
|
2007-10-22 02:43:07 +00:00
|
|
|
}
|
|
|
|
else if (cob->type == CONSTRAINT_OBTYPE_OBJECT) {
|
2008-12-24 11:08:15 +00:00
|
|
|
Object workob;
|
2009-01-29 11:22:34 +00:00
|
|
|
|
2007-10-22 02:43:07 +00:00
|
|
|
/* evaluate using workob */
|
2009-04-15 12:59:49 +00:00
|
|
|
// FIXME: we don't have any consistent standards on limiting effects on object...
|
|
|
|
what_does_obaction(cob->scene, cob->ob, &workob, NULL, data->act, NULL, t);
|
2007-10-22 02:43:07 +00:00
|
|
|
object_to_mat4(&workob, ct->matrix);
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* behaviour undefined... */
|
|
|
|
puts("Error: unknown owner type for Action Constraint");
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void actcon_evaluate (bConstraint *UNUSED(con), bConstraintOb *cob, ListBase *targets)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float temp[4][4];
|
|
|
|
|
|
|
|
/* Nice and simple... we just need to multiply the matrices, as the get_target_matrix
|
|
|
|
* function has already taken care of everything else.
|
|
|
|
*/
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(temp, cob->matrix);
|
|
|
|
mul_m4_m4m4(cob->matrix, ct->matrix, temp);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_ACTION = {
|
|
|
|
CONSTRAINT_TYPE_ACTION, /* type */
|
|
|
|
sizeof(bActionConstraint), /* size */
|
|
|
|
"Action", /* name */
|
|
|
|
"bActionConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
actcon_relink, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
actcon_id_looper, /* id looper */
|
2010-02-08 17:30:35 +00:00
|
|
|
NULL, /* copy data */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
actcon_new_data, /* new data */
|
|
|
|
actcon_get_tars, /* get constraint targets */
|
|
|
|
actcon_flush_tars, /* flush constraint targets */
|
|
|
|
actcon_get_tarmat, /* get target matrix */
|
|
|
|
actcon_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* --------- Locked Track ---------- */
|
|
|
|
|
|
|
|
static void locktrack_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bLockTrackConstraint *data= (bLockTrackConstraint *)cdata;
|
|
|
|
|
|
|
|
data->trackflag = TRACK_Y;
|
2007-10-26 11:37:29 +00:00
|
|
|
data->lockflag = LOCK_Z;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void locktrack_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bLockTrackConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int locktrack_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bLockTrackConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void locktrack_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bLockTrackConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void locktrack_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bLockTrackConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float vec[3],vec2[3];
|
|
|
|
float totmat[3][3];
|
|
|
|
float tmpmat[3][3];
|
|
|
|
float invmat[3][3];
|
|
|
|
float tmat[4][4];
|
|
|
|
float mdet;
|
|
|
|
|
|
|
|
/* Vector object -> target */
|
2009-11-10 20:43:45 +00:00
|
|
|
sub_v3_v3v3(vec, ct->matrix[3], cob->matrix[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
switch (data->lockflag){
|
|
|
|
case LOCK_X: /* LOCK X */
|
|
|
|
{
|
|
|
|
switch (data->trackflag) {
|
|
|
|
case TRACK_Y: /* LOCK X TRACK Y */
|
|
|
|
{
|
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[0]);
|
|
|
|
sub_v3_v3v3(totmat[1], vec, vec2);
|
|
|
|
normalize_v3(totmat[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the x axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[0], cob->matrix[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[2], totmat[0], totmat[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case TRACK_Z: /* LOCK X TRACK Z */
|
|
|
|
{
|
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[0]);
|
|
|
|
sub_v3_v3v3(totmat[2], vec, vec2);
|
|
|
|
normalize_v3(totmat[2]);
|
2007-10-26 11:37:29 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* the x axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[0], cob->matrix[0]);
|
2007-10-26 11:37:29 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* the z axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[1], totmat[2], totmat[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case TRACK_nY: /* LOCK X TRACK -Y */
|
|
|
|
{
|
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[0]);
|
|
|
|
sub_v3_v3v3(totmat[1], vec, vec2);
|
|
|
|
normalize_v3(totmat[1]);
|
|
|
|
negate_v3(totmat[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the x axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[0], cob->matrix[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[2], totmat[0], totmat[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case TRACK_nZ: /* LOCK X TRACK -Z */
|
|
|
|
{
|
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[0]);
|
|
|
|
sub_v3_v3v3(totmat[2], vec, vec2);
|
|
|
|
normalize_v3(totmat[2]);
|
|
|
|
negate_v3(totmat[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the x axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[0], cob->matrix[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[1], totmat[2], totmat[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
{
|
2010-08-15 15:14:08 +00:00
|
|
|
unit_m3(totmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
2002-10-12 11:37:38 +00:00
|
|
|
}
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
|
|
|
case LOCK_Y: /* LOCK Y */
|
2003-10-21 13:22:07 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
switch (data->trackflag) {
|
|
|
|
case TRACK_X: /* LOCK Y TRACK X */
|
2007-07-03 00:58:38 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[1]);
|
|
|
|
sub_v3_v3v3(totmat[0], vec, vec2);
|
|
|
|
normalize_v3(totmat[0]);
|
2007-07-03 00:58:38 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* the y axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[1], cob->matrix[1]);
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* the z axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[2], totmat[0], totmat[1]);
|
2007-07-03 00:58:38 +00:00
|
|
|
}
|
2003-10-21 13:22:07 +00:00
|
|
|
break;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
case TRACK_Z: /* LOCK Y TRACK Z */
|
2007-07-03 00:58:38 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[1]);
|
|
|
|
sub_v3_v3v3(totmat[2], vec, vec2);
|
|
|
|
normalize_v3(totmat[2]);
|
2009-10-30 06:33:40 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* the y axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[1], cob->matrix[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[0], totmat[1], totmat[2]);
|
2007-07-03 00:58:38 +00:00
|
|
|
}
|
2003-10-21 13:22:07 +00:00
|
|
|
break;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
case TRACK_nX: /* LOCK Y TRACK -X */
|
2007-07-03 00:58:38 +00:00
|
|
|
{
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[1]);
|
|
|
|
sub_v3_v3v3(totmat[0], vec, vec2);
|
|
|
|
normalize_v3(totmat[0]);
|
|
|
|
negate_v3(totmat[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the y axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[1], cob->matrix[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[2], totmat[0], totmat[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case TRACK_nZ: /* LOCK Y TRACK -Z */
|
|
|
|
{
|
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[1]);
|
|
|
|
sub_v3_v3v3(totmat[2], vec, vec2);
|
|
|
|
normalize_v3(totmat[2]);
|
|
|
|
negate_v3(totmat[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the y axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[1], cob->matrix[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[0], totmat[1], totmat[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
{
|
2010-08-15 15:14:08 +00:00
|
|
|
unit_m3(totmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
case LOCK_Z: /* LOCK Z */
|
|
|
|
{
|
|
|
|
switch (data->trackflag) {
|
|
|
|
case TRACK_X: /* LOCK Z TRACK X */
|
|
|
|
{
|
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[2]);
|
|
|
|
sub_v3_v3v3(totmat[0], vec, vec2);
|
|
|
|
normalize_v3(totmat[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[2], cob->matrix[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the x axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[1], totmat[2], totmat[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case TRACK_Y: /* LOCK Z TRACK Y */
|
|
|
|
{
|
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[2]);
|
|
|
|
sub_v3_v3v3(totmat[1], vec, vec2);
|
|
|
|
normalize_v3(totmat[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[2], cob->matrix[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the x axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[0], totmat[1], totmat[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case TRACK_nX: /* LOCK Z TRACK -X */
|
|
|
|
{
|
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[2]);
|
|
|
|
sub_v3_v3v3(totmat[0], vec, vec2);
|
|
|
|
normalize_v3(totmat[0]);
|
|
|
|
negate_v3(totmat[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[2], cob->matrix[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the x axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[1], totmat[2], totmat[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case TRACK_nY: /* LOCK Z TRACK -Y */
|
|
|
|
{
|
|
|
|
/* Projection of Vector on the plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
project_v3_v3v3(vec2, vec, cob->matrix[2]);
|
|
|
|
sub_v3_v3v3(totmat[1], vec, vec2);
|
|
|
|
normalize_v3(totmat[1]);
|
|
|
|
negate_v3(totmat[1]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the z axis is fixed */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[2], cob->matrix[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* the x axis gets mapped onto a third orthogonal vector */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(totmat[0], totmat[1], totmat[2]);
|
2007-07-03 00:58:38 +00:00
|
|
|
}
|
2003-10-21 13:22:07 +00:00
|
|
|
break;
|
|
|
|
default:
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
2010-08-15 15:14:08 +00:00
|
|
|
unit_m3(totmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
break;
|
|
|
|
default:
|
2009-10-30 06:33:40 +00:00
|
|
|
{
|
2010-08-15 15:14:08 +00:00
|
|
|
unit_m3(totmat);
|
2009-10-30 06:33:40 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
|
|
|
}
|
|
|
|
/* Block to keep matrix heading */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_m3_m4(tmpmat, cob->matrix);
|
|
|
|
normalize_m3(tmpmat);
|
2009-11-10 20:43:45 +00:00
|
|
|
invert_m3_m3(invmat, tmpmat);
|
|
|
|
mul_m3_m3m3(tmpmat, totmat, invmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
totmat[0][0] = tmpmat[0][0];totmat[0][1] = tmpmat[0][1];totmat[0][2] = tmpmat[0][2];
|
|
|
|
totmat[1][0] = tmpmat[1][0];totmat[1][1] = tmpmat[1][1];totmat[1][2] = tmpmat[1][2];
|
|
|
|
totmat[2][0] = tmpmat[2][0];totmat[2][1] = tmpmat[2][1];totmat[2][2] = tmpmat[2][2];
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tmat, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
mdet = determinant_m3( totmat[0][0],totmat[0][1],totmat[0][2],
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
totmat[1][0],totmat[1][1],totmat[1][2],
|
|
|
|
totmat[2][0],totmat[2][1],totmat[2][2]);
|
|
|
|
if (mdet==0) {
|
2010-08-15 15:14:08 +00:00
|
|
|
unit_m3(totmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* apply out transformaton to the object */
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m3m4(cob->matrix, totmat, tmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_LOCKTRACK = {
|
|
|
|
CONSTRAINT_TYPE_LOCKTRACK, /* type */
|
|
|
|
sizeof(bLockTrackConstraint), /* size */
|
|
|
|
"Locked Track", /* name */
|
|
|
|
"bLockTrackConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
locktrack_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
locktrack_new_data, /* new data */
|
|
|
|
locktrack_get_tars, /* get constraint targets */
|
|
|
|
locktrack_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
locktrack_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
2008-01-25 11:23:36 +00:00
|
|
|
/* ---------- Limit Distance Constraint ----------- */
|
|
|
|
|
|
|
|
static void distlimit_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bDistLimitConstraint *data= (bDistLimitConstraint *)cdata;
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
data->dist= 0.0f;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void distlimit_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bDistLimitConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
2008-01-25 11:23:36 +00:00
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int distlimit_get_tars (bConstraint *con, ListBase *list)
|
2008-01-25 11:23:36 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bDistLimitConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
2008-01-25 11:23:36 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
2008-01-25 11:23:36 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void distlimit_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bDistLimitConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void distlimit_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bDistLimitConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
/* only evaluate if there is a target */
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float dvec[3], dist=0.0f, sfac=1.0f;
|
|
|
|
short clamp_surf= 0;
|
|
|
|
|
|
|
|
/* calculate our current distance from the target */
|
2009-11-10 20:43:45 +00:00
|
|
|
dist= len_v3v3(cob->matrix[3], ct->matrix[3]);
|
2008-01-25 11:23:36 +00:00
|
|
|
|
|
|
|
/* set distance (flag is only set when user demands it) */
|
|
|
|
if (data->dist == 0)
|
|
|
|
data->dist= dist;
|
|
|
|
|
|
|
|
/* check if we're which way to clamp from, and calculate interpolation factor (if needed) */
|
|
|
|
if (data->mode == LIMITDIST_OUTSIDE) {
|
|
|
|
/* if inside, then move to surface */
|
|
|
|
if (dist <= data->dist) {
|
|
|
|
clamp_surf= 1;
|
|
|
|
sfac= data->dist / dist;
|
|
|
|
}
|
|
|
|
/* if soft-distance is enabled, start fading once owner is dist+softdist from the target */
|
|
|
|
else if (data->flag & LIMITDIST_USESOFT) {
|
|
|
|
if (dist <= (data->dist + data->soft)) {
|
|
|
|
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else if (data->mode == LIMITDIST_INSIDE) {
|
|
|
|
/* if outside, then move to surface */
|
|
|
|
if (dist >= data->dist) {
|
|
|
|
clamp_surf= 1;
|
|
|
|
sfac= data->dist / dist;
|
|
|
|
}
|
|
|
|
