* Refactored F-Curve Modifier stack evaluation into functions which can be reused for F-Curve Modifiers on NLA-Strips
* Started setting up temporary accumulation buffer system for use when evaluating strips
* Loading files saved with 2.5 works again now without crashing. Was crashing when trying to load Grease Pencil data (even though there was none)
* Fixed some msvc compiler warnings in own code
Time-Modifying F-Curve Modifiers now get special callbacks to allow them to specify what frame they need to be evaluated on, instead of forcing a re-evaluation of the preceeding curve + modifier-stack. This should be more robust than the old way in general.
It still remains to be seen if some tweaks to this are still needed, as the full consequences of the propogation of modified time-spaces have yet to be fully explored.
For now though, evaluation works by finding the last modifier on the stack which modifies time, and asks it what time it modifies the given time to. This modified time is used to evaluate the F-Curve data only. The modifier stack gets evaluated using the original time instead.
Restored the code to get this working. I haven't tested this to verify it, but it 'should' work...
Also, deleted old gpencil.c file in editors, since it was causing compiling errors.
Now it is possible to write a Python Expression using the variable names for driver targets (see mockup from initial commit) to substitute the appropriate values into the expression.
In the __global__ namespace for PyDriver evaluation, the following modules are available:
* __builtins__ - i.e. the builtin Python functions
* bpy - new Python API
* math or m - math module
For example:
Consider a driver with three targets, named: A, B, C
Now, you could write an expression like:
C if A < 5 else B
or even:
2*C if A < 5 or A > 20 else m.PI*B
Of course, you don't have to have three targets, the above was just an example.
TODO:
* Bring back way to load pydrivers.py
* Blender.Noise equivalent would be nice to have
P.S. I hope I haven't made any terrible Python API coding errors here (i.e. mem leaks, etc.)
The Python API to define Panels and Operators is based on subclassing,
this makes that system more generic, and based on RNA. Hopefully that
will make it easy to make various parts of Blender more extensible.
* The system simply uses RNA properties and functions and marks them
with REGISTER to make them part of the type registration process.
Additionally, the struct must provide a register/unregister callback
to create/free the PanelType or similar.
* From the python side there were some small changes, mainly that
registration now goes trough bpy.types.register instead of
bpy.ui.addPanel.
* Only Panels have been wrapped this way now. Check rna_ui.c to see
how this code works. There's still some rough edges and possibilities
to make it cleaner, though it works without any manual python code.
* Started some docs here:
http://wiki.blender.org/index.php/BlenderDev/Blender2.5/RNATypeRegistration
* Also changed some RNA_property and RNA_struct functions to not
require a PointerRNA anymore, where they were not required (which
is actually the cause of most changed files).
* Added extra parameter to generic unique name finding function BLI_uniquename() for specifying the delimeter between non-unique parts of the name and digits.
* Driver target variables now get unique names by default.
Drivers now support multiple targets which act as 'variables'. The targets have a short 'name' (see later), and reference some property (in much the same way as F-Curves do, using RNA-Paths) which acts as the 'value'.
These named variables can then be used in a Python Expression which relates them to each other for more fine-grained control over the result of the driver. By using only the names of these variables in the expressions, we are able to define expressions/relationships in a much more readable way, as data access is separated from data use. This makes the underlying relationships easier to understand.
By default, if no Python Expression is given, the variables are simply averaged together, so old files won't break. :)
For example, check the following diagram (thanks Cessen/Nathan V from Peach team):
http://download.blender.org/ftp/incoming/250_drivers_mockup_cessen.png
TODO List:
* Depsgraph building for new driver relationships doesn't work yet. This needs to be recoded again, but this new system makes this much easier, since the targets are clearly defined (i.e. no need to parse py expressions to get list of objects)
* Graph Editor interface for editing these needs to be rewritten
* Python function for evaluating these expressions is needed (Campbell?)
This new modifier clamps the values of the F-Curve to lie within specified bounds, much like Limit Location/Rotation/Scale constraints do. You can limit by time range(s) and/or value range(s).
* Copying objects with animation data now works correctly. Previously, actions were not getting copied correctly, leading to loss of data in some cases.
* Action and Graph editors now display the name of the Action concerned (for 'Action' folder channels), making it easier to tell which AnimData blocks are sharing the same actions
* Added some code to make relative KeyingSets (converted from absolute ones) work better.
- I key over a button inserts a keyframe.
- Alt+I removes a keyframe.
- With right mouse button a menu with these options pops up.
- Buttons are colored green if the property is animated, yellow
if it is on a keyframe. I followed the colors from the UI
mockups, but the flicker on keyframes seems too distracting in
practice?
- This only works for properties on the ID itself at the moment,
path callbacks need to be filled in for all structs but mesh
still.
- It doesn't work when you're over a related label, that needs to
be made to work.
- I made it insert keyframes outside of any keyingset. Not sure
how this is supposed to integrate?
When inserting keyframes in the 3D-View (support will be extended to other editors in due course) using the IKEY hotkey, the menu which appears will now consist of 3 parts:
* 'Active Keying Set' - this option allows you to use the user-defined KeyingSet which is active for the current scene (i.e. the one seen in the TimeLine/Outliner headers)
* User defined Keying Sets - a list of all such available KeyingSets is included, and entries can be chosen from there
* Built-In Keying Sets - see later...
To achieve this, several changes needed to be made first:
* Added support for 'relative' in addition to 'absolute' Keying Sets. Relative Keying Sets are Keying Sets which operate on data from the current context (i.e. a 'location' KeyingSet will add location keyframes for selected objects/bones/nodes as opposed to location keyframes for some particular object). The is a tentative 'templates' requirement system here, which still needs to be fully fleshed out.
