----------------------------------
Here's my changelog:
-fixed Rand() so that it doesn't seed everytime and should generate better random numbers
- changed a few error return types to something more appropriate
- clean up of uninitialized variables & removal of unneccessary objects
- NMesh returns wrapped vectors now
- World returns wrapped matrices now
- Object.getEuler() and Object.getBoundingBox() return Wrapped data when data is present
- Object.getMatrix() returns wrapped data if it's worldspace, 'localspace' returns a new matrix
- Vector, Euler, Mat, Quat, call all now internally wrap object without destroying internal datablocks
- Removed memory allocation (unneeded) from all methods
- Vector's resize methods are only applicable to new vectors not wrapped data.
- Matrix(), Quat(), Euler(), Vector() now accepts ANY sequence list, including tuples, list, or a self object to copy - matrices accept multiple sequences
- Fixed Slerp() so that it now works correctly values are clamped between 0 and 1
- Euler.rotate does internal rotation now
- Slice assignment now works better for all types
- Vector * Vector and Quat * Quat are defined and return the DOT product
- Mat * Vec and Vec * Mat are defined now
- Moved #includes to .c file from headers. Also fixed prototypes in mathutils
- Added new helper functions for incref'ing to genutils
- Major cleanup of header files includes - include Mathutils.h for access to math types
- matrix.toQuat() and .toEuler() now fixed take appropriate matrix sizes
- Matrix() with no parameters now returns an identity matrix by default not a zero matrix
- printf() now prints with 6 digits instead of 4
- printf() now prints output with object descriptor
- Matrices now support [x][y] assignment (e.g. matrix[x][y] = 5.4)
- Matrix[index] = value now expectes a sequence not an integer. This will now set a ROW of the matrix through a sequence. index cannot go above the row size of the matrix.
- slice operations on matrices work with sequences now (rows of the matrix) example: mymatrix[0:2] returns a list of 2 wrapped vectors with access to the matrix data.
- slice assignment will no longer modify the data if the assignment operation fails
- fixed error in matrix * scalar multiplication
- euler.toMatrix(), toQuat() no longer causes "creep" from repeated use
- Wrapped data will generate wrapped objects when toEuler(), toQuat(), toMatrix() is used
- Quats can be created with angle/axis, axis/angle
- 4x4 matrices can be multiplied by 3D vectors (by popular demand :))
- vec *quat / quat * vec is now defined
- vec.magnitude alias for vec.length
- all self, internal methods return a pointer to self now so you can do print vector.internalmethod() or vector.internalmethod().nextmethod() (no more print matrix.inverse() returning 'none')
- these methods have been deprecated (still functioning but suggested to use the corrected functionality):
* CopyVec() - replaced by Vector() functionality
* CopyMat() - replaced by Matrix() functionality
* CopyQuat() - replace by Quaternion() functionality
* CopyEuler() - replaced by Euler() functionality
* RotateEuler() - replaced by Euler.rotate() funtionality
* MatMultVec() - replaced by matrix * vector
* VecMultMat() - replaced by vector * matrix
- New struct containers references to python object data or internally allocated blender data for wrapping
* Explaination here: math structs now function as a 'simple wrapper' or a 'py_object' - data that is created on the fly will now be a 'py_object' with its memory managed by python
* otherwise if the data is returned by blender's G.main then the math object is a 'simple wrapper' and data can be accessed directly from the struct just like other python objects.
nicer, but be aware that these locations are 'virtual', derived from
the actual Mesh Cage vertices you don't see.
Normal transform (and actual transform with Manipulators) remains on
the original vertices, which also shows with dashed line towards pivot
on scale/rotate.
1:
* when a blendfile gets loaded, paths are corrected with OS specific slashes (see blender.c)
* made available BLI_char_switch(char *string, char from, char to)
* made available BLI_clean(char *string);. This function should be called whenever you're doing path stuff, so paths are correctly saved, and thus avoiding other path functions stopping to work
2:
* relative paths work now in sequencer too (due to slash mess that didn't work all too well).
- fixed error in option "Enable Goal" which didn't work with Vertex groups
assigned
- renamed some buttons & fixed tooltips. A doc online will be there soon
Added check for audiostream_pos() - which should return current frame
corrected for audio playback - to be at least 1 frame larger than current
frame. It caused in some situations even a backwards playback!
feature with a header button.
Works like audio playback on ALT+A, meaning you have to add the audio
strips in Sequencer. The 'only sync' option for ALT+A doesn't work here.
Also: fixed crash on ESC in audio playback in sequencer.
When a new space (window) type gets added, older Blender binaries will
crash on reading files saved with such spaces. This breaks the so nice
upward compility on files... With addition of SpaceTime I got an occasion
to verify & fix that. If Blender detects a non-existant space it will
degrade it to a 3D window now. Tested on a full 2.34 compile.
It fails in all criticial tests, not to mention the confusing UI even. :)
Code is still there, just the options for zblur are removed now, and
Output Panel restored.