/* if soft-distance is enabled, start fading once owner is dist-soft from the target */
|
|
|
|
else if (data->flag & LIMITDIST_USESOFT) {
|
|
|
|
// FIXME: there's a problem with "jumping" when this kicks in
|
|
|
|
if (dist >= (data->dist - data->soft)) {
|
2008-10-20 06:39:08 +00:00
|
|
|
sfac = (float)( data->soft*(1.0 - exp(-(dist - data->dist)/data->soft)) + data->dist );
|
2008-01-25 11:23:36 +00:00
|
|
|
sfac /= dist;
|
|
|
|
|
|
|
|
clamp_surf= 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
if (IS_EQ(dist, data->dist)==0) {
|
|
|
|
clamp_surf= 1;
|
|
|
|
sfac= data->dist / dist;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* clamp to 'surface' (i.e. move owner so that dist == data->dist) */
|
|
|
|
if (clamp_surf) {
|
|
|
|
/* simply interpolate along line formed by target -> owner */
|
2009-11-10 20:43:45 +00:00
|
|
|
interp_v3_v3v3(dvec, ct->matrix[3], cob->matrix[3], sfac);
|
2008-01-25 11:23:36 +00:00
|
|
|
|
|
|
|
/* copy new vector onto owner */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(cob->matrix[3], dvec);
|
2008-01-25 11:23:36 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_DISTLIMIT = {
|
|
|
|
CONSTRAINT_TYPE_DISTLIMIT, /* type */
|
|
|
|
sizeof(bDistLimitConstraint), /* size */
|
|
|
|
"Limit Distance", /* name */
|
|
|
|
"bDistLimitConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
distlimit_id_looper, /* id looper */
|
2008-01-25 11:23:36 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
distlimit_new_data, /* new data */
|
|
|
|
distlimit_get_tars, /* get constraint targets */
|
|
|
|
distlimit_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get a target matrix */
|
|
|
|
distlimit_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* ---------- Stretch To ------------ */
|
|
|
|
|
|
|
|
static void stretchto_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bStretchToConstraint *data= (bStretchToConstraint *)cdata;
|
|
|
|
|
|
|
|
data->volmode = 0;
|
|
|
|
data->plane = 0;
|
|
|
|
data->orglength = 0.0;
|
|
|
|
data->bulge = 1.0;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void stretchto_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bStretchToConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int stretchto_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bStretchToConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void stretchto_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bStretchToConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void stretchto_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bStretchToConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
/* only evaluate if there is a target */
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float size[3], scale[3], vec[3], xx[3], zz[3], orth[3];
|
|
|
|
float totmat[3][3];
|
|
|
|
float tmat[4][4];
|
|
|
|
float dist;
|
|
|
|
|
|
|
|
/* store scaling before destroying obmat */
|
2010-05-27 11:56:31 +00:00
|
|
|
mat4_to_size(size, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* store X orientation before destroying obmat */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(xx, cob->matrix[0]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* store Z orientation before destroying obmat */
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(zz, cob->matrix[2]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
sub_v3_v3v3(vec, cob->matrix[3], ct->matrix[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
vec[0] /= size[0];
|
|
|
|
vec[1] /= size[1];
|
|
|
|
vec[2] /= size[2];
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
dist = normalize_v3(vec);
|
|
|
|
//dist = len_v3v3( ob->obmat[3], targetmat[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* data->orglength==0 occurs on first run, and after 'R' button is clicked */
|
|
|
|
if (data->orglength == 0)
|
|
|
|
data->orglength = dist;
|
|
|
|
if (data->bulge == 0)
|
|
|
|
data->bulge = 1.0;
|
|
|
|
|
|
|
|
scale[1] = dist/data->orglength;
|
|
|
|
switch (data->volmode) {
|
|
|
|
/* volume preserving scaling */
|
|
|
|
case VOLUME_XZ :
|
|
|
|
scale[0] = 1.0f - (float)sqrt(data->bulge) + (float)sqrt(data->bulge*(data->orglength/dist));
|
|
|
|
scale[2] = scale[0];
|
|
|
|
break;
|
|
|
|
case VOLUME_X:
|
|
|
|
scale[0] = 1.0f + data->bulge * (data->orglength /dist - 1);
|
|
|
|
scale[2] = 1.0;
|
|
|
|
break;
|
|
|
|
case VOLUME_Z:
|
|
|
|
scale[0] = 1.0;
|
|
|
|
scale[2] = 1.0f + data->bulge * (data->orglength /dist - 1);
|
|
|
|
break;
|
|
|
|
/* don't care for volume */
|
|
|
|
case NO_VOLUME:
|
|
|
|
scale[0] = 1.0;
|
|
|
|
scale[2] = 1.0;
|
|
|
|
break;
|
|
|
|
default: /* should not happen, but in case*/
|
|
|
|
return;
|
|
|
|
} /* switch (data->volmode) */
|
|
|
|
|
|
|
|
/* Clear the object's rotation and scale */
|
|
|
|
cob->matrix[0][0]=size[0]*scale[0];
|
|
|
|
cob->matrix[0][1]=0;
|
|
|
|
cob->matrix[0][2]=0;
|
|
|
|
cob->matrix[1][0]=0;
|
|
|
|
cob->matrix[1][1]=size[1]*scale[1];
|
|
|
|
cob->matrix[1][2]=0;
|
|
|
|
cob->matrix[2][0]=0;
|
|
|
|
cob->matrix[2][1]=0;
|
|
|
|
cob->matrix[2][2]=size[2]*scale[2];
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
sub_v3_v3v3(vec, cob->matrix[3], ct->matrix[3]);
|
|
|
|
normalize_v3(vec);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* new Y aligns object target connection*/
|
2010-08-15 15:14:08 +00:00
|
|
|
negate_v3_v3(totmat[1], vec);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
switch (data->plane) {
|
|
|
|
case PLANE_X:
|
|
|
|
/* build new Z vector */
|
|
|
|
/* othogonal to "new Y" "old X! plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(orth, vec, xx);
|
|
|
|
normalize_v3(orth);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* new Z*/
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(totmat[2], orth);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* we decided to keep X plane*/
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(xx, orth, vec);
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[0], xx);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
|
|
|
case PLANE_Z:
|
|
|
|
/* build new X vector */
|
|
|
|
/* othogonal to "new Y" "old Z! plane */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(orth, vec, zz);
|
|
|
|
normalize_v3(orth);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* new X */
|
2010-08-15 15:14:08 +00:00
|
|
|
negate_v3_v3(totmat[0], orth);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* we decided to keep Z */
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(zz, orth, vec);
|
2010-08-15 15:14:08 +00:00
|
|
|
normalize_v3_v3(totmat[2], zz);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
|
|
|
} /* switch (data->plane) */
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(tmat, cob->matrix);
|
|
|
|
mul_m4_m3m4(cob->matrix, totmat, tmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_STRETCHTO = {
|
|
|
|
CONSTRAINT_TYPE_STRETCHTO, /* type */
|
|
|
|
sizeof(bStretchToConstraint), /* size */
|
|
|
|
"Stretch To", /* name */
|
|
|
|
"bStretchToConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
stretchto_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
stretchto_new_data, /* new data */
|
|
|
|
stretchto_get_tars, /* get constraint targets */
|
|
|
|
stretchto_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
stretchto_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* ---------- Floor ------------ */
|
|
|
|
|
|
|
|
static void minmax_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bMinMaxConstraint *data= (bMinMaxConstraint *)cdata;
|
|
|
|
|
|
|
|
data->minmaxflag = TRACK_Z;
|
|
|
|
data->offset = 0.0f;
|
|
|
|
data->cache[0] = data->cache[1] = data->cache[2] = 0.0f;
|
|
|
|
data->flag = 0;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void minmax_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bMinMaxConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int minmax_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bMinMaxConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void minmax_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bMinMaxConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void minmax_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bMinMaxConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
/* only evaluate if there is a target */
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float obmat[4][4], imat[4][4], tarmat[4][4], tmat[4][4];
|
|
|
|
float val1, val2;
|
|
|
|
int index;
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(obmat, cob->matrix);
|
|
|
|
copy_m4_m4(tarmat, ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
if (data->flag & MINMAX_USEROT) {
|
|
|
|
/* take rotation of target into account by doing the transaction in target's localspace */
|
2009-11-10 20:43:45 +00:00
|
|
|
invert_m4_m4(imat, tarmat);
|
|
|
|
mul_m4_m4m4(tmat, obmat, imat);
|
|
|
|
copy_m4_m4(obmat, tmat);
|
|
|
|
unit_m4(tarmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
switch (data->minmaxflag) {
|
|
|
|
case TRACK_Z:
|
|
|
|
val1 = tarmat[3][2];
|
|
|
|
val2 = obmat[3][2]-data->offset;
|
|
|
|
index = 2;
|
|
|
|
break;
|
|
|
|
case TRACK_Y:
|
|
|
|
val1 = tarmat[3][1];
|
|
|
|
val2 = obmat[3][1]-data->offset;
|
|
|
|
index = 1;
|
|
|
|
break;
|
|
|
|
case TRACK_X:
|
|
|
|
val1 = tarmat[3][0];
|
|
|
|
val2 = obmat[3][0]-data->offset;
|
|
|
|
index = 0;
|
|
|
|
break;
|
|
|
|
case TRACK_nZ:
|
|
|
|
val2 = tarmat[3][2];
|
|
|
|
val1 = obmat[3][2]-data->offset;
|
|
|
|
index = 2;
|
|
|
|
break;
|
|
|
|
case TRACK_nY:
|
|
|
|
val2 = tarmat[3][1];
|
|
|
|
val1 = obmat[3][1]-data->offset;
|
|
|
|
index = 1;
|
|
|
|
break;
|
|
|
|
case TRACK_nX:
|
|
|
|
val2 = tarmat[3][0];
|
|
|
|
val1 = obmat[3][0]-data->offset;
|
|
|
|
index = 0;
|
|
|
|
break;
|
|
|
|
default:
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (val1 > val2) {
|
|
|
|
obmat[3][index] = tarmat[3][index] + data->offset;
|
|
|
|
if (data->flag & MINMAX_STICKY) {
|
|
|
|
if (data->flag & MINMAX_STUCK) {
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(obmat[3], data->cache);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
else {
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(data->cache, obmat[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
data->flag |= MINMAX_STUCK;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
if (data->flag & MINMAX_USEROT) {
|
|
|
|
/* get out of localspace */
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_m4_m4m4(tmat, obmat, ct->matrix);
|
|
|
|
copy_m4_m4(cob->matrix, tmat);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
else {
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(cob->matrix[3], obmat[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
data->flag &= ~MINMAX_STUCK;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_MINMAX = {
|
|
|
|
CONSTRAINT_TYPE_MINMAX, /* type */
|
|
|
|
sizeof(bMinMaxConstraint), /* size */
|
|
|
|
"Floor", /* name */
|
|
|
|
"bMinMaxConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
minmax_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
minmax_new_data, /* new data */
|
|
|
|
minmax_get_tars, /* get constraint targets */
|
|
|
|
minmax_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
minmax_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* ------- RigidBody Joint ---------- */
|
|
|
|
|
|
|
|
static void rbj_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bRigidBodyJointConstraint *data= (bRigidBodyJointConstraint *)cdata;
|
|
|
|
|
|
|
|
// removed code which set target of this constraint
|
2010-03-22 09:30:00 +00:00
|
|
|
data->type=1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void rbj_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bRigidBodyJointConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int rbj_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bRigidBodyJointConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints without subtargets */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGETNS_GET_TARS(con, data->tar, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void rbj_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bRigidBodyJointConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGETNS_FLUSH_TARS(con, data->tar, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_RIGIDBODYJOINT = {
|
|
|
|
CONSTRAINT_TYPE_RIGIDBODYJOINT, /* type */
|
|
|
|
sizeof(bRigidBodyJointConstraint), /* size */
|
|
|
|
"RigidBody Joint", /* name */
|
|
|
|
"bRigidBodyJointConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
rbj_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
rbj_new_data, /* new data */
|
|
|
|
rbj_get_tars, /* get constraint targets */
|
|
|
|
rbj_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
NULL /* evaluate - this is not solved here... is just an interface for game-engine */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* -------- Clamp To ---------- */
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void clampto_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bClampToConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int clampto_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bClampToConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints without subtargets */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGETNS_GET_TARS(con, data->tar, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void clampto_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bClampToConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGETNS_FLUSH_TARS(con, data->tar, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void clampto_get_tarmat (bConstraint *UNUSED(con), bConstraintOb *cob, bConstraintTarget *ct, float UNUSED(ctime))
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
Curve *cu= ct->tar->data;
|
|
|
|
|
|
|
|
/* note: when creating constraints that follow path, the curve gets the CU_PATH set now,
|
|
|
|
* currently for paths to work it needs to go through the bevlist/displist system (ton)
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* only happens on reload file, but violates depsgraph still... fix! */
|
2009-09-12 14:12:37 +00:00
|
|
|
if (cu->path==NULL || cu->path->data==NULL)
|
2009-01-04 14:14:06 +00:00
|
|
|
makeDispListCurveTypes(cob->scene, ct->tar, 0);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* technically, this isn't really needed for evaluation, but we don't know what else
|
|
|
|
* might end up calling this...