* Added support for builtin Keying Sets (i.e. 'Location', 'Rotation', 'Scaling', and 'LocRot' as a few initial demonstrations), which replaces the temporary Insert Keyframe operator for the 3D-View (IKEY). These are effectively relative Keying Set definitions which are included in Blender by default and stored in a list separate from user-defined ones. Volunteer help in defining a few more of these for other editors will be welcome soon.
* Removed/replaced much of the crappy temporary Keyframing operator code, though a few tweaks could still be done.
This setting means that instead of blindly replacing the existing values, the generator modifier will instead apply its effects on top of any existing modifiers (and/or curve data).
Thus, it is now possible to apply effects such sin/cos-based oscillations on top of keyframed motion.
* Envelope modifier now works correctly. Previously, the mapping was being done in the wrong order.
* Extra controls for modifiers now only draw when the F-Curve is the active one, so the view will be less cluttered in other cases.
Got the basic envelope modifier code working, including primitive drawing of relevant helper info in the graph view. It doesn't work in a very intuitive way yet, so I will recode it soon.
* Added GUI and fixed bugs in the Cycles Modifier. This replaces the old Cyclic Extrapolation settings, giving more fine-grained control over the results. You can now specify whether the keyframes are repeated before and/or after the range independently, also, the maximum number of cycles on either side can be controlled.
* TODO: it would be nice to have the last value held for cyclic+offset.
* Deleting modifiers now works
* Finished code for Expanded Polynomial and Factorised Polynomial UI's.
* Started UI code for 'Builtin Function' mode. There are still 4 controls to add there to use something other than simple mapping
* Finished/fixed up verification code for these so that values get initialised ok.
* Currently, this only works for the 'Expanded polynomial' mode, but this will be expanded to include the other modes too. Now you can modify the values and interactively see the graph in the view change.
* Disabled the backdrops (modifier 'panels') temporarily, as ROUNDBOX UI elements currently swallow all events, which is not good.
Note: the code here still uses the old-style UI definition code since the new stuff is still under heavy construction.
* Rewrote the Generator modifier to be more efficient and support more options
* A few UI tweaks for this, but the UI for this is still not yet functional though.
- Completed cleaning up the drawing code so that F-Curves with modifiers now get drawn to reflect this.
- Added a temporary operator to add modifiers (hotkey Ctrl-Shift-M)
This operator can be activated using the 'Alt-C' hotkey for now, and operates on selected + editable F-Curves. This is currently still highly experimental, and does crash
I've implemented this as a way to test out the FPoints/samples code, which will be used to provide better support of the dense F-Curves which result from importing Mocap/BVH data. These should use considerably less memory + have a few additional benefits over keyframes when they're working in a stable fashion.
* RNA_blender.h is now generated along with the other files. It is not
used anywhere yet, and still located quite hidden next to the other
rna_*_gen.c files. Read only access for now.
* Inherited properties are not copied from the base anymore but
iterated over. Patch by Vekoon, thanks!
* Array get/set callbacks now do the whole array instead of getting an
index. This is needed for some layers for example so python can set
the array as a whole, otherwise the check that one layer has to be
enabled at all times gets in the way. Also nicer for the C API.
* Also some changes to returning pointers to make the API cleaner, got
rid of the type() callback and instead let get() return PointerRNA
with the type included.
The C API looks like this currently:
http://users.pandora.be/blendix/RNA_blender.h
* Drivers view in Graph Editor now displays drivers only, instead of displaying normal Animation data.
* 'Materials' channel is now only shown under an Object when there are Materials with animation data...
* Hid more debug prints behind debug flag. These should be removed...
Only object layers have been done for now. The implementation of this should provide a good indicator for how this should be done. Unforunately, the code for animating layer visibility is not that straightforward, since scene->base->layer (not set) settings need to be synchronised with the object->layer (set by Animato) settings.
* ShapeKey conversions should now go to the correct places
* Added fix for drivers to prevent crashes when no driver object is provided (Franky from BBB seems to have one such driver for some reason).
* Temporarily disabled a check when evaluating drivers for whether the driver is tagged for evaluation. Drivers still need to be correctly tagged AND ordered by the Depsgraph to work correctly.
* FModifier data is now allocated correctly, solving crashes on loading files containing IPO-curves with cyclic extrapolation
* Fixed memory leaks with FModifiers and their data not being freed correctly. Not sure exactly why the old code didn't work for this.
This is work-in-progress patching support for converting animation saved in old system to work in the new one.
* Only IPOs/Actions directly attached to Objects + Shapekeys are converted for now. More types will follow...
* This is currently done as a step outside of do_versions() due to problems with various pointers not having been resolved yet, but which are necessary for correct resolution of issues such as drivers... However, the current code does illustrate how the data should be converted to give best results/compat between the two systems.
* Still need to get the converted data working with depsgraph correctly. Currently, some of my testfiles work, but the BBB files still don't.
* Added back Auto Preview-Range tool (i.e. set preview-range from keyframe extents)
* Restored delete keyframe tool. For now, this doesn't delete empty F-Curves, even though its keyframe-api counterpart still does. I still need to figure out how to do this in the best way.
* Fixed crashes when selecting keyframes in 'object' summary channels
* Removed prototypes for a few unused/depreceated functions...
* Removed nAction struct. We'll be using good ol' bAction structs again, but putting new data in a different list. Apart from that, the data is similar enough to do so.
* Rearranged code in DNA_action_types.h while renaming the structs to avoid confusion over what is currently in use...
* Added freeing and AnimData execution loops for many other ID-types too. (NOTE: I've added AnimData in NodeTree struct too, but it's not clear to me where the relevant data-management calls should go in Nodes code).
* File writing code should now only write the new data to files
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!).