Belongs actually as an effect in the Sequencer, nice todo for a next
release.
This time, probably for good since I've really tested it through and the z factor used there never gives overflow and is correctly calibrated with both the real zfac and the zoom level.
It tries to use the shaded colors, but these are from the subsurfed mesh.
Didn't feel like hacking here, so for this case colors are initialized
as black.
This combo didnt render OK:
1 Armature has Empties parented to Bones
2 Curve has these Empties as Hooks
3 Mesh is deformed with this Curve
Could be solved with calling makeDispList() earlier on, and it seems like
the depgraph already does a great job here. :)
The ALT+F "Beauty Fill" could crash, for example on the already legendary
monkey head CTRL+T,J test. :)
Found out there was a necessary check for edge/face selection flags missing
My fix to make subloops more idling was not needed in the case for
drawing the knife cut 'mouse trail'. This uses a proper qread() which
has a wait builtin for events.
- fixing bug reported by Paolo Colombo: space handler slinks set for a 3d view were not set when the area got maximized;
- Blender.Object: added object.isSB() method to know if an object is a soft body (has ob->soft != NULL). Used in fixfromarmature.py.
Scripts:
- updates: batch_name_edit (Campbell), fixfromarmature (JMS);
- additions:
X3D exporter by Bart;
Envelope Suite by Jonas Petersen;
BVH 2 Armature by Jean-Baptiste Perin;
Camera Changer by Regis Montoya (3R);
Interactive Console by Campbell (ideasman).
- tiny updates in other scripts.
His fix was good, except that it switched to project_float instead of project_short, and that's not good because we're renormalizing the vector after that, which maximizes the float errors. Since it's scalling the axis anyway with the zfac method, it can use project_short safely (I tested it with the bug tracker file that triggered the fix in the first place).
Used the official list: http://docs.python.org/ref/keywords.html
While "as" is not a keyword, it is (should be) always used as such, so I've added it to the list too.
MEMSET was clearing the TransInfo struct after the constraint setup call has been made.
Temporary fix until the manipulator transform init is split off, like normal transform.
when zooming in with frame enabled, partial of frame were drawn with
multiple same int labels. confusing if harmless.
Ton, can you check this fix is the proper one ?
Patch provided by Ricki Myers. Works quite obvious, see Theme editor too!
Notes about the provided code;
- The default syntax colors were screaming! Toned it down to match the
default dark on lightgrey background better.
- Added: initializing colors in saved themes (usiblender.c)
- The implementation of the button for this option was quite clumsy...
Blender UI options services this a lot easier.
(Same fixed for 'line numbers' option)
- Bracket matching now uses as color a mix of backdrop and the
selected-text color. Noticed my texteditor did it too...
-> I really miss comments in code describing a little bit the thought
process behind the code. Like a short introduction on this feature
in the top of the drawtext.c, little remarks on new functions.
ALso in patch tracker or the mailinglist no docs was mentioned?
-> drawtext.c now gets messy quickly... lack of overview, structure,
and confusing mix of personal coding styles.
For not-active supported code dangerous...
should disable that, since animation systems remain in their own local
system time. Framemap is only for playback!
Not doing this caused corrupt baking, bug #2531 reported it.
- Bug #2530
The MMB Transform constraint 'chooser' used projection code that didnt
work when zoomed in extreme. Needed a correction to scale projected
coordinate according zoomlevel.
- NKEY panel for objects: click on left/right side of scale button now
goes with increments of 0.1 (was 1.0)
- Moving window edges in Screen: prevented top header from becoming too
small due to grid-snapping. Was visible now with using Texture Font.
Armatures with > 100 bones can deliver a 'parent IK menu' that doesn't
fit on screen anymore. Fixed it with changes;
- when menu doesnt fit to right/left, it places it aligned to left hand
of window
- when more than 8 collums in a menu exist, it makes collums 25 items
instead of 20.
Also fixed for the popup menu on Bone parenting. On a 1280 screen we now
can have a menu with over 300 bones (even more when you zoom down the
buttons window a bit). Still not optimal, but at least it's less annoying.
CTRL+V on buttons
NKEY Panels
join mesh
join curve
editing buttons
boolean
vpaint
faceselect
Manipulator
Also; Transform() got in useless loop when you entered without anything
selected. Not sure why Martin recoded it this way... maybe as a first
step to handlerify it? For evil Python Aussie Bosses? :P
Fun, the error is in code committed by Kent Mein, in a patch provided by
Campbell, who also reported the bug. :)
There was a line that set ALL selected objects (on click) to become active,
in succession, in a loop... very bad.
Activating in Oops better be restricted to ctrl+click. Fixed!
Using as Pivot the "individual object centers" didn't do any rotation or
scaling in editmode. I think it was supposed to give some kind of
'connected' centers, for individual faces/edges rotate. That'll be some
more coding work to get to work. :)
