* With multiple objects selected, only one of the objects got keyframed. The code which was checking for duplicate paths was wrongly assuming to ignore the ID-block used still.
* Not registering a Keying Set as 'builtin' would crash on startup. I've made all Keying Sets fallback to adding as if they were local for now, but a better solution is coming soon.
* Fixed a typo in RNA function wrappers for the generator callback, since it was looking for the iterator only. This doesn't seem to have caused any problems (thankfully).
After a few days of wrong turns and learning the finer points of RNA-type-subclassing the hard way, this commit finally presents a refactored version of the Keying Sets system (now version 2) based on some requirements from Cessen.
For a more thorough discussion of this commit, see
http://sites.google.com/site/aligorith/keyingsets_2.pdf?attredirects=0&d=1
------
The main highlight of this refactor is that relative Keying Sets have now been recoded so that Python callbacks are run to generate the Keying Set's list of paths everytime the Keying Set is used (to insert or delete keyframes), allowing complex heuristics to be used to determine whether a property gets keyframed based on the current context. These checks may include checking on selection status of related entities, or transform locks.
Built-In KeyingSets have also been recoded, and moved from C and out into Python. These are now coded as Relative Keying Sets, and can to some extent serve as basis for adding new relative Keying Sets. However, these have mostly been coded in a slightly 'modular' way which may be confusing for those not so familiar with Python in general. A usable template will be added soon for more general usage.
Keyframing settings (i.e. 'visual', 'needed') can now be specified on a per-path basis now, which is especially useful for Absolute Keying Sets, where control over this is often beneficial.
Most of the places where Auto-Keyframing is performed have been tidied up for consistency. I'm sure quite a few issues still exist there, but these I'll clean up over the next few days.
Added a check in RNA_property_animateable() which checks if the base ID-block can have animation data or not. Screen data currently cannot have animation data, so this solves that problem (where there were non-functional entries there in the menu).
F-Curves now internally store radians again instead of degrees.
- This solves problems with inconsistencies when working with drivers.
- No need to version patch old files, potentially screwing them up. As such, removed the version patching for F-Curves.
- Is better suited to optionally showing radians throughout the UI instead or degrees.
As a result, values are now converted on the fly in the Graph Editor for display and operators that operate on values. I've made the conversion system for this rather general, so that other unit type conversions can also be hooked up with the type conversion backend.
Also, made some tweaks to F-Curve RNA wrapping to make it represent the data better.
TODO:
- Transform code currently still needs to be corrected to work with these changes. Currently moving keyframes for rotation curves will make them change too rapidly vertically when using degrees.
Rotations are now stored internally as radians, while exposing degrees in the UI -
in the graph editor and UI controls. This is done in two areas:
1) Using the unit system to convert RNA data to display as degrees in the UI controls
2) FCurves now use degrees for rotation, so you can edit in the graph editor what
you see in the UI.
All rotation data is consistently accessible in DNA and RNA as radians, degrees are only
used for the UI controls and graph editor.
This commit includes conversions will convert old files (stored data and also fcurve data)
to the new units, hopefully everything should go smoothly!
Part of this also changes a few properties that were hard-coded as degrees before (such
as IK pole angle and brush texture rotation) to also use the same consistent system of
radians (dna/rna) and degrees (ui).
Thanks to Joshua for hints and review here too.
* Mostly revert #24880: Transform Locks affecting animation system. This
was implemented as a feature request from Nathan but was not the
intended functionality, and actually gives problems instead. The transform
locks should really only affect what the user can do, not if the property
can be animated internally.
* Revert #25868: fix for lib linked scenes not animating. This was needed
due to the above feature, but meant that all lib linked properties were
editable in the UI.
* Make bone properties of on proxy-protected layers not editable. They will
be overriden on reload/redo like lib linked data, so should not be edited.
* Fixed Driver version-patching code to work correctly again with the new system.
* Fix for bug #20484, by adding a new driver variable type ('Transform Channel') which makes it easier to use object/bone transforms as in the past. The main differences with using this (compared with the 'Single Prop' type) are that this allows for 'final' transforms to get used instead (i.e. constraints are also taken into account), and also that this variable type can only be used for transforms (more limited scope -> less flexibility -> point-n-click goodies can follow). Mancandy now loads correctly again.
