7 years agoEnabling smooth modifier for curves
Daniel Salazar [Wed, 6 Jun 2012 00:03:39 +0000 (00:03 +0000)]
Enabling smooth modifier for curves

7 years agoSmoke: Fix rna names as suggested by Thomas Dinges.
Daniel Genrich [Tue, 5 Jun 2012 22:50:02 +0000 (22:50 +0000)]
Smoke: Fix rna names as suggested by Thomas Dinges.

7 years agoSmoke: Make Smoke density available to python via rna.
Daniel Genrich [Tue, 5 Jun 2012 22:43:29 +0000 (22:43 +0000)]
Smoke: Make Smoke density available to python via rna.

7 years agostyle cleanup
Campbell Barton [Tue, 5 Jun 2012 22:12:17 +0000 (22:12 +0000)]
style cleanup

7 years agostyle cleanup
Campbell Barton [Tue, 5 Jun 2012 21:54:21 +0000 (21:54 +0000)]
style cleanup

7 years agooperator to reset feather weights on all shape keys
Campbell Barton [Tue, 5 Jun 2012 19:39:12 +0000 (19:39 +0000)]
operator to reset feather weights on all shape keys

7 years agoedge split modifier now works more like pre-bmesh --- 3+ faces using an edge always...
Campbell Barton [Tue, 5 Jun 2012 19:24:01 +0000 (19:24 +0000)]
edge split modifier now works more like pre-bmesh --- 3+ faces using an edge always split.

7 years agoEnable cuda sm_30 kernels for buildbot
Sergey Sharybin [Tue, 5 Jun 2012 19:23:52 +0000 (19:23 +0000)]
Enable cuda sm_30 kernels for buildbot

7 years agoConfigurable start frame for movie clip datablocks as alternative to automatic start...
Sergey Sharybin [Tue, 5 Jun 2012 18:38:09 +0000 (18:38 +0000)]
Configurable start frame for movie clip datablocks as alternative to automatic start frame number

Number of start frame in opened image sequence used to be distinguished automatically
in a way that file name used on open would be displayed at scene frame #1.

But sometimes it's useful to have it manually configurable (like in cases when you're
processing image sequence and replacing clip's filepath to postprocessed image sequence
and want new clip to show at the same frame range as it was rendered from).

Added Custom Start Frame flag to movie clip (could be accessed from Footage panel in
clip editor) and Start Frame which means number of frame from sequence which would
be displayed at scene frame #1.

For example if you've got clip pointing to file render_00100.png and Start Frame of 100
this file would be displayed at scene frame #1, if Start Frame is 1 then this image
would be displayed at scene frame #100,

7 years agoCycles / OSL:
Thomas Dinges [Tue, 5 Jun 2012 15:40:02 +0000 (15:40 +0000)]
Cycles / OSL:
* Fixes for r46114, object_fetch_transform missed time argument.
* Syntax fixes for Checker texture.

7 years agoshrink/fatten now works with proportional editmode on curves.
Campbell Barton [Tue, 5 Jun 2012 14:19:13 +0000 (14:19 +0000)]
shrink/fatten now works with proportional editmode on curves.

7 years agoshrink/fatten transform for mask - adjusts the feather weight.
Campbell Barton [Tue, 5 Jun 2012 14:13:14 +0000 (14:13 +0000)]
shrink/fatten transform for mask - adjusts the feather weight.
works in proportional editmode too.

7 years agomask point slide now accounts for scaled bezier weights,
Campbell Barton [Tue, 5 Jun 2012 12:51:11 +0000 (12:51 +0000)]
mask point slide now accounts for scaled bezier weights,

7 years agochange how the weight for feather points are calculated with masking, instead of...
Campbell Barton [Tue, 5 Jun 2012 12:22:02 +0000 (12:22 +0000)]
change how the weight for feather points are calculated with masking, instead of using absolute weights, multiply by by the weights of the adjacent beziers.

without this - there was no way to animate the overall feather influence of the feather.

will update tools to account for this next.

7 years agocode cleanup: mask feather weight - add in checks for u==1 or u==0
Campbell Barton [Tue, 5 Jun 2012 12:09:24 +0000 (12:09 +0000)]
code cleanup: mask feather weight - add in checks for u==1 or u==0

7 years agocode cleanup: helper functions for masking.
Campbell Barton [Tue, 5 Jun 2012 11:46:55 +0000 (11:46 +0000)]
code cleanup: helper functions for masking.

7 years agocode cleanup: var names in mask code
Campbell Barton [Tue, 5 Jun 2012 11:28:54 +0000 (11:28 +0000)]
code cleanup: var names in mask code

7 years agoFix #31713: Mask editor: RMB drag crashes Blender
Sergey Sharybin [Tue, 5 Jun 2012 11:28:15 +0000 (11:28 +0000)]
Fix #31713: Mask editor: RMB drag crashes Blender

Simple missed NULL check in TransData creation.

7 years agoFix #31702: Drag and Drop parenting crashes Blender
Sergey Sharybin [Tue, 5 Jun 2012 09:57:19 +0000 (09:57 +0000)]
Fix #31702: Drag and Drop parenting crashes Blender

Crash was caused by recent changes in parent drop operator which were
aimed to prevent parenting objects between different scenes (which probably
makes sense).

The problem was how it was checked if objects belongs to the same scene --
outliner tree with type ID_SCE was used for this which works pretty nice
for All Scenes outliner view. But in other view modes there is no scene
element in outliner tree which lead to some NULL pointer dereferences.

Currently resolved this by assuming that if there's no Scene parent element
in outliner tree parent and child belongs to the same scene which is active
scene. This is truth for current view modes of outliner but if it'll be
changed in the future this assumption shall be updated and re-implemented
with some smarter checks of which scene object from outliner belongs to.

7 years agomask editing
Campbell Barton [Tue, 5 Jun 2012 09:37:44 +0000 (09:37 +0000)]
mask editing
- clear feather weights (alt+s)
- fix for glitch where placing the feather would jitter.

7 years agoCycles / OSL:
Thomas Dinges [Tue, 5 Jun 2012 09:29:47 +0000 (09:29 +0000)]
Cycles / OSL:
* Remove oslexec_pvt.h header and some typo