* Particle collision with size was broken since raytrace optimizations by jaguarandi, now the collision code falls back to old slower method when the collision ray has a radius.
* Single goal/avoid object now works for boids.
* Some tiny improvements on collision avoidance for boids.
Basically two simple changes, changes, I pulled in the faster
ghash in bmesh (which uses mempools for allocation, providing
a substanstial speedup in some cases, and also I inlined some
of the functions), and I changed __inline to __forceinline for inlining
of math functions.
I also removed the timer in the view3d zoom op (ctrl-middlemouse)
that was making it nonfunctional. Why was that there?
* Selected keyframes in unselected F-Curves were still being modified, although not shown
* Selection tools for keyframes were changing the selection status of F-Curves, making keyframes sometimes disappear after trying to select another keyframe
* Warning fixes in a few other files
problem is that where_is_object is being called from multiple threads
but is not thread-safe, added a note about this problem, this commit
only solves the crash. Also remove the pushdata/popdata mechanism
that was being used here, using this kind of system is bound to give
problems with threading.
replaces another so it can do updates (e.g. dopesheet editor can
sync channel selection).
Also coded a simple optimization for allocating small objects,
based on mempools. It's #ifdef'd out, you can enabled it by
defining OPTIMIZE_SMALL_BLOCKS (e.g. adding -DDOPTIMIZE_SMALL_BLOCKS to
your compiler flags).
We suffer from a great deal of performance loss from the system allocator
(vgroups, ghash, edgehash, the singly-linked list implementation in blenlib,
editmesh, and likely a great many areas I'm forgetting), and this is the
common solution for handling the many-small-objects problem. It's not
really production-ready yet (it's long-term memory consequencers need to
be profiled first, and the implementation tweaked as necassary), but for
people on systems with slow system allocators it's worth trying.
Note that since this creates a guardedalloc<->blenlib link, the build systems
need to be updated accordingly (I've already done this for scons, though I'm
not sure if the player builds).
Was very quick to do, now re-aquainted with node editor.
http://mke3.net/blender/devel/2.5/hue_correct_node.jpg
Todo: modes for affecting hue and value on the vertical axis as well as just saturation - or if an enterprising coder wants to give it a go, let me know and
I can help :)
similar to sequence editor.
--> http://mke3.net/blender/devel/2.5/color_balance_node.jpg
Also added 0 key (zero key) shortcut when mouse is over a button, to reset it to its default value.
Same as the RMB menu ->Reset to Default, except for color wheels, it only resets the hue/sat/value
components that that widget affects.
Peter/Xavier: The existing color balance code can generate NaNs (fractional power of a negative),
which causes havoc along the image pipeline. I added a check in the node code to prevent this.
Still plenty of potential for lots of better colour correction tools in the compositor, just needs time...
* Current value for drivers didn't get stored, which meant that the debug value never got updated to reflect the current state.
* Min/Max variable types were not working
Feature request for ZanQdo, which shows the intermediate values used in driver calculations (i.e. current value of driver, and current value of variables), allowing drivers expressions to be debugged. This is a per-driver setting...
Multiple background images displaying each on a different axis.
Changes made from the original patch.
- Use an enum rather then multiple booleans.
- Reduced the space taken up by the user interface.
- Made the image template compact display not show fields & premul options.
- Added readfile.c lines so old blendfile images are loaded.
- Option to hide BGpic UI (like modifiers & constraints)
- Use the index rather then a bgpic from the context for the remove operator.
note: could be good to use 1 image for both left+right, for eg, but for this to work as intended we would need to add image flipping depending on the axis so left this commented out for now.
do this on drawing instead since SCREEN_OT_animation_step isnt calling the notifier (assume this is to be more efficient?). this isnt slow so is ok to do on drawing.
rename BKE_image_user_calc_imanr to BKE_image_user_calc_frame
This commit makes the new-style Motion Paths work for Objects and Bones. Motion Paths can either be added for Objects (Object buttons) or for Selected Bones in PoseMode (Armature Buttons), and/or removed from these panels too.
Changes:
* Changed the way the baking code worked, since it was better to be able to bake a bunch of objects at once, instead of doing it per object
* Fixed a variety of bugs regarding initialising defaults and reading old files
* Added operators for Objects (like for bones), and replaced the existing code for bones.
* Fixed bug with baking code that was causing it to bake the wrong ranges
Todos:
* Frame number drawing is currently messed up, since the "cached" text drawing takes into account the object transforms.
* The new MotionPath panels currently appear as the first panels in the respective contexts, probably due to the order in which the files are included. This needs some fixing, though not sure what the best way is yet.
* Added a generic 'histogram' ui control, currently available in new image editor
'scopes' region (shortcut P). Shows the histogram of the currently viewed image.
It's a baby step in unifying the functionality and code from the sequence editor,
so eventually we can migrate the sequence preview to the image editor too,
like compositor.
Still a couple of rough edges to tweak, regarding when it updates. Also would
be very nice to have this region as a partially transparent overlapping region...
- selected bones on hidden layers would be selected for weight painting (confusing to the user). use the 'active' bone instead.
- when no weight group was set, the index used for the bone group was off by 1.
Hopefully this time I've finally found the proper fix for this. At least in the mini test-suite I made for this, the new inverse process seemed stable enough.
---
Also, tweaked the keyframe drawing code so that it doesn't convert the handles too when performing NLA mapping. This should provide some minor speed boosts when drawing keyframes under these conditions...
blender supports type changing for textures in a way that python doesnt.
add a new general function.
Example usage:
tex = bpy.data.textures.new("Foo")
tex.type = 'IMAGE'
tex = tex.recast_type()
Macro to give the number of users accounting for fake user.
ID_REAL_USERS(id)
Use this so you can remove a datablock if it has a fake users as well as apply transformations to it in the 3D view.
Move api function bpy.data.add_texture() --> bpy.data.textures.new()/remove()
Code-wise it looks now (again) that tab is the default. I hope that'ok. For bitwise operation it's cleaner IMO if the check is for positive values on them.
* TXT_TABSIZE is still harcoded to 4 spaces *
fix for freeing the undo buffer from a copied text block. (the copy had the old undo pointer)
since I only got an error, not a segfault Im not sure this is the real cause of the crash.
- particle set weight operator (Shift + K) and from the menu.
- mirror vertex groups operator can also flip weight group names.
a number of utility functions for weight groups added
int *get_defgroup_flip_map(struct Object *ob);
void flip_vertexgroup_name (char *name_r, const char *name, int strip_number); // moved from modifier.c
void copy_defvert (struct MDeformVert *dvert_r, const struct MDeformVert *dvert);
void flip_defvert (struct MDeformVert *dvert, int *flip_map);
So now tab is not ALWAYS converted to spaces.
This is stored by text datablock (what allows to do nice things in the future, as automatic check for the indentation type of the file).
Ideally we should redraw the other Text Editor windows after changing that (in case the same file is opened and the Property panel is also open). Not sure how to do that though.
I'm using TABSTOSPACES as the DEFINE flag because TABSASSPACES sounds too ugly.
(also fix for interface divisor bug)
transforming, due to the evaluation of the object transform without
constraints, and the driver not finding the property.
Constraints are now disabled in a way that should avoid this problem
* 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.
