Revision 5850e62 by Bastien Montagne November 7, 2012, 12:24 (GMT) |
Attempt to fix problems with local built py caused by own r51976. |
Revision d8443cd by Bastien Montagne November 7, 2012, 12:14 (GMT) |
New words... |
Revision d7fe029 by jens verwiebe November 7, 2012, 11:57 (GMT) |
Fix ITASK warnings |
Revision fff3f95 by Bastien Montagne November 7, 2012, 11:44 (GMT) |
Update to make it work with new debian-like py3.3 packages. A bit hackish (this arch-dependent stuff makes things a bit complex), and probably only working for amd64 arch, but at least should not create any regression... |
Revision 9a365d0 by Sergey Sharybin November 7, 2012, 11:41 (GMT) |
Motion Tracker: disable stabilization operators when active object is not a camera |
Revision 14dea2c by Campbell Barton November 7, 2012, 11:37 (GMT) |
fix for own error in recent orig-index refactor (reported as [#33112]) |
Revision 73d83fd by Sergey Sharybin November 7, 2012, 11:32 (GMT) |
Affine tracker do_versions were missed for objects tracs Run this code again, so files from 2.63 would be opened correct now. |
Revision f6c1596 by Campbell Barton November 7, 2012, 11:28 (GMT) |
prevent floating point exceptions from being raised in soft_range_round_up/soft_range_round_down(), caused by log(0). |
Revision a10ed84 by Campbell Barton November 7, 2012, 11:19 (GMT) |
fix [#33094] Even edge slide on multiple loops destroys geometry |
Revision 1c450d7 by Sergey Sharybin November 7, 2012, 11:18 (GMT) |
Fixed crashes when motion tracks are zero-sized This is probably versioning issue happened when both trunk and tomato were mixed to work on the same file. Anyway, there're few files here locally and it's probably other users do have the same files, so lets keep things safe here :) |
Revision 552c0e7 by jens verwiebe November 7, 2012, 10:52 (GMT) |
Fix a typo and more descriptve commet |
Revision a68416e by jens verwiebe November 7, 2012, 10:43 (GMT) |
OSX: some more version target cleanups and corrections |
Revision 53ede28 by Sergey Sharybin November 7, 2012, 10:18 (GMT) |
Invalidate sequencer cache when movie clip's input color space is changing Currently the only way is to invalidate the whole cache, for something smarter we'll need a dependency graph which would be able to deal with relations between clip and strip. |
Revision f727448 by Campbell Barton November 7, 2012, 09:28 (GMT) |
fix [#33106] Decimate modifier/collapse give bad result FLT_EPSILON was too small to use when checking if the edge collapse result is an error. add invert_m3_m3_ex(), invert_m3_ex() functiosn which take an epsilon to check the determinant, saves calculating it twice per edge collapse. |
Revision efc44d0 by Campbell Barton November 7, 2012, 07:16 (GMT) |
decimator - interpolate vertex normals rather then re-calculating them. faster and means degenerate faces wont `explode`. |
Revision be36789 by Campbell Barton November 7, 2012, 06:28 (GMT) |
decimator - no need to check abs() on normal length. |
Revision 8740b6c by Campbell Barton November 7, 2012, 04:13 (GMT) |
fix [#33108] Running save_as_mainfile breaks relative texture paths save-as with path remapping left the paths relate to the file written. |
Revision cdce3e0 by Campbell Barton November 7, 2012, 01:32 (GMT) |
code cleanup: - remove double promotions - use UI_DPI_ICON_FAC macro rather then dpi/72 - add assert to prevent creating jobs with WM_JOB_TYPE_ANY, which should only be used for finding jobs. |
Revision 852a41f by Campbell Barton November 7, 2012, 01:02 (GMT) |
style cleanup, also remove redundant call to set_listbasepointers in free_main(). |
Revision 5cf9d5e by Campbell Barton November 7, 2012, 01:00 (GMT) |
for utility checkers, use QUIET env var, not to print progress. |
|