|
|
|
|
*/
|
|
|
|
if (ct)
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void clampto_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bClampToConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
/* only evaluate if there is a target and it is a curve */
|
|
|
|
if (VALID_CONS_TARGET(ct) && (ct->tar->type == OB_CURVE)) {
|
|
|
|
Curve *cu= data->tar->data;
|
|
|
|
float obmat[4][4], targetMatrix[4][4], ownLoc[3];
|
|
|
|
float curveMin[3], curveMax[3];
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(obmat, cob->matrix);
|
|
|
|
unit_m4(targetMatrix);
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(ownLoc, obmat[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
INIT_MINMAX(curveMin, curveMax)
|
|
|
|
minmax_object(ct->tar, curveMin, curveMax);
|
|
|
|
|
|
|
|
/* get targetmatrix */
|
|
|
|
if (cu->path && cu->path->data) {
|
|
|
|
float vec[4], dir[3], totmat[4][4];
|
|
|
|
float curvetime;
|
|
|
|
short clamp_axis;
|
|
|
|
|
|
|
|
/* find best position on curve */
|
|
|
|
/* 1. determine which axis to sample on? */
|
|
|
|
if (data->flag == CLAMPTO_AUTO) {
|
|
|
|
float size[3];
|
2009-11-10 20:43:45 +00:00
|
|
|
sub_v3_v3v3(size, curveMax, curveMin);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* find axis along which the bounding box has the greatest
|
|
|
|
* extent. Otherwise, default to the x-axis, as that is quite
|
|
|
|
* frequently used.
|
|
|
|
*/
|
|
|
|
if ((size[2]>size[0]) && (size[2]>size[1]))
|
|
|
|
clamp_axis= CLAMPTO_Z - 1;
|
|
|
|
else if ((size[1]>size[0]) && (size[1]>size[2]))
|
|
|
|
clamp_axis= CLAMPTO_Y - 1;
|
|
|
|
else
|
|
|
|
clamp_axis = CLAMPTO_X - 1;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
clamp_axis= data->flag - 1;
|
|
|
|
|
|
|
|
/* 2. determine position relative to curve on a 0-1 scale based on bounding box */
|
|
|
|
if (data->flag2 & CLAMPTO_CYCLIC) {
|
|
|
|
/* cyclic, so offset within relative bounding box is used */
|
|
|
|
float len= (curveMax[clamp_axis] - curveMin[clamp_axis]);
|
|
|
|
float offset;
|
|
|
|
|
2008-12-19 11:45:46 +00:00
|
|
|
/* check to make sure len is not so close to zero that it'll cause errors */
|
|
|
|
if (IS_EQ(len, 0) == 0) {
|
|
|
|
/* find bounding-box range where target is located */
|
|
|
|
if (ownLoc[clamp_axis] < curveMin[clamp_axis]) {
|
|
|
|
/* bounding-box range is before */
|
|
|
|
offset= curveMin[clamp_axis];
|
|
|
|
|
|
|
|
while (ownLoc[clamp_axis] < offset)
|
|
|
|
offset -= len;
|
|
|
|
|
|
|
|
/* now, we calculate as per normal, except using offset instead of curveMin[clamp_axis] */
|
|
|
|
curvetime = (ownLoc[clamp_axis] - offset) / (len);
|
|
|
|
}
|
|
|
|
else if (ownLoc[clamp_axis] > curveMax[clamp_axis]) {
|
|
|
|
/* bounding-box range is after */
|
|
|
|
offset= curveMax[clamp_axis];
|
|
|
|
|
|
|
|
while (ownLoc[clamp_axis] > offset) {
|
|
|
|
if ((offset + len) > ownLoc[clamp_axis])
|
|
|
|
break;
|
|
|
|
else
|
|
|
|
offset += len;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* now, we calculate as per normal, except using offset instead of curveMax[clamp_axis] */
|
|
|
|
curvetime = (ownLoc[clamp_axis] - offset) / (len);
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* as the location falls within bounds, just calculate */
|
|
|
|
curvetime = (ownLoc[clamp_axis] - curveMin[clamp_axis]) / (len);
|
2003-10-21 13:22:07 +00:00
|
|
|
}
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
else {
|
2008-12-19 11:45:46 +00:00
|
|
|
/* as length is close to zero, curvetime by default should be 0 (i.e. the start) */
|
|
|
|
curvetime= 0.0f;
|
2003-10-21 13:22:07 +00:00
|
|
|
}
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
else {
|
|
|
|
/* no cyclic, so position is clamped to within the bounding box */
|
|
|
|
if (ownLoc[clamp_axis] <= curveMin[clamp_axis])
|
2008-12-19 11:45:46 +00:00
|
|
|
curvetime = 0.0f;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
else if (ownLoc[clamp_axis] >= curveMax[clamp_axis])
|
2008-12-19 11:45:46 +00:00
|
|
|
curvetime = 1.0f;
|
|
|
|
else if ( IS_EQ((curveMax[clamp_axis] - curveMin[clamp_axis]), 0) == 0 )
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
curvetime = (ownLoc[clamp_axis] - curveMin[clamp_axis]) / (curveMax[clamp_axis] - curveMin[clamp_axis]);
|
2008-12-19 11:45:46 +00:00
|
|
|
else
|
|
|
|
curvetime = 0.0f;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* 3. position on curve */
|
2010-04-21 11:59:47 +00:00
|
|
|
if (where_on_path(ct->tar, curvetime, vec, dir, NULL, NULL, NULL) ) {
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(totmat);
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(totmat[3], vec);
|
2007-01-14 23:42:06 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_serie_m4(targetMatrix, ct->tar->obmat, totmat, NULL, NULL, NULL, NULL, NULL, NULL);
|
2003-10-21 13:22:07 +00:00
|
|
|
}
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* obtain final object position */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(cob->matrix[3], targetMatrix[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
2006-08-13 14:18:15 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static bConstraintTypeInfo CTI_CLAMPTO = {
|
|
|
|
CONSTRAINT_TYPE_CLAMPTO, /* type */
|
|
|
|
sizeof(bClampToConstraint), /* size */
|
|
|
|
"Clamp To", /* name */
|
|
|
|
"bClampToConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
clampto_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
NULL, /* new data */
|
|
|
|
clampto_get_tars, /* get constraint targets */
|
|
|
|
clampto_flush_tars, /* flush constraint targets */
|
|
|
|
clampto_get_tarmat, /* get target matrix */
|
|
|
|
clampto_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
|
|
|
/* ---------- Transform Constraint ----------- */
|
|
|
|
|
|
|
|
static void transform_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bTransformConstraint *data= (bTransformConstraint *)cdata;
|
|
|
|
|
|
|
|
data->map[0]= 0;
|
|
|
|
data->map[1]= 1;
|
|
|
|
data->map[2]= 2;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void transform_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bTransformConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2008-03-08 02:16:37 +00:00
|
|
|
static int transform_get_tars (bConstraint *con, ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bTransformConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 1;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2008-03-08 02:16:37 +00:00
|
|
|
|
|
|
|
return 0;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void transform_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bTransformConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
2007-10-24 14:58:31 +00:00
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void transform_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bTransformConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
/* only evaluate if there is a target */
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float loc[3], eul[3], size[3];
|
|
|
|
float dvec[3], sval[3];
|
2009-06-22 18:19:18 +00:00
|
|
|
int i;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* obtain target effect */
|
|
|
|
switch (data->from) {
|
2008-02-10 10:01:57 +00:00
|
|
|
case 2: /* scale */
|
2009-11-10 20:43:45 +00:00
|
|
|
mat4_to_size( dvec,ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
2008-02-15 04:42:48 +00:00
|
|
|
case 1: /* rotation (convert to degrees first) */
|
2010-05-27 11:56:31 +00:00
|
|
|
mat4_to_eulO(dvec, cob->rotOrder, ct->matrix);
|
2008-02-15 04:42:48 +00:00
|
|
|
for (i=0; i<3; i++)
|
2008-10-20 06:39:08 +00:00
|
|
|
dvec[i] = (float)(dvec[i] / M_PI * 180);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
|
|
|
default: /* location */
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_v3_v3(dvec, ct->matrix[3]);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
2006-08-13 14:18:15 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* extract components of owner's matrix */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(loc, cob->matrix[3]);
|
2010-05-27 11:56:31 +00:00
|
|
|
mat4_to_eulO(eul, cob->rotOrder, cob->matrix);
|
|
|
|
mat4_to_size(size, cob->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* determine where in range current transforms lie */
|
|
|
|
if (data->expo) {
|
|
|
|
for (i=0; i<3; i++) {
|
|
|
|
if (data->from_max[i] - data->from_min[i])
|
|
|
|
sval[i]= (dvec[i] - data->from_min[i]) / (data->from_max[i] - data->from_min[i]);
|
|
|
|
else
|
|
|
|
sval[i]= 0.0f;
|
2006-08-13 14:18:15 +00:00
|
|
|
}
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
else {
|
|
|
|
/* clamp transforms out of range */
|
|
|
|
for (i=0; i<3; i++) {
|
|
|
|
CLAMP(dvec[i], data->from_min[i], data->from_max[i]);
|
|
|
|
if (data->from_max[i] - data->from_min[i])
|
|
|
|
sval[i]= (dvec[i] - data->from_min[i]) / (data->from_max[i] - data->from_min[i]);
|
|
|
|
else
|
|
|
|
sval[i]= 0.0f;
|
2006-08-13 14:18:15 +00:00
|
|
|
}
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
|
|
|
|
/* apply transforms */
|
|
|
|
switch (data->to) {
|
|
|
|
case 2: /* scaling */
|
|
|
|
for (i=0; i<3; i++)
|
2009-06-22 18:19:18 +00:00
|
|
|
size[i]= data->to_min[i] + (sval[(int)data->map[i]] * (data->to_max[i] - data->to_min[i]));
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
|
|
|
case 1: /* rotation */
|
|
|
|
for (i=0; i<3; i++) {
|
|
|
|
float tmin, tmax;
|
2007-04-07 03:32:57 +00:00
|
|
|
|
2008-04-16 07:33:54 +00:00
|
|
|
tmin= data->to_min[i];
|
|
|
|
tmax= data->to_max[i];
|
2007-09-26 07:33:31 +00:00
|
|
|
|
2008-04-16 07:33:54 +00:00
|
|
|
/* all values here should be in degrees */
|
2009-06-22 18:19:18 +00:00
|
|
|
eul[i]= tmin + (sval[(int)data->map[i]] * (tmax - tmin));
|
2008-04-16 07:33:54 +00:00
|
|
|
|
|
|
|
/* now convert final value back to radians */
|
2008-10-20 06:39:08 +00:00
|
|
|
eul[i] = (float)(eul[i] / 180 * M_PI);
|
2007-09-26 07:33:31 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
|
|
|
default: /* location */
|
|
|
|
/* get new location */
|
|
|
|
for (i=0; i<3; i++)
|
2009-06-22 18:19:18 +00:00
|
|
|
loc[i]= (data->to_min[i] + (sval[(int)data->map[i]] * (data->to_max[i] - data->to_min[i])));
|
2007-04-07 03:32:57 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* add original location back on (so that it can still be moved) */
|
2009-11-10 20:43:45 +00:00
|
|
|
add_v3_v3v3(loc, cob->matrix[3], loc);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* apply to matrix */
|
2009-11-10 20:43:45 +00:00
|
|
|
loc_eulO_size_to_mat4(cob->matrix, loc, eul, size, cob->rotOrder);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_TRANSFORM = {
|
|
|
|
CONSTRAINT_TYPE_TRANSFORM, /* type */
|
|
|
|
sizeof(bTransformConstraint), /* size */
|
|
|
|
"Transform", /* name */
|
|
|
|
"bTransformConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
transform_id_looper, /* id looper */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
transform_new_data, /* new data */
|
|
|
|
transform_get_tars, /* get constraint targets */
|
|
|
|
transform_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get a target matrix */
|
|
|
|
transform_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
/* ---------- Shrinkwrap Constraint ----------- */
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void shrinkwrap_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bShrinkwrapConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->target, userdata);
|
|
|
|
}
|
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
static int shrinkwrap_get_tars (bConstraint *con, ListBase *list)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bShrinkwrapConstraint *data = con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
SINGLETARGETNS_GET_TARS(con, data->target, ct, list)
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static void shrinkwrap_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bShrinkwrapConstraint *data = con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
SINGLETARGETNS_FLUSH_TARS(con, data->target, ct, list, nocopy)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void shrinkwrap_get_tarmat (bConstraint *con, bConstraintOb *cob, bConstraintTarget *ct, float UNUSED(ctime))
|
2009-04-20 15:06:46 +00:00
|
|
|
{
|
|
|
|
bShrinkwrapConstraint *scon = (bShrinkwrapConstraint *) con->data;
|
|
|
|
|
|
|
|
if( VALID_CONS_TARGET(ct) && (ct->tar->type == OB_MESH) )
|
|
|
|
{
|
|
|
|
int fail = FALSE;
|
|
|
|
float co[3] = {0.0f, 0.0f, 0.0f};
|
|
|
|
float no[3] = {0.0f, 0.0f, 0.0f};
|
|
|
|
float dist;
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
SpaceTransform transform;
|
|
|