* Added toggle for local vs worldspace transforms when working with Rot/Loc Diff variable types, and also for the newly added Transform Channel
* Removed some dead code from sequencer...
Highlights:
* Support for Multi-Target Variables
This was the main reason for this recode. Previously, variables could only be used to give some RNA property used as an input source to the driver a name. However, this meant that effects such as Rotational Difference couldn't be used in conjunction with other effects and/or settings to achieve the powerful results. Now, a variable can take several input targets, perform some interesting operations on them, and spit out a representative value based on that.
* New Variable Types
With the introduction of multi-target variables, there are now 3 types of variable that can be used: single property (i.e. the only type previously), Rotational Difference (angle between two bones), and Distance (distance between two objects or bones).
* New Driver Types
In addition to the existing 'Average', 'Sum', and 'Expression' types, there is now the additional options of 'Minimum' and 'Maximum'. These take the smallest/largest value that one of the variables evaluates to.
* Fix for Driver F-Curve colouring bug
Newly added drivers did not get automatically coloured in the Graph Editor properly. Was caused by inappropriate notifiers being used.
Notes:
* This commit breaks existing 2.5 files with drivers (in other words, they are lost forever).
* Rigify has been corrected to work with the new system. The PyAPI for accessing targets used for the variables could still be made nicer (using subclassing to directly access?), but that is left for later.
* Version patching for 2.49 files still needs to be put back in place.
* Mesh data/settings can now be animated. It is not recommended that geometry be animated directly, but other settings such as autosmooth, etc. can be...
* Code cleanups for depsgraph, making sure that drivers get included for all object data types.
The Animation System now respects the Transform Locks too (i.e. lock x-location, etc.) when writing settings. This means that it is no longer necessary to set up "constant drivers" to make sure some values don't get accidentally animated.
Internally, added a new callback for properties in RNA, which is responsible for checking if the item at some array-index is editable. This needs to be manually called for each place which uses rna to set settings for arrays (see the code changes in anim_sys.c for changes how to do this; the same thing needs to be done in the UI code too, and probably in py-api too)
* Convert all code to use new functions.
* Branch maintainers may want to skip this commit, and run this
conversion script instead, if they use a lot of math functions
in new code:
http://www.pasteall.org/9052/python
* RNA Path fixing when renaming data now checks if a path in question cannot be resolved before trying to fix it. This should reduce the number of misindentified cases I hope.
* Silenced compiler warnings for EdgeSlide stuff that mingw was making about unused variables.
* Now the old/new names get tagged with [" "] before the search and replace operation, which should alleviate problems with searching for 'bone' and ending up with all instances of 'boney' 'boney.r' etc. also getting renamed.
* Cleaned up some compiler warnings, and removed an unused function from an earlier attempt at this work.
RNA Paths used in F-Curve, Drivers, etc. now get renamed when some data that they use gets renamed. This only works when things like Bones, Constraints, Shape Keys, and Modifiers get renamed, but other cases can get added easily.
The code here only performs simple string replacements, so there is the potential for problems when several sets of data with the same names are present. For example, if there are multiple armatures with bones that have the same names, renaming a bone on one armature (with a bone on another armature having the same name) will break all the drivers on the other one, even though they aren't really connected. However, I don't expect the aforementioned scenario to really be a problem in most production scenarios.
Added a way to view and edit Keying Sets via the Scene Buttons. These are still some tweaks needed to make this really workable, but should still work well enough for simply viewing and tweaking existing Keying Sets created using other means.
Additional bugfixes:
* Adjusted the size of labels on properties that had a 'label' for their name. Now it uses 1/3 of the total width instead, which looks much better for most cases.
* Added missing entries for adding Force Fields from the Info-header 'Add' menu. At some point we should unify this menu with the popup operator's one, since this is exactly the kind of situation we had hoped in avoid with new UI architectures.
* Moved all the operator defines for keyframing stuff to the 'intern' anim header instead
* #19501: Only the first user of multi-user IPO's were getting converted to AnimData.
Now, this AnimData gets converted multiple times - once for each user. This will mean that multi-user actions will no longer be multi-user after conversion though, although this could be fixed manually if there really is such a need.
