* I've replaced the long lists of #defines in header_action.c with enums, which should make that easier to maintain/add new items as needed.
* Added a few comments also
Multiple selected bones are now able to be parented to the active bone in EditMode. Previously, only one selected bone could be parented to another at a time.
The Action Editor can now display timing in seconds too. By default, it still displays timing in frames, but it is possible to switch the time display to seconds.
The hotkey to toggle this is Ctrl T.
Notes:
- Many of the Action Editor's tools have yet to be made aware of this. Therefore, they will still work as though frames are used. Only Transform, Snap, and Mirror should really be affected. This will be fixed when I wake up.
- Nothing *should* be broken, but I might have missed something.
Made inserting new keyframes use a coarser threshold for determining when to replace existing keyframes with new ones. Now, it shouldn't create so many keyframes stacked on top of each other, but yet still keep some bizzare cases working.
- Removing evil InitTransModeFlags (I would always forget to update that when adding new stuff). It's properly done in the per transform init now.
- Add toggleable mode for Shear (with MMB) for horizontal or vertical shear. Toggle shear mode also changes the input mode from vertical to horizontal and vice versa (it's intuitiver that way). (This was suggested in a really old patch but never implemented because I disliked adding transform specific handling in the main event switch).
- Add custom event handlers per transform (currently only used for Shear)
- Add custom data pointer and freeing flag (only (a)bused by Shear).
NOTE to Levi Schooley: Some stuff in there might be useful for you.
After complaints from wavez, Snap Cursor to Selected for Bones in EditMode now results in the cursor being snapped to the midpoint of the selected joint(s) like it did before my no zero-length bones commit.
This problem was resulted from the handles on keyframes not being mirrored as well. I thought that the handle resetting function would solve that, but obviously it didn't.
If you look at the log, the last change to this file I was trying to fix
game engine bug #6667. In so doing I added a print usage if you do not
pass any arguments to the gameplayer.
Things should work better now.
Kent
I've finally come to the conclusion that the old code used to clean up IPO-curves was utterly unsafe/unstable. It used to have a lot of memory-related bugs early on, which I had resolved, but still I found a file where this broke quite badly (thanks Cessen).
Now the code should be saner, as it does away with that evil memory copying that seems to be at fault. It would be advisable to have this properly tested for a few days before considering it for inclusion in 2.45 (stable) branch.
Preview Range could get set with frame numbers less than 1. This caused problems with playback in a few cases, and also gave gibblish in the counter-cursor thing.
This fixes [#6996] Sequence Nesting and enables sequence nesting for the
first time :)
The old hack done by Ton prevented eternal loops by preventing Sequence
tracks being added to Sequence scenes.
We now disable "Do sequence" temporarily for the _current_ scene, which
has the same effect but leaves the possibility of sequence nesting
untouched.
Also fixes a warning in editseq (uninitialized variable).
This adds "[#6766] Transform Sequencer effect" by damiles.
* Configurable interpolation: no, linear, bicubic
* Makes it possible to specify coordinates in units of pixels or by percent
of the picture size
It makes the code a lot more clean, since new imbuf functions are used.
This patch also fixes
[#6829] Limit on VSE Transform Effect
bsystem_time was being called with an extra variable, which was useless. Most of the places that called it, were passing NULL for that variable anyway.
I've also cleaned up that function a bit, but the underlying problems with that part of the code still exist (EVIL GLOBALS that are exported for frame_to_float), for mblur and fields rendering features. That remains for another time.
In this commit, a new way to set the bone roll is introduced. It aligns the z-axis of the selected bone(s) to the 3D-Cursor, so that the cursor intersects with the YZ plane.
This makes it easier to align bone rolls to a target, for use in rigs where arms/legs are not horizontal/vertical. To use this, just press Ctrl N, and choose the "Align Z-Axis to 3D-Cursor" option.
Credits go to Robert Christian (wavez) for providing the method as a working BPy script, that demonstrated this in action.
* Now, when X-Mirror is turned on for the armature, the roll from bones being 'rolled' is also mirrored.
* Disabled transform-constraints for this mode, as it didn't really make sense.
Now the Outliner can be scrolled horizontally too. This was the first patch I ever submitted, but it's gone through many revisions due to ugly Blender bugs that needed to be fixed.
Code Notes:
* I discovered an ancient bug which would cause Blender to hang when loading a file saved with horizontal scrollbars turned on for the Outliner/OOPS.
* Therefore, I've added special B_SCROLLO and HOR_SCROLLO defines for use by the Outliner only. These are used in place of B_SCROLL and HOR_SCROLL so that older Blender's won't choke on this stuff. Thanks for this suggestion Ton.
* The hanging occurred in draw_scroll in draw_ipo.c
Recent font preview commit didn't have the proper enclosing #ifdef INTERNATIONAL
blocks. Honestly, I have no idea why we bother, but added them anyway.
===============================
This is a new armature deform interpolation method using Dual Quaternions,
which reduces the artifacts of linear blend skinning:
http://www.blender.org/development/current-projects/changes-since-244/skinning/
Based on the paper and provided code:
Skinning with Dual Quaternions
Ladislav Kavan, Steven Collins, Jiri Zara, Carol O'Sullivan.
Symposium on Interactive 3D Graphics and Games, 2007.
This code was always kludgy and this clean it up a bit.
insertmatrixkey() has been properly renamed insertkey_float().
Matrix calculations have been excised from the interface code,
and placed in insertmatrixkey(), so that you can call it from
anywhere without preliminaries and the proper values are calced
based on the passed adr code.
By much request, several semi-bug reports and discussion with
production animators, visual keying is now used automatic
for objects and bones that have constraints that cause them
to ignore their Ipos (CopyLoc, TrackTo, etc.). In those cases,
visual keying is used instead of the normal Ipo insertion
method. This "auto" functionality is toggled from the
"Use Visual Keying" button found along with "Needed" and
"Available" in the Edit Methods prefs.
Logic as to which constraints trigger visual keying on
which adrcodes can be tweaked in match_adr_constraint()
src/editipo.c
This has been tested by a couple of people, myself included,
but I may not have hit every constraint case, so evolutionary
feedback is welcome.
* Swapped the order of the CSpace menus. This is so that it makes more sense with the Transform Constraint, to follow a source (target) -> destination (owner) flow.
* 'Local Space' option for Objects has been renamed 'Local (Without Parent) Space'
Previous bugfix of spec not being included in only shadow lamps had 2 errors.
There weren't any checks for if shi->spec and shi->shad go below 0 because of
an only shadow lamp, and also the code for including spec in only shadow lamps
was slightly wrong.