|
DerivedMesh *target = object_get_derived_final(cob->scene, ct->tar, CD_MASK_BAREMESH);
|
|
|
|
BVHTreeRayHit hit;
|
|
|
|
BVHTreeNearest nearest;
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
BVHTreeFromMesh treeData;
|
2010-05-27 11:56:31 +00:00
|
|
|
memset(&treeData, 0, sizeof(treeData));
|
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
nearest.index = -1;
|
|
|
|
nearest.dist = FLT_MAX;
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
hit.index = -1;
|
|
|
|
hit.dist = 100000.0f; //TODO should use FLT_MAX.. but normal projection doenst yet supports it
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
if(target != NULL)
|
|
|
|
{
|
|
|
|
space_transform_from_matrixs(&transform, cob->matrix, ct->tar->obmat);
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
switch(scon->shrinkType)
|
|
|
|
{
|
|
|
|
case MOD_SHRINKWRAP_NEAREST_SURFACE:
|
|
|
|
case MOD_SHRINKWRAP_NEAREST_VERTEX:
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
if(scon->shrinkType == MOD_SHRINKWRAP_NEAREST_VERTEX)
|
|
|
|
bvhtree_from_mesh_verts(&treeData, target, 0.0, 2, 6);
|
|
|
|
else
|
|
|
|
bvhtree_from_mesh_faces(&treeData, target, 0.0, 2, 6);
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
if(treeData.tree == NULL)
|
|
|
|
{
|
|
|
|
fail = TRUE;
|
|
|
|
break;
|
|
|
|
}
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
space_transform_apply(&transform, co);
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
BLI_bvhtree_find_nearest(treeData.tree, co, &nearest, treeData.nearest_callback, &treeData);
|
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
dist = len_v3v3(co, nearest.co);
|
|
|
|
interp_v3_v3v3(co, co, nearest.co, (dist - scon->dist)/dist); /* linear interpolation */
|
2009-04-20 15:06:46 +00:00
|
|
|
space_transform_invert(&transform, co);
|
|
|
|
break;
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
case MOD_SHRINKWRAP_PROJECT:
|
|
|
|
if(scon->projAxis & MOD_SHRINKWRAP_PROJECT_OVER_X_AXIS) no[0] = 1.0f;
|
|
|
|
if(scon->projAxis & MOD_SHRINKWRAP_PROJECT_OVER_Y_AXIS) no[1] = 1.0f;
|
|
|
|
if(scon->projAxis & MOD_SHRINKWRAP_PROJECT_OVER_Z_AXIS) no[2] = 1.0f;
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
if(INPR(no,no) < FLT_EPSILON)
|
|
|
|
{
|
|
|
|
fail = TRUE;
|
|
|
|
break;
|
|
|
|
}
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
normalize_v3(no);
|
2010-05-27 11:56:31 +00:00
|
|
|
|
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
bvhtree_from_mesh_faces(&treeData, target, scon->dist, 4, 6);
|
|
|
|
if(treeData.tree == NULL)
|
|
|
|
{
|
|
|
|
fail = TRUE;
|
|
|
|
break;
|
|
|
|
}
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
if(normal_projection_project_vertex(0, co, no, &transform, treeData.tree, &hit, treeData.raycast_callback, &treeData) == FALSE)
|
|
|
|
{
|
|
|
|
fail = TRUE;
|
|
|
|
break;
|
|
|
|
}
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(co, hit.co);
|
2009-04-20 15:06:46 +00:00
|
|
|
break;
|
|
|
|
}
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
free_bvhtree_from_mesh(&treeData);
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
target->release(target);
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
if(fail == TRUE)
|
|
|
|
{
|
|
|
|
/* Don't move the point */
|
|
|
|
co[0] = co[1] = co[2] = 0.0f;
|
|
|
|
}
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2009-04-20 15:06:46 +00:00
|
|
|
/* co is in local object coordinates, change it to global and update target position */
|
2010-04-12 00:36:50 +00:00
|
|
|
mul_m4_v3(cob->matrix, co);
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(ct->matrix[3], co);
|
2009-04-20 15:06:46 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void shrinkwrap_evaluate (bConstraint *UNUSED(con), bConstraintOb *cob, ListBase *targets)
|
2009-04-20 15:06:46 +00:00
|
|
|
{
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
/* only evaluate if there is a target */
|
|
|
|
if (VALID_CONS_TARGET(ct))
|
|
|
|
{
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(cob->matrix[3], ct->matrix[3]);
|
2009-04-20 15:06:46 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_SHRINKWRAP = {
|
|
|
|
CONSTRAINT_TYPE_SHRINKWRAP, /* type */
|
|
|
|
sizeof(bShrinkwrapConstraint), /* size */
|
|
|
|
"Shrinkwrap", /* name */
|
|
|
|
"bShrinkwrapConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
shrinkwrap_id_looper, /* id looper */
|
2009-04-20 15:06:46 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
NULL, /* new data */
|
|
|
|
shrinkwrap_get_tars, /* get constraint targets */
|
|
|
|
shrinkwrap_flush_tars, /* flush constraint targets */
|
|
|
|
shrinkwrap_get_tarmat, /* get a target matrix */
|
|
|
|
shrinkwrap_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
2009-10-30 06:33:40 +00:00
|
|
|
/* --------- Damped Track ---------- */
|
|
|
|
|
|
|
|
static void damptrack_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bDampTrackConstraint *data= (bDampTrackConstraint *)cdata;
|
|
|
|
|
|
|
|
data->trackflag = TRACK_Y;
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void damptrack_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bDampTrackConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2009-10-30 06:33:40 +00:00
|
|
|
static int damptrack_get_tars (bConstraint *con, ListBase *list)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bDampTrackConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
2009-04-20 15:06:46 +00:00
|
|
|
|
2009-10-30 06:33:40 +00:00
|
|
|
static void damptrack_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bDampTrackConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* array of direction vectors for the tracking flags */
|
|
|
|
static const float track_dir_vecs[6][3] = {
|
|
|
|
{+1,0,0}, {0,+1,0}, {0,0,+1}, /* TRACK_X, TRACK_Y, TRACK_Z */
|
|
|
|
{-1,0,0}, {0,-1,0}, {0,0,-1} /* TRACK_NX, TRACK_NY, TRACK_NZ */
|
|
|
|
};
|
|
|
|
|
|
|
|
static void damptrack_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bDampTrackConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
float obvec[3], tarvec[3], obloc[3];
|
|
|
|
float raxis[3], rangle;
|
|
|
|
float rmat[3][3], tmat[4][4];
|
|
|
|
|
|
|
|
/* find the (unit) direction that the axis we're interested in currently points
|
2009-11-10 20:43:45 +00:00
|
|
|
* - mul_mat3_m4_v3() only takes the 3x3 (rotation+scaling) components of the 4x4 matrix
|
2009-10-30 06:33:40 +00:00
|
|
|
* - the normalisation step at the end should take care of any unwanted scaling
|
|
|
|
* left over in the 3x3 matrix we used
|
|
|
|
*/
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(obvec, track_dir_vecs[data->trackflag]);
|
2009-11-10 20:43:45 +00:00
|
|
|
mul_mat3_m4_v3(cob->matrix, obvec);
|
2009-10-30 06:33:40 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
if (normalize_v3(obvec) == 0.0f) {
|
2009-10-30 06:33:40 +00:00
|
|
|
/* exceptional case - just use the track vector as appropriate */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(obvec, track_dir_vecs[data->trackflag]);
|
2009-10-30 06:33:40 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* find the (unit) direction vector going from the owner to the target */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(obloc, cob->matrix[3]);
|
2009-11-10 20:43:45 +00:00
|
|
|
sub_v3_v3v3(tarvec, ct->matrix[3], obloc);
|
2009-10-30 06:33:40 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
if (normalize_v3(tarvec) == 0.0f) {
|
2009-10-30 06:33:40 +00:00
|
|
|
/* the target is sitting on the owner, so just make them use the same direction vectors */
|
|
|
|
// FIXME: or would it be better to use the pure direction vector?
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(tarvec, obvec);
|
|
|
|
//copy_v3_v3(tarvec, track_dir_vecs[data->trackflag]);
|
2009-10-30 06:33:40 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* determine the axis-angle rotation, which represents the smallest possible rotation
|
|
|
|
* between the two rotation vectors (i.e. the 'damping' referred to in the name)
|
|
|
|
* - we take this to be the rotation around the normal axis/vector to the plane defined
|
|
|
|
* by the current and destination vectors, which will 'map' the current axis to the
|
|
|
|
* destination vector
|
|
|
|
* - the min/max wrappers around (obvec . tarvec) result (stored temporarily in rangle)
|
|
|
|
* are used to ensure that the smallest angle is chosen
|
|
|
|
*/
|
2009-11-10 20:43:45 +00:00
|
|
|
cross_v3_v3v3(raxis, obvec, tarvec);
|
2009-10-30 06:33:40 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
rangle= dot_v3v3(obvec, tarvec);
|
2009-10-30 06:33:40 +00:00
|
|
|
rangle= acos( MAX2(-1.0f, MIN2(1.0f, rangle)) );
|
|
|
|
|
|
|
|
/* construct rotation matrix from the axis-angle rotation found above
|
|
|
|
* - this call takes care to make sure that the axis provided is a unit vector first
|
|
|
|
*/
|
2010-05-27 11:56:31 +00:00
|
|
|
axis_angle_to_mat3(rmat, raxis, rangle);
|
2009-10-30 06:33:40 +00:00
|
|
|
|
|
|
|
/* rotate the owner in the way defined by this rotation matrix, then reapply the location since
|
|
|
|
* we may have destroyed that in the process of multiplying the matrix
|
|
|
|
*/
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(tmat);
|
|
|
|
mul_m4_m3m4(tmat, rmat, cob->matrix); // m1, m3, m2
|
2009-10-30 06:33:40 +00:00
|
|
|
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(cob->matrix, tmat);
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(cob->matrix[3], obloc);
|
2009-10-30 06:33:40 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_DAMPTRACK = {
|
|
|
|
CONSTRAINT_TYPE_DAMPTRACK, /* type */
|
|
|
|
sizeof(bDampTrackConstraint), /* size */
|
|
|
|
"Damped Track", /* name */
|
|
|
|
"bDampTrackConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
damptrack_id_looper, /* id looper */
|
2009-10-30 06:33:40 +00:00
|
|
|
NULL, /* copy data */
|
|
|
|
damptrack_new_data, /* new data */
|
|
|
|
damptrack_get_tars, /* get constraint targets */
|
|
|
|
damptrack_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
damptrack_evaluate /* evaluate */
|
|
|
|
};
|
2009-04-20 15:06:46 +00:00
|
|
|
|
2009-11-01 11:29:40 +00:00
|
|
|
/* ----------- Spline IK ------------ */
|
|
|
|
|
|
|
|
static void splineik_free (bConstraint *con)
|
|
|
|
{
|
|
|
|
bSplineIKConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* binding array */
|
|
|
|
if (data->points)
|
|
|
|
MEM_freeN(data->points);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void splineik_copy (bConstraint *con, bConstraint *srccon)
|
|
|
|
{
|
|
|
|
bSplineIKConstraint *src= srccon->data;
|
|
|
|
bSplineIKConstraint *dst= con->data;
|
|
|
|
|
|
|
|
/* copy the binding array */
|
|
|
|
dst->points= MEM_dupallocN(src->points);
|
2010-04-25 13:27:52 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static void splineik_new_data (void *cdata)
|
|
|
|
{
|
|
|
|
bSplineIKConstraint *data= (bSplineIKConstraint *)cdata;
|
|
|
|
|
|
|
|
data->chainlen= 1;
|
2009-11-01 11:29:40 +00:00
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
static void splineik_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bSplineIKConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
2009-11-01 11:29:40 +00:00
|
|
|
static int splineik_get_tars (bConstraint *con, ListBase *list)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bSplineIKConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints without subtargets */
|
|
|
|
SINGLETARGETNS_GET_TARS(con, data->tar, ct, list)
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void splineik_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bSplineIKConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
|
|
|
SINGLETARGETNS_FLUSH_TARS(con, data->tar, ct, list, nocopy)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-10-16 14:32:17 +00:00
|
|
|
static void splineik_get_tarmat (bConstraint *UNUSED(con), bConstraintOb *cob, bConstraintTarget *ct, float UNUSED(ctime))
|
2009-11-01 11:29:40 +00:00
|
|
|
{
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
Curve *cu= ct->tar->data;
|
|
|
|
|
|
|
|
/* note: when creating constraints that follow path, the curve gets the CU_PATH set now,
|
|
|
|
* currently for paths to work it needs to go through the bevlist/displist system (ton)
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* only happens on reload file, but violates depsgraph still... fix! */
|
|
|
|
if (cu->path==NULL || cu->path->data==NULL)
|
|
|
|
makeDispListCurveTypes(cob->scene, ct->tar, 0);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* technically, this isn't really needed for evaluation, but we don't know what else
|
|
|
|
* might end up calling this...