* #19503: Nasty memory leaks when duplicating objects with AnimData
Fixed a few little oversights made when coding the copying code for NLA-data (which resulted in exponential copying-loops of doom), and sanitised the AnimData copying code for ID-blocks to be simpler to manage.
Main Feature:
* It is now possible to choose which AnimData block is the 'active' one for editing, and/or select them too. AnimData blocks are generally the dark blue and lighter-blue expanders (i.e. Scene, Object, Camera, Lamp, Curve, Armature, etc.)
* Objects are no longer selected/deselected when AKEY is used to toggle selection of channels. This was getting a bit annoying.
* Following on from selection of AnimData blocks, it is now possible to select/make active an AnimData block in the animation editors, and change the active action for that block via the 'Animation Data' panel in NLA Editor's properties region.
--> Be aware that user-counts are not totally handled correctly there yet, so some funky behaviour might be seen...
--> It is possible to assign a new action, or to assign an existing one, allowing to switch between actions as in the past with Actions/IPO Editors...
Other tweaks:
* Some code tweaks towards making the 'Euler Filter' feature for Graph Editor working sometime soon
* Added some backend code for snapping the values of keyframes to a single value. Still need to work out some UI for it though.
* Shuffled the code for ACT_OT_new() around, and removed the poll() callback so that it worked in NLA too.
* Fixed some more notifier bugs with deleting bones and a few other editmode operations for Armatures.
* delta-transforms for objects should work again. These were basically extra transforms that could get added on top of the values calculated from animation values.
* Added some skeleton code for fixing paths when some data needs to be renamed.
* Updating Preview Range settings results in correct updates for Animation Editors
* Compositing nodes now correctly animate when the values were set by IKEY/RMB on suitable node parameters. Beware that these nodes are not relinkable node-trees, hence the standard code not working.
* Ctrl-P to parent objects to bones now works in Pose Mode too. I needed to add a special keymap entry for this, though I thought this would have been better to be automatically inherited/present from Object keymap already?
* Ctrl-P -> Parent to Bone option now works correctly again. 1.5 lines of code missing here...
* Breakdowns tool now shows custom cursor during 'modal' phase so that it's not that confusing what's going on.
Added a utility function to check which transforms for an object or bone are animated, returning these as bitflags and/or optionally retrieving the relevant F-Curves too. Beware that this method may not be working correctly yet, but it shouldn't hurt anyone in the meantime :)
Also, split RNA-path building function up into a version which only creates the path up to the given struct, with the other parts being added later.
* Split object_edit.c into multiple files:
object_add.c, object_edit.c, object_hook.c, object_relations.c,
object_select.c, object_transform.c.
* Rename files to have consistent object_ and mball_ prefix:
object_shapekey.c, object_lattice.c, object_vgroup.c, mball_edit.c.
* Added operators:
* vertex group menu and set active
* apply location, rotation, scale, visual transform (location is new)
* make local
* make vertex parent
* move to layer
* convert to curve/mesh (not finished yet)
* Many small fixes for marked issues, but still much code to be cleaned
up here...
Example code: http://www.pasteall.org/7332/c.
New API functions: http://www.pasteall.org/7330/c.
Maximum number of dimensions is currently limited to 3, but can be increased arbitrarily if needed.
What this means for ID property access:
* MeshFace.verts - dynamic array, size 3 or 4 depending on MFace.v4
* MeshTextureFace.uv - dynamic, 2-dimensional array, size depends on MFace.v4
* Object.matrix - 2-dimensional array
What this means for functions:
* more intuitive API possibility, for example:
Mesh.add_vertices([(x, y, z), (x, y, z), ...])
Mesh.add_faces([(1, 2, 3), (4, 5, 6), ...])
Python part is not complete yet, e.g. it is possible to:
MeshFace.verts = (1, 2, 3) # even if Mesh.verts is (1, 2, 3, 4) and vice-versa
MeshTextureFace.uv = [(0.0, 0.0)] * 4 # only if a corresponding MFace is a quad
but the following won't work:
MeshTextureFace.uv[3] = (0.0, 0.0) # setting uv[3] modifies MTFace.uv[1][0] instead of MTFace.uv[3]
Wrapped the Keying Sets API with RNA Functions so that they can now be called from Py-Scripts. This will ultimately be useful for riggers to create Keying Sets which can get loaded up/created for animators to use after importing their rig.