|
|
|
|
*/
|
|
|
|
if (ct)
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
2009-11-01 11:29:40 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_SPLINEIK = {
|
|
|
|
CONSTRAINT_TYPE_SPLINEIK, /* type */
|
|
|
|
sizeof(bSplineIKConstraint), /* size */
|
|
|
|
"Spline IK", /* name */
|
|
|
|
"bSplineIKConstraint", /* struct name */
|
|
|
|
splineik_free, /* free data */
|
|
|
|
NULL, /* relink data */
|
2010-02-06 11:28:22 +00:00
|
|
|
splineik_id_looper, /* id looper */
|
2009-11-01 11:29:40 +00:00
|
|
|
splineik_copy, /* copy data */
|
2010-04-25 13:27:52 +00:00
|
|
|
splineik_new_data, /* new data */
|
2009-11-01 11:29:40 +00:00
|
|
|
splineik_get_tars, /* get constraint targets */
|
|
|
|
splineik_flush_tars, /* flush constraint targets */
|
|
|
|
splineik_get_tarmat, /* get target matrix */
|
|
|
|
NULL /* evaluate - solved as separate loop */
|
|
|
|
};
|
|
|
|
|
== Pivot Constraint ==
This constraint allows an object or bone to have their rotations applied as if their origin/pivot-point was located elsewhere. The most obvious uses include foot-roll, see-saws, but could also include more complicated rolling-box examples.
== Usage Examples ==
=== Foot Roll ===
1. Add 'Pivot' Constraint to the bone without any target.
2. Set the 'Y' value of the offset to the length of the bone. Usually this should be negative (if you rig with feet facing 'forwards' along -Y axis). This gives you a pivot point relative to the bone's (preconstraint) location, which should be at the tip of the bone here. Disabling the 'Use Relative Offset' would make this offset be relative to 0,0,0 instead of to the owner/bone-head.
3. Ensure that the 'Pivot When' setting is set to '-X Rot', (default) which means that the pivot will only used when the rotation on the X-Axis is negative to get tip-toe 'roll'.
=== See Saw ===
1. Add a 'Pivot' constraint too see-saw plank object, this time with a target that you wish to have as the pivot-point. It's possible to do this without too (as before), but is less intuitive.
2. Optionally, if you want the plank slightly raised, set the z-offset value, which should make the pivot-point used to be relative to the target with the z-offset applied.
3. Ensure that 'Pivot When' is set to 'Always', which means that the pivot will always be used, irrespective of the rotation.
== Notes ==
* The 'Pivot When' setting has been integrated in the constraint, since this is something that will often be required for these setups. Having to set up additional drivers to drive the constraint to do this kindof beats the purpose of providing this.
* The 'Offset' functionality is probably not presented as clearly as it could be. We may need to go over this again.
* For foot-roll - if any scaling of the foot is required, simply set up a driver on the y-offset to make this dynamically respond to the "scale" RNA property of the bones (don't use the "Transform Channel" vartype since that won't work correct here). However, this shouldn't be common enough to warrant special treatment.
2010-05-27 10:50:06 +00:00
|
|
|
/* ----------- Pivot ------------- */
|
|
|
|
|
|
|
|
static void pivotcon_id_looper (bConstraint *con, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bPivotConstraint *data= con->data;
|
|
|
|
|
|
|
|
/* target only */
|
|
|
|
func(con, (ID**)&data->tar, userdata);
|
|
|
|
}
|
|
|
|
|
|
|
|
static int pivotcon_get_tars (bConstraint *con, ListBase *list)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bPivotConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* standard target-getting macro for single-target constraints */
|
|
|
|
SINGLETARGET_GET_TARS(con, data->tar, data->subtarget, ct, list)
|
|
|
|
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
static void pivotcon_flush_tars (bConstraint *con, ListBase *list, short nocopy)
|
|
|
|
{
|
|
|
|
if (con && list) {
|
|
|
|
bPivotConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= list->first;
|
|
|
|
|
|
|
|
/* the following macro is used for all standard single-target constraints */
|
|
|
|
SINGLETARGET_FLUSH_TARS(con, data->tar, data->subtarget, ct, list, nocopy)
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static void pivotcon_evaluate (bConstraint *con, bConstraintOb *cob, ListBase *targets)
|
|
|
|
{
|
|
|
|
bPivotConstraint *data= con->data;
|
|
|
|
bConstraintTarget *ct= targets->first;
|
|
|
|
|
|
|
|
float pivot[3], vec[3];
|
|
|
|
float rotMat[3][3];
|
|
|
|
|
|
|
|
/* firstly, check if pivoting should take place based on the current rotation */
|
|
|
|
if (data->rotAxis != PIVOTCON_AXIS_NONE) {
|
|
|
|
float rot[3];
|
|
|
|
|
|
|
|
/* extract euler-rotation of target */
|
|
|
|
mat4_to_eulO(rot, cob->rotOrder, cob->matrix);
|
|
|
|
|
|
|
|
/* check which range might be violated */
|
|
|
|
if (data->rotAxis < PIVOTCON_AXIS_X) {
|
|
|
|
/* negative rotations (data->rotAxis = 0 -> 2) */
|
|
|
|
if (rot[data->rotAxis] > 0.0f)
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* positive rotations (data->rotAxis = 3 -> 5 */
|
|
|
|
if (rot[data->rotAxis - PIVOTCON_AXIS_X] < 0.0f)
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* find the pivot-point to use */
|
|
|
|
if (VALID_CONS_TARGET(ct)) {
|
|
|
|
/* apply offset to target location */
|
|
|
|
add_v3_v3v3(pivot, ct->matrix[3], data->offset);
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* no targets to worry about... */
|
|
|
|
if ((data->flag & PIVOTCON_FLAG_OFFSET_ABS) == 0) {
|
|
|
|
/* offset is relative to owner */
|
|
|
|
add_v3_v3v3(pivot, cob->matrix[3], data->offset);
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* directly use the 'offset' specified as an absolute position instead */
|
2010-08-15 15:14:08 +00:00
|
|
|
copy_v3_v3(pivot, data->offset);
|
== Pivot Constraint ==
This constraint allows an object or bone to have their rotations applied as if their origin/pivot-point was located elsewhere. The most obvious uses include foot-roll, see-saws, but could also include more complicated rolling-box examples.
== Usage Examples ==
=== Foot Roll ===
1. Add 'Pivot' Constraint to the bone without any target.
2. Set the 'Y' value of the offset to the length of the bone. Usually this should be negative (if you rig with feet facing 'forwards' along -Y axis). This gives you a pivot point relative to the bone's (preconstraint) location, which should be at the tip of the bone here. Disabling the 'Use Relative Offset' would make this offset be relative to 0,0,0 instead of to the owner/bone-head.
3. Ensure that the 'Pivot When' setting is set to '-X Rot', (default) which means that the pivot will only used when the rotation on the X-Axis is negative to get tip-toe 'roll'.
=== See Saw ===
1. Add a 'Pivot' constraint too see-saw plank object, this time with a target that you wish to have as the pivot-point. It's possible to do this without too (as before), but is less intuitive.
2. Optionally, if you want the plank slightly raised, set the z-offset value, which should make the pivot-point used to be relative to the target with the z-offset applied.
3. Ensure that 'Pivot When' is set to 'Always', which means that the pivot will always be used, irrespective of the rotation.
== Notes ==
* The 'Pivot When' setting has been integrated in the constraint, since this is something that will often be required for these setups. Having to set up additional drivers to drive the constraint to do this kindof beats the purpose of providing this.
* The 'Offset' functionality is probably not presented as clearly as it could be. We may need to go over this again.
* For foot-roll - if any scaling of the foot is required, simply set up a driver on the y-offset to make this dynamically respond to the "scale" RNA property of the bones (don't use the "Transform Channel" vartype since that won't work correct here). However, this shouldn't be common enough to warrant special treatment.
2010-05-27 10:50:06 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* get rotation matrix representing the rotation of the owner */
|
|
|
|
// TODO: perhaps we might want to include scaling based on the pivot too?
|
|
|
|
copy_m3_m4(rotMat, cob->matrix);
|
|
|
|
normalize_m3(rotMat);
|
|
|
|
|
|
|
|
/* perform the pivoting... */
|
|
|
|
/* 1. take the vector from owner to the pivot */
|
|
|
|
sub_v3_v3v3(vec, pivot, cob->matrix[3]);
|
|
|
|
/* 2. rotate this vector by the rotation of the object... */
|
|
|
|
mul_m3_v3(rotMat, vec);
|
|
|
|
/* 3. make the rotation in terms of the pivot now */
|
|
|
|
add_v3_v3v3(cob->matrix[3], pivot, vec);
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static bConstraintTypeInfo CTI_PIVOT = {
|
|
|
|
CONSTRAINT_TYPE_PIVOT, /* type */
|
|
|
|
sizeof(bPivotConstraint), /* size */
|
|
|
|
"Pivot", /* name */
|
|
|
|
"bPivotConstraint", /* struct name */
|
|
|
|
NULL, /* free data */
|
|
|
|
NULL, /* relink data */
|
|
|
|
pivotcon_id_looper, /* id looper */
|
|
|
|
NULL, /* copy data */
|
|
|
|
NULL, /* new data */ // XXX: might be needed to get 'normal' pivot behaviour...
|
|
|
|
pivotcon_get_tars, /* get constraint targets */
|
|
|
|
pivotcon_flush_tars, /* flush constraint targets */
|
|
|
|
default_get_tarmat, /* get target matrix */
|
|
|
|
pivotcon_evaluate /* evaluate */
|
|
|
|
};
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* ************************* Constraints Type-Info *************************** */
|
|
|
|
/* All of the constraints api functions use bConstraintTypeInfo structs to carry out
|
2009-03-28 11:44:07 +00:00
|
|
|
* and operations that involve constraint specific code.