I've created a demo for this, which can be found at:
http://www.pasteall.org/blend/552
Notes:
- Kazanbas, I've had to create a rna_scene_api.c here to hold some of the relevant functions. Hopefully this won't cause you too much pain when you do your next merge from 2.5 to your branch ;)
- I've noticed that there seem to be a few cases mentioned in the demo which don't totally work yet. I'll commit some fixes for those later.
* F-Modifiers on F-Curves can now taken into account when calculating the extents of actions. This is used when there are some NLA strips and some action with some F-Modifiers is being played back on top of those.
* The toggles in the NLA channels list now respect the width of the list instead of using a hardcoded position. This means that clicking on these toggles when the list is resized works again.
Implementation Note:
* Moved the scene copy/unlink code back into blenkernel, with
the exception of the copy single user stuff which is still in
object_edit.c.
* Uses SCREENDELETE notifier like SCREENBROWSE, seems only clean
way to do this now.
When there were no strips but some empty tracks, the active action should be evaluated normally. This is not an optimal solution (more user testing + suggestions regarding the best way to do this is welcome).
* Added some optimisations to avoid having to try evaluating some data that won't have any effect.
* Converted playback buttons in timeline header to use operators too
* When doing realtime recording of animation (i.e. transforming objects + bones while animation playback is running, and auto-keying is enabled), animation will be added to a new NLA Track+Strip combo everytime a single 'loop' of the frame range has finished. This will allow 'passes' over the animation to be less destructive.
* Made the evaluation of the active action (when NLA data is present), be handled as part of the normal NLA system evaluation code (as if it were just another strip in a track at the end). The immediate benefit is that there are now some settings (available in the "Animation Data" panel in the NLA Editor with a strip selected) which allow for the way the active action is combined with the NLA stack results. For instance, the way that the action extrapolates is used in the recording tweaks above.
* Fixed crash when anim-playback is running and a strip beside a transition gets transformed.
Transition strips no-longer assume that their neighbours are action-clips, using the standard NLA-strip evaluation function instead to evaluate their neighbours. However, a check for ping-pong recursion needed to be added there, so that a transition beside a meta-strip, with the meta having a transition nested at the start of one of its levels, wouldn't fail with stack overflow.
* Moved 'Tweak Mode' menu entry to Edit menu, since it's not really that modal.
* Renamed the 'blend' blending mode to 'replace', since that's what it usually does
* Drawing a darkened rect behind the keyframes shown in the action line
--
* Fixed typo made last night which broke compiling
* Consolidated all the keyframe-shape drawing code to use a single codebase. Even if we don't ultimately go with OpenGL keyframes, there's always a tidy option for that now.
F-Modifiers now work on 'all' types of NLA Strip. To get them working on Meta-strips, F-Modifiers on the Meta strip must be applied on top of those for child strips. For now, this is done by joining the lists of modifiers, and evaluating the joined list.
Currently, Transitions don't work that great with F-Modifiers yet (only the endpoints get evaluated with the F-Modifiers). Solving this is for another time...
Previously, the quick-hack I coded for Meta-strips evaluation didn't really take into account the possibilities to use Meta-Strips as strips in their own right - i.e. reversed, muted, time-mapping-curves, influence-blending - were all unavailable.
This commit makes the aforementioned capabilities of other strips available for Meta-Strips too. The evaluation is now kind-of recursive (as with most of the other methods which take them into account) so that each 'level' of strips get evaluated correctly within their frame-of-reference.
TODO:
* F-Modifier support for Metas...
Refactored the backend code/API's to support 'meta' strips (i.e. strips containing other strips). These have been implemented to be nested to 'unlimited' depths (in terms of common usages that is, though there is a very remote chance of stack-overflow in theoretrical evil cases only that shouldn't ever be encountered in production).
This paves the way for implementing the necessary tweaks needed for the transform code (in addition to some cool user-level tricks)
Groups (the Action variety) can now be muted, and also be set to not be drawn in the Graph Editor. These settings get applied to all the F-Curves within the group, overriding any settings individual F-Curves might have, allowing users to quickly mute or hide groups of curves without having to go through clicking on all of them.
Also, added a flag that can be used to set the curve visiblity on AnimData level too. This will be enabled in a future commit.