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
*/
|
|
|
|
|
|
|
|
/* These globals only ever get directly accessed in this file */
|
2009-11-01 11:29:40 +00:00
|
|
|
static bConstraintTypeInfo *constraintsTypeInfo[NUM_CONSTRAINT_TYPES];
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
static short CTI_INIT= 1; /* when non-zero, the list needs to be updated */
|
|
|
|
|
|
|
|
/* This function only gets called when CTI_INIT is non-zero */
|
|
|
|
static void constraints_init_typeinfo () {
|
|
|
|
constraintsTypeInfo[0]= NULL; /* 'Null' Constraint */
|
|
|
|
constraintsTypeInfo[1]= &CTI_CHILDOF; /* ChildOf Constraint */
|
|
|
|
constraintsTypeInfo[2]= &CTI_TRACKTO; /* TrackTo Constraint */
|
|
|
|
constraintsTypeInfo[3]= &CTI_KINEMATIC; /* IK Constraint */
|
|
|
|
constraintsTypeInfo[4]= &CTI_FOLLOWPATH; /* Follow-Path Constraint */
|
|
|
|
constraintsTypeInfo[5]= &CTI_ROTLIMIT; /* Limit Rotation Constraint */
|
|
|
|
constraintsTypeInfo[6]= &CTI_LOCLIMIT; /* Limit Location Constraint */
|
|
|
|
constraintsTypeInfo[7]= &CTI_SIZELIMIT; /* Limit Scaling Constraint */
|
|
|
|
constraintsTypeInfo[8]= &CTI_ROTLIKE; /* Copy Rotation Constraint */
|
|
|
|
constraintsTypeInfo[9]= &CTI_LOCLIKE; /* Copy Location Constraint */
|
|
|
|
constraintsTypeInfo[10]= &CTI_SIZELIKE; /* Copy Scaling Constraint */
|
|
|
|
constraintsTypeInfo[11]= &CTI_PYTHON; /* Python/Script Constraint */
|
|
|
|
constraintsTypeInfo[12]= &CTI_ACTION; /* Action Constraint */
|
|
|
|
constraintsTypeInfo[13]= &CTI_LOCKTRACK; /* Locked-Track Constraint */
|
2008-01-25 11:23:36 +00:00
|
|
|
constraintsTypeInfo[14]= &CTI_DISTLIMIT; /* Limit Distance Constraint */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
constraintsTypeInfo[15]= &CTI_STRETCHTO; /* StretchTo Constaint */
|
|
|
|
constraintsTypeInfo[16]= &CTI_MINMAX; /* Floor Constraint */
|
|
|
|
constraintsTypeInfo[17]= &CTI_RIGIDBODYJOINT; /* RigidBody Constraint */
|
|
|
|
constraintsTypeInfo[18]= &CTI_CLAMPTO; /* ClampTo Constraint */
|
2008-01-25 11:23:36 +00:00
|
|
|
constraintsTypeInfo[19]= &CTI_TRANSFORM; /* Transformation Constraint */
|
2009-04-20 15:06:46 +00:00
|
|
|
constraintsTypeInfo[20]= &CTI_SHRINKWRAP; /* Shrinkwrap Constraint */
|
2009-10-30 06:33:40 +00:00
|
|
|
constraintsTypeInfo[21]= &CTI_DAMPTRACK; /* Damped TrackTo Constraint */
|
2009-11-01 11:29:40 +00:00
|
|
|
constraintsTypeInfo[22]= &CTI_SPLINEIK; /* Spline IK Constraint */
|
2010-01-02 04:14:17 +00:00
|
|
|
constraintsTypeInfo[23]= &CTI_TRANSLIKE; /* Copy Transforms Constraint */
|
2010-03-16 12:55:56 +00:00
|
|
|
constraintsTypeInfo[24]= &CTI_SAMEVOL; /* Maintain Volume Constraint */
|
== Pivot Constraint ==
This constraint allows an object or bone to have their rotations applied as if their origin/pivot-point was located elsewhere. The most obvious uses include foot-roll, see-saws, but could also include more complicated rolling-box examples.
== Usage Examples ==
=== Foot Roll ===
1. Add 'Pivot' Constraint to the bone without any target.
2. Set the 'Y' value of the offset to the length of the bone. Usually this should be negative (if you rig with feet facing 'forwards' along -Y axis). This gives you a pivot point relative to the bone's (preconstraint) location, which should be at the tip of the bone here. Disabling the 'Use Relative Offset' would make this offset be relative to 0,0,0 instead of to the owner/bone-head.
3. Ensure that the 'Pivot When' setting is set to '-X Rot', (default) which means that the pivot will only used when the rotation on the X-Axis is negative to get tip-toe 'roll'.
=== See Saw ===
1. Add a 'Pivot' constraint too see-saw plank object, this time with a target that you wish to have as the pivot-point. It's possible to do this without too (as before), but is less intuitive.
2. Optionally, if you want the plank slightly raised, set the z-offset value, which should make the pivot-point used to be relative to the target with the z-offset applied.
3. Ensure that 'Pivot When' is set to 'Always', which means that the pivot will always be used, irrespective of the rotation.
== Notes ==
* The 'Pivot When' setting has been integrated in the constraint, since this is something that will often be required for these setups. Having to set up additional drivers to drive the constraint to do this kindof beats the purpose of providing this.
* The 'Offset' functionality is probably not presented as clearly as it could be. We may need to go over this again.
* For foot-roll - if any scaling of the foot is required, simply set up a driver on the y-offset to make this dynamically respond to the "scale" RNA property of the bones (don't use the "Transform Channel" vartype since that won't work correct here). However, this shouldn't be common enough to warrant special treatment.
2010-05-27 10:50:06 +00:00
|
|
|
constraintsTypeInfo[25]= &CTI_PIVOT; /* Pivot Constraint */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* This function should be used for getting the appropriate type-info when only
|
|
|
|
* a constraint type is known
|
|
|
|
*/
|
|
|
|
bConstraintTypeInfo *get_constraint_typeinfo (int type)
|
|
|
|
{
|
|
|
|
/* initialise the type-info list? */
|
|
|
|
if (CTI_INIT) {
|
|
|
|
constraints_init_typeinfo();
|
|
|
|
CTI_INIT = 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* only return for valid types */
|
|
|
|
if ( (type >= CONSTRAINT_TYPE_NULL) &&
|
2009-11-01 11:29:40 +00:00
|
|
|
(type < NUM_CONSTRAINT_TYPES ) )
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
/* there shouldn't be any segfaults here... */
|
|
|
|
return constraintsTypeInfo[type];
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
printf("No valid constraint type-info data available. Type = %i \n", type);
|
|
|
|
}
|
|
|
|
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* This function should always be used to get the appropriate type-info, as it
|
|
|
|
* has checks which prevent segfaults in some weird cases.
|
|
|
|
*/
|
|
|
|
bConstraintTypeInfo *constraint_get_typeinfo (bConstraint *con)
|
|
|
|
{
|
|
|
|
/* only return typeinfo for valid constraints */
|
|
|
|
if (con)
|
|
|
|
return get_constraint_typeinfo(con->type);
|
|
|
|
else
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* ************************* General Constraints API ************************** */
|
|
|
|
/* The functions here are called by various parts of Blender. Very few (should be none if possible)
|
|
|
|
* constraint-specific code should occur here.
|
|
|
|
*/
|
|
|
|
|
|
|
|
/* ---------- Data Management ------- */
|
|
|
|
|
|
|
|
/* Free data of a specific constraint if it has any info */
|
|
|
|
void free_constraint_data (bConstraint *con)
|
|
|
|
{
|
|
|
|
if (con->data) {
|
|
|
|
bConstraintTypeInfo *cti= constraint_get_typeinfo(con);
|
|
|
|
|
|
|
|
/* perform any special freeing constraint may have */
|
|
|
|
if (cti && cti->free_data)
|
|
|
|
cti->free_data(con);
|
|
|
|
|
|
|
|
/* free constraint data now */
|
|
|
|
MEM_freeN(con->data);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Free all constraints from a constraint-stack */
|
2009-03-31 22:36:13 +00:00
|
|
|
void free_constraints (ListBase *list)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
bConstraint *con;
|
|
|
|
|
|
|
|
/* Free constraint data and also any extra data */
|
2009-03-31 22:36:13 +00:00
|
|
|
for (con= list->first; con; con= con->next)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
free_constraint_data(con);
|
|
|
|
|
|
|
|
/* Free the whole list */
|
2009-03-31 22:36:13 +00:00
|
|
|
BLI_freelistN(list);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
2009-11-16 12:33:42 +00:00
|
|
|
|
|
|
|
/* Remove the specified constraint from the given constraint stack */
|
|
|
|
int remove_constraint (ListBase *list, bConstraint *con)
|
|
|
|
{
|
|
|
|
if (con) {
|
|
|
|
free_constraint_data(con);
|
|
|
|
BLI_freelinkN(list, con);
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
else
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2010-02-21 11:05:52 +00:00
|
|
|
/* Remove all the constraints of the specified type from the given constraint stack */
|
|
|
|
void remove_constraints_type (ListBase *list, short type, short last_only)
|
|
|
|
{
|
|
|
|
bConstraint *con, *conp;
|
|
|
|
|
|
|
|
if (list == NULL)
|
|
|
|
return;
|
|
|
|
|
|
|
|
/* remove from the end of the list to make it faster to find the last instance */
|
|
|
|
for (con= list->last; con; con= conp) {
|
|
|
|
conp= con->prev;
|
|
|
|
|
|
|
|
if (con->type == type) {
|
|
|
|
remove_constraint(list, con);
|
|
|
|
if (last_only)
|
|
|
|
return;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-11-16 12:33:42 +00:00
|
|
|
/* ......... */
|
|
|
|
|
|
|
|
/* Creates a new constraint, initialises its data, and returns it */
|
|
|
|
static bConstraint *add_new_constraint_internal (const char *name, short type)
|
|
|
|
{
|
|
|
|
bConstraint *con= MEM_callocN(sizeof(bConstraint), "Constraint");
|
|
|
|
bConstraintTypeInfo *cti= get_constraint_typeinfo(type);
|
|
|
|
const char *newName;
|
2009-11-11 19:58:30 +00:00
|
|
|
|
|
|
|
/* Set up a generic constraint datablock */
|
|
|
|
con->type = type;
|
|
|
|
con->flag |= CONSTRAINT_EXPAND;
|
|
|
|
con->enforce = 1.0f;
|
|
|
|
|
2009-11-16 12:33:42 +00:00
|
|
|
/* Determine a basic name, and info */
|
2009-11-11 19:58:30 +00:00
|
|
|
if (cti) {
|
2009-11-16 12:33:42 +00:00
|
|
|
/* initialise constraint data */
|
2009-11-11 19:58:30 +00:00
|
|
|
con->data = MEM_callocN(cti->size, cti->structName);
|
2009-11-16 12:33:42 +00:00
|
|
|
|
2009-11-11 19:58:30 +00:00
|
|
|
/* only constraints that change any settings need this */
|
|
|
|
if (cti->new_data)
|
|
|
|
cti->new_data(con->data);
|
2009-11-16 12:33:42 +00:00
|
|
|
|
|
|
|
/* if no name is provided, use the type of the constraint as the name */
|
|
|
|
newName= (name && name[0]) ? name : cti->name;
|
2009-11-11 19:58:30 +00:00
|
|
|
}
|
2009-11-16 12:33:42 +00:00
|
|
|
else {
|
|
|
|
/* if no name is provided, use the generic "Const" name */
|
|
|
|
// NOTE: any constraint type that gets here really shouldn't get added...
|
|
|
|
newName= (name && name[0]) ? name : "Const";
|
|
|
|
}
|
|
|
|
|
|
|
|
/* copy the name */
|
|
|
|
BLI_strncpy(con->name, newName, sizeof(con->name));
|
|
|
|
|
|
|
|
/* return the new constraint */
|
2009-11-11 19:58:30 +00:00
|
|
|
return con;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* if pchan is not NULL then assume we're adding a pose constraint */
|
2009-11-16 12:33:42 +00:00
|
|
|
static bConstraint *add_new_constraint (Object *ob, bPoseChannel *pchan, const char *name, short type)
|
2009-11-11 19:58:30 +00:00
|
|
|
{
|
|
|
|
bConstraint *con;
|
|
|
|
ListBase *list;
|
2009-11-16 12:33:42 +00:00
|
|
|
|
|
|
|
/* add the constraint */
|
2009-11-11 19:58:30 +00:00
|
|
|
con= add_new_constraint_internal(name, type);
|
2009-11-16 12:33:42 +00:00
|
|
|
|
|
|
|
/* find the constraint stack - bone or object? */
|
|
|
|
list = (pchan) ? (&pchan->constraints) : (&ob->constraints);
|
|
|
|
|
2009-11-11 19:58:30 +00:00
|
|
|
if (list) {
|
|
|
|
/* add new constraint to end of list of constraints before ensuring that it has a unique name
|
|
|
|
* (otherwise unique-naming code will fail, since it assumes element exists in list)
|
|
|
|
*/
|
|
|
|
BLI_addtail(list, con);
|
|
|
|
unique_constraint_name(con, list);
|
2009-11-16 12:33:42 +00:00
|
|
|
|
2009-11-11 19:58:30 +00:00
|
|
|
/* if the target list is a list on some PoseChannel belonging to a proxy-protected
|
|
|
|
* Armature layer, we must tag newly added constraints with a flag which allows them
|
|
|
|
* to persist after proxy syncing has been done
|
|
|
|
*/
|
|
|
|
if (proxylocked_constraints_owner(ob, pchan))
|
|
|
|
con->flag |= CONSTRAINT_PROXY_LOCAL;
|
2009-11-16 12:33:42 +00:00
|
|
|
|
|
|
|
/* make this constraint the active one */
|
|
|
|
constraints_set_active(list, con);
|
2009-11-11 19:58:30 +00:00
|
|
|
}
|
2009-11-16 12:33:42 +00:00
|
|
|
|
2009-11-11 19:58:30 +00:00
|
|
|
return con;
|
|
|
|
}
|
|
|
|
|
2009-11-16 12:33:42 +00:00
|
|
|
/* ......... */
|
|
|
|
|
|
|
|
/* Add new constraint for the given bone */
|
|
|
|
bConstraint *add_pose_constraint (Object *ob, bPoseChannel *pchan, const char *name, short type)
|
2009-11-11 19:58:30 +00:00
|
|
|
{
|
2009-11-16 12:33:42 +00:00
|
|
|
if (pchan == NULL)
|
2009-11-11 19:58:30 +00:00
|
|
|
return NULL;
|
2009-11-16 12:33:42 +00:00
|
|
|
|
2009-11-11 19:58:30 +00:00
|
|
|
return add_new_constraint(ob, pchan, name, type);
|
|
|
|
}
|
|
|
|
|
2009-11-16 12:33:42 +00:00
|
|
|
/* Add new constraint for the given object */
|
2009-11-11 19:58:30 +00:00
|
|
|
bConstraint *add_ob_constraint(Object *ob, const char *name, short type)
|
|
|
|
{
|
|
|
|
return add_new_constraint(ob, NULL, name, type);
|
|
|
|
}
|
|
|
|
|
2009-11-16 12:33:42 +00:00
|
|
|
/* ......... */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* Reassign links that constraints have to other data (called during file loading?) */
|
|
|
|
void relink_constraints (ListBase *conlist)
|
|
|
|
{
|
|
|
|
bConstraint *con;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
for (con= conlist->first; con; con= con->next) {
|
|
|
|
bConstraintTypeInfo *cti= constraint_get_typeinfo(con);
|
|
|
|
|
|
|
|
if (cti) {
|
|
|
|
/* relink any targets */
|
|
|
|
if (cti->get_constraint_targets) {
|
|
|
|
ListBase targets = {NULL, NULL};
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
cti->get_constraint_targets(con, &targets);
|
|
|
|
for (ct= targets.first; ct; ct= ct->next) {
|
|
|
|
ID_NEW(ct->tar);
|
|
|
|
}
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (cti->flush_constraint_targets)
|
|
|
|
cti->flush_constraint_targets(con, &targets, 0);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
2007-07-21 07:26:15 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* relink any other special data */
|
|
|
|
if (cti->relink_data)
|
|
|
|
cti->relink_data(con);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2010-02-06 11:28:22 +00:00
|
|
|
/* Run the given callback on all ID-blocks in list of constraints */
|
|
|
|
void id_loop_constraints (ListBase *conlist, ConstraintIDFunc func, void *userdata)
|
|
|
|
{
|
|
|
|
bConstraint *con;
|
|
|
|
|
|
|
|
for (con= conlist->first; con; con= con->next) {
|
|
|
|
bConstraintTypeInfo *cti= constraint_get_typeinfo(con);
|
|
|
|
|
|
|
|
if (cti) {
|
|
|
|
if (cti->id_looper)
|
|
|
|
cti->id_looper(con, func, userdata);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-11-16 12:33:42 +00:00
|
|
|
/* ......... */
|
|
|
|
|
2010-02-09 00:02:22 +00:00
|
|
|
/* helper for copy_constraints(), to be used for making sure that ID's are valid */
|
2010-10-16 14:32:17 +00:00
|
|
|
static void con_extern_cb(bConstraint *UNUSED(con), ID **idpoin, void *UNUSED(userData))
|
2010-02-08 17:30:35 +00:00
|
|
|
{
|
2010-02-10 14:47:39 +00:00
|
|
|
if (*idpoin && (*idpoin)->lib)
|
2010-02-08 17:30:35 +00:00
|
|
|
id_lib_extern(*idpoin);
|
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* duplicate all of the constraints in a constraint stack */
|
2010-04-29 15:31:53 +00:00
|
|
|
void copy_constraints (ListBase *dst, const ListBase *src, int do_extern)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
bConstraint *con, *srccon;
|
|
|
|
|
|
|
|
dst->first= dst->last= NULL;
|
2008-01-01 18:29:19 +00:00
|
|
|
BLI_duplicatelist(dst, src);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
2009-01-12 22:54:30 +00:00
|
|
|
for (con=dst->first, srccon=src->first; con && srccon; srccon=srccon->next, con=con->next) {
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
bConstraintTypeInfo *cti= constraint_get_typeinfo(con);
|
|
|
|
|
|
|
|
/* make a new copy of the constraint's data */
|
|
|
|
con->data = MEM_dupallocN(con->data);
|
|
|
|
|
|
|
|
/* only do specific constraints if required */
|
2010-02-08 17:30:35 +00:00
|
|
|
if (cti) {
|
2010-02-09 00:02:22 +00:00
|
|
|
/* perform custom copying operations if needed */
|
|
|
|
if (cti->copy_data)
|
2010-02-08 17:30:35 +00:00
|
|
|
cti->copy_data(con, srccon);
|
2010-05-27 11:56:31 +00:00
|
|
|
|
2010-04-29 15:31:53 +00:00
|
|
|
/* for proxies we dont want to make extern */
|
2010-05-27 11:56:31 +00:00
|
|
|
if (do_extern) {
|
2010-04-29 15:31:53 +00:00
|
|
|
/* go over used ID-links for this constraint to ensure that they are valid for proxies */
|
|
|
|
if (cti->id_looper)
|
|
|
|
cti->id_looper(con, con_extern_cb, NULL);
|
|
|
|
}
|
2010-02-08 17:30:35 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2009-11-16 12:33:42 +00:00
|
|
|
/* ......... */
|
|
|
|
|
2010-04-26 03:42:38 +00:00
|
|
|
bConstraint *constraints_findByName(ListBase *list, const char *name)
|
|
|
|
{
|
|
|
|
return BLI_findstring(list, name, offsetof(bConstraint, name));
|
|
|
|
}
|
|
|
|
|
2009-03-31 22:36:13 +00:00
|
|
|
/* finds the 'active' constraint in a constraint stack */
|
|
|
|
bConstraint *constraints_get_active (ListBase *list)
|
|
|
|
{
|
|
|
|
bConstraint *con;
|
|
|
|
|
|
|
|
/* search for the first constraint with the 'active' flag set */
|
|
|
|
if (list) {
|
|
|
|
for (con= list->first; con; con= con->next) {
|
|
|
|
if (con->flag & CONSTRAINT_ACTIVE)
|
|
|
|
return con;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* no active constraint found */
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2009-11-16 12:33:42 +00:00
|
|
|
/* Set the given constraint as the active one (clearing all the others) */
|
|
|
|
void constraints_set_active (ListBase *list, bConstraint *con)
|
|
|
|
{
|
|
|
|
bConstraint *c;
|
|
|
|
|
2010-02-05 06:05:24 +00:00
|
|
|
if (list) {
|
|
|
|
for (c= list->first; c; c= c->next) {
|
|
|
|
if (c == con)
|
|
|
|
c->flag |= CONSTRAINT_ACTIVE;
|
|
|
|
else
|
|
|
|
c->flag &= ~CONSTRAINT_ACTIVE;
|
|
|
|
}
|
2009-11-16 12:33:42 +00:00
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2008-01-04 11:21:50 +00:00
|
|
|
/* -------- Constraints and Proxies ------- */
|
|
|
|
|
|
|
|
/* Rescue all constraints tagged as being CONSTRAINT_PROXY_LOCAL (i.e. added to bone that's proxy-synced in this file) */
|
|
|
|
void extract_proxylocal_constraints (ListBase *dst, ListBase *src)
|
|
|
|
{
|
|
|
|
bConstraint *con, *next;
|
|
|
|
|
|
|
|
/* for each tagged constraint, remove from src and move to dst */
|
|
|
|
for (con= src->first; con; con= next) {
|
|
|
|
next= con->next;
|
|
|
|
|
|
|
|
/* check if tagged */
|
|
|
|
if (con->flag & CONSTRAINT_PROXY_LOCAL) {
|
|
|
|
BLI_remlink(src, con);
|
|
|
|
BLI_addtail(dst, con);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Returns if the owner of the constraint is proxy-protected */
|
|
|
|
short proxylocked_constraints_owner (Object *ob, bPoseChannel *pchan)
|
|
|
|
{
|
|
|
|
/* Currently, constraints can only be on object or bone level */
|
|
|
|
if (ob && ob->proxy) {
|
|
|
|
if (ob->pose && pchan) {
|
|
|
|
bArmature *arm= ob->data;
|
|
|
|
|
|
|
|
/* On bone-level, check if bone is on proxy-protected layer */
|
|
|
|
if ((pchan->bone) && (pchan->bone->layer & arm->layer_protected))
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* FIXME: constraints on object-level are not handled well yet */
|
|
|
|
return 1;
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* -------- Target-Matrix Stuff ------- */
|
|
|
|
|
|
|
|
/* This function is a relic from the prior implementations of the constraints system, when all
|
|
|
|
* constraints either had one or no targets. It used to be called during the main constraint solving
|
|
|
|
* loop, but is now only used for the remaining cases for a few constraints.
|
|
|
|
*
|
|
|
|
* None of the actual calculations of the matricies should be done here! Also, this function is
|
|
|
|
* not to be used by any new constraints, particularly any that have multiple targets.
|
|
|
|
*/
|
2009-10-12 09:39:57 +00:00
|
|
|
void get_constraint_target_matrix (struct Scene *scene, bConstraint *con, int n, short ownertype, void *ownerdata, float mat[][4], float ctime)
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
{
|
|
|
|
bConstraintTypeInfo *cti= constraint_get_typeinfo(con);
|
|
|
|
ListBase targets = {NULL, NULL};
|
|
|
|
bConstraintOb *cob;
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
if (cti && cti->get_constraint_targets) {
|
|
|
|
/* make 'constraint-ob' */
|
|
|
|
cob= MEM_callocN(sizeof(bConstraintOb), "tempConstraintOb");
|
|
|
|
cob->type= ownertype;
|
2009-10-12 09:39:57 +00:00
|
|
|
cob->scene = scene;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
switch (ownertype) {
|
|
|
|
case CONSTRAINT_OBTYPE_OBJECT: /* it is usually this case */
|
|
|
|
{
|
|
|
|
cob->ob= (Object *)ownerdata;
|
|
|
|
cob->pchan= NULL;
|
|
|
|
if (cob->ob) {
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(cob->matrix, cob->ob->obmat);
|
|
|
|
copy_m4_m4(cob->startmat, cob->matrix);
|
2007-07-21 07:26:15 +00:00
|
|
|
}
|
|
|
|
else {
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(cob->matrix);
|
|
|
|
unit_m4(cob->startmat);
|
2007-07-21 07:26:15 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
break;
|
|
|
|
case CONSTRAINT_OBTYPE_BONE: /* this may occur in some cases */
|
|
|
|
{
|
|
|
|
cob->ob= NULL; /* this might not work at all :/ */
|
|
|
|
cob->pchan= (bPoseChannel *)ownerdata;
|
|
|
|
if (cob->pchan) {
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(cob->matrix, cob->pchan->pose_mat);
|
|
|
|
copy_m4_m4(cob->startmat, cob->matrix);
|
2007-07-21 07:26:15 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
else {
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(cob->matrix);
|
|
|
|
unit_m4(cob->startmat);
|
2007-07-21 07:26:15 +00:00
|
|
|
}
|
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
break;
|
2007-07-21 07:26:15 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* get targets - we only need the first one though (and there should only be one) */
|
|
|
|
cti->get_constraint_targets(con, &targets);
|
|
|
|
|
|
|
|
/* only calculate the target matrix on the first target */
|
|
|
|
ct= (bConstraintTarget *)targets.first;
|
2007-10-24 14:58:31 +00:00
|
|
|
while(ct && n-- > 0)
|
|
|
|
ct= ct->next;
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
if (ct) {
|
|
|
|
if (cti->get_target_matrix)
|
|
|
|
cti->get_target_matrix(con, cob, ct, ctime);
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(mat, ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
|
|
|
|
|
|
|
/* free targets + 'constraint-ob' */
|
|
|
|
if (cti->flush_constraint_targets)
|
|
|
|
cti->flush_constraint_targets(con, &targets, 1);
|
|
|
|
MEM_freeN(cob);
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
/* invalid constraint - perhaps... */
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(mat);
|
2002-10-12 11:37:38 +00:00
|
|
|
}
|
2005-09-12 04:46:34 +00:00
|
|
|
}
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
|
|
|
|
/* ---------- Evaluation ----------- */
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* This function is called whenever constraints need to be evaluated. Currently, all
|
|
|
|
* constraints that can be evaluated are everytime this gets run.
|
|
|
|
*
|
|
|
|
* constraints_make_evalob and constraints_clear_evalob should be called before and
|
|
|
|
* after running this function, to sort out cob
|
|
|
|
*/
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
void solve_constraints (ListBase *conlist, bConstraintOb *cob, float ctime)
|
|
|
|
{
|
|
|
|
bConstraint *con;
|
2010-09-28 10:03:56 +00:00
|
|
|
float oldmat[4][4];
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
float enf;
|
|
|
|
|
|
|
|
/* check that there is a valid constraint object to evaluate */
|
|
|
|
if (cob == NULL)
|
|
|
|
return;
|
|
|
|
|
|
|
|
/* loop over available constraints, solving and blending them */
|
|
|
|
for (con= conlist->first; con; con= con->next) {
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
bConstraintTypeInfo *cti= constraint_get_typeinfo(con);
|
|
|
|
ListBase targets = {NULL, NULL};
|
|
|
|
|
|
|
|
/* these we can skip completely (invalid constraints...) */
|
|
|
|
if (cti == NULL) continue;
|
2009-09-24 21:22:24 +00:00
|
|
|
if (con->flag & (CONSTRAINT_DISABLE|CONSTRAINT_OFF)) continue;
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* these constraints can't be evaluated anyway */
|
|
|
|
if (cti->evaluate_constraint == NULL) continue;
|
2007-07-23 11:28:43 +00:00
|
|
|
/* influence == 0 should be ignored */
|
|
|
|
if (con->enforce == 0.0f) continue;
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* influence of constraint
|
2.5: Blender "Animato" - New Animation System
Finally, here is the basic (functional) prototype of the new animation system which will allow for the infamous "everything is animatable", and which also addresses several of the more serious shortcomings of the old system. Unfortunately, this will break old animation files (especially right now, as I haven't written the version patching code yet), however, this is for the future.
Highlights of the new system:
* Scrapped IPO-Curves/IPO/(Action+Constraint-Channels)/Action system, and replaced it with F-Curve/Action.
- F-Curves (animators from other packages will feel at home with this name) replace IPO-Curves.
- The 'new' Actions, act as the containers for F-Curves, so that they can be reused. They are therefore more akin to the old 'IPO' blocks, except they do not have the blocktype restriction, so you can store materials/texture/geometry F-Curves in the same Action as Object transforms, etc.
* F-Curves use RNA-paths for Data Access, hence allowing "every" (where sensible/editable that is) user-accessible setting from RNA to be animated.
* Drivers are no longer mixed with Animation Data, so rigs will not be that easily broken and several dependency problems can be eliminated. (NOTE: drivers haven't been hooked up yet, but the code is in place)
* F-Curve modifier system allows useful 'large-scale' manipulation of F-Curve values, including (I've only included implemented ones here): envelope deform (similar to lattices to allow broad-scale reshaping of curves), curve generator (polynomial or py-expression), cycles (replacing the old cyclic extrapolation modes, giving more control over this). (NOTE: currently this cannot be tested, as there's not access to them, but the code is all in place)
* NLA system with 'tracks' (i.e. layers), and multiple strips per track. (NOTE: NLA system is not yet functional, as it's only partially coded still)
There are more nice things that I will be preparing some nice docs for soon, but for now, check for more details:
http://lists.blender.org/pipermail/bf-taskforce25/2009-January/000260.html
So, what currently works:
* I've implemented two basic operators for the 3D-view only to Insert and Delete Keyframes. These are tempolary ones only that will be replaced in due course with 'proper' code.
* Object Loc/Rot/Scale can be keyframed. Also, the colour of the 'active' material (Note: this should really be for nth material instead, but that doesn't work yet in RNA) can also be keyframed into the same datablock.
* Standard animation refresh (i.e. animation resulting from NLA and Action evaluation) is now done completely separate from drivers before anything else is done after a frame change. Drivers are handled after this in a separate pass, as dictated by depsgraph flags, etc.
Notes:
* Drivers haven't been hooked up yet
* Only objects and data directly linked to objects can be animated.
* Depsgraph will need further tweaks. Currently, I've only made sure that it will update some things in the most basic cases (i.e. frame change).
* Animation Editors are currently broken (in terms of editing stuff). This will be my next target (priority to get Dopesheet working first, then F-Curve editor - i.e. old IPO Editor)
* I've had to put in large chunks of XXX sandboxing for old animation system code all around the place. This will be cleaned up in due course, as some places need special review.
In particular, the particles and sequencer code have far too many manual calls to calculate + flush animation info, which is really bad (this is a 'please explain yourselves' call to Physics coders!).
2009-01-17 03:12:50 +00:00
|
|
|
* - value should have been set from animation data already
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
*/
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
enf = con->enforce;
|
|
|
|
|
2007-08-18 06:17:50 +00:00
|
|
|
/* move owner matrix into right space */
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
constraint_mat_convertspace(cob->ob, cob->pchan, cob->matrix, CONSTRAINT_SPACE_WORLD, con->ownspace);
|
2009-11-10 20:43:45 +00:00
|
|
|
copy_m4_m4(oldmat, cob->matrix);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* prepare targets for constraint solving */
|
|
|
|
if (cti->get_constraint_targets) {
|
|
|
|
bConstraintTarget *ct;
|
|
|
|
|
|
|
|
/* get targets
|
|
|
|
* - constraints should use ct->matrix, not directly accessing values
|
|
|
|
* - ct->matrix members have not yet been calculated here!
|
|
|
|
*/
|
|
|
|
cti->get_constraint_targets(con, &targets);
|
2007-08-18 06:17:50 +00:00
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* set matrices
|
|
|
|
* - calculate if possible, otherwise just initialise as identity matrix
|
|
|
|
*/
|
|
|
|
if (cti->get_target_matrix) {
|
|
|
|
for (ct= targets.first; ct; ct= ct->next)
|
|
|
|
cti->get_target_matrix(con, cob, ct, ctime);
|
|
|
|
}
|
|
|
|
else {
|
|
|
|
for (ct= targets.first; ct; ct= ct->next)
|
2009-11-10 20:43:45 +00:00
|
|
|
unit_m4(ct->matrix);
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
}
|
2007-08-18 06:17:50 +00:00
|
|
|
}
|
|
|
|
|
2010-09-24 17:47:28 +00:00
|
|
|
/* Solve the constraint and put result in cob->matrix */
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
cti->evaluate_constraint(con, cob, &targets);
|
|
|
|
|
|
|
|
/* clear targets after use
|
|
|
|
* - this should free temp targets but no data should be copied back
|
|
|
|
* as constraints may have done some nasty things to it...
|
|
|
|
*/
|
|
|
|
if (cti->flush_constraint_targets) {
|
|
|
|
cti->flush_constraint_targets(con, &targets, 1);
|
|
|
|
}
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
|
|
|
|
/* Interpolate the enforcement, to blend result of constraint into final owner transform */
|
2010-09-24 17:47:28 +00:00
|
|
|
/* Note: all kind of stuff here before (caused trouble), much easier to just interpolate, or did I miss something? -jahka */
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
if (enf < 1.0) {
|
2010-09-24 17:47:28 +00:00
|
|
|
float solution[4][4];
|
|
|
|
copy_m4_m4(solution, cob->matrix);
|
|
|
|
blend_m4_m4m4(cob->matrix, oldmat, solution, enf);
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
}
|
|
|
|
|
== Constraints System - Recode 2 ==
Once again, I've recoded the constraints system. This time, the goals were:
* To make it more future-proof by 'modernising' the coding style. The long functions filled with switch statements, have given way to function-pointers with smaller functions for specific purposes.
* To make it support constraints which use multiple targets more readily that it did. In the past, it was assumed that constraints could only have at most one target.
As a result, a lot of code has been shuffled around, and modified. Also, the subversion number has been bumped up.
Known issues:
* PyConstraints, which were the main motivation for supporting multiple-targets, are currently broken. There are some bimport() error that keeps causing problems. I've also temporarily removed the doDriver support, although it may return in another form soon.
* Constraints BPy-API is currently has a few features which currently don't work yet
* Outliner currently only displays the names of the constraints instead of the fancy subtarget/target/constraint-name display it used to do. What gets displayed here needs further investigation, as the old way was certainly not that great (and is not compatible with the new system too)
2007-10-21 23:00:29 +00:00
|
|
|
/* move owner back into worldspace for next constraint/other business */
|
== Constraints System ==
After just over a week of coding, I've finished doing a major refactor/cleanup of the constraints code. In the process, quite a few old kludges and ugly hacks have been removed. Also, some new features which will greatly benefit riggers have been implemented.
=== What's New ===
* The long-awaited ``ChildOf Constraint'':
This allows you to animate parent influences, and choose which transformation channels the parent affects the child on (i.e. no translation/rotation/scaling). It should be noted that disabling some combinations may not totally work as expected. Also, the 'Set Inverse' and 'Clear Inverse' buttons at the bottom of this constraint's panel set/clear the inverse correction for the parent's effects. Use these to make the owner not stick/be glued to the parent.
* Constraint/Target Evaluation Spaces:
In some constraints, there are now 1-2 combo boxes at the bottom of their panel, which allows you to pick which `co-ordinate space' they are evaluated in. This is much more flexible than the old 'local' options for bones only were.
* Action Constraint - Loc/Rot/Size Inputs
The Action Constraint can finally use the target's location/rotation/scaling transforms as input, to control the owner of the constraint. This should work much more reliably than it used to. The target evaluation should now also be more accurate due to the new space conversion stuff.
* Transform - No longer in Crazy Space (TM)
Transforming objects/bones with constraints applied should no longer occur in Crazy Space. They are now correctly inverse-corrected. This also applies to old-style object tracking.
=== General Code Changes ===
* solve_constraints is now in constraints.c. I've removed the old `blend consecutive constraints of same type' junk, which made the code more complex than it needed to be.
* evaluate_constraint is now only passed the constraint, and two matrices. A few unused variables have been removed from here.
* A tempolary struct, bConstraintOb, is now passed to solve_constraints instead of relying on an ugly, static workobject in some cases. This works much better.
* Made the formatting of constraint code consistent
* There's a version patch for older files so that constraint settings are correctly converted to the new system. This is currently done for MajorVersion <= 244, and SubVersion < 3. I've bumped up the subversion to 3 for this purpose. However, with the imminent 2.45 release, this may need to be adjusted accordingly.
* LocEulSizeToMat4 and LocQuatSizeToMat4 now work in the order Size, Rot, Location. I've also added a few other math functions.
* Mat4BlendMat4 is now in arithb. I've modified it's method slightly, to use other arithb functions, instead of its crazy blending scheme.
* Moved some of the RigidBodyJoint constraint's code out of blenkernel, and into src. It shouldn't be setting its target in its data initialisation function based + accessing scene stuff where it was doing so.
=== Future Work ===
* Geometry to act as targets for constraints. A space has been reserved for this already.
* Tidy up UI buttons of constraints
2007-07-15 03:35:37 +00:00
|
|
|
if ((con->flag & CONSTRAINT_SPACEONCE) == 0)
|
|
|
|
constraint_mat_convertspace(cob->ob, cob->pchan, cob->matrix, con->ownspace, CONSTRAINT_SPACE_WORLD);
|
|
|
|
}
|
|
|
|
}
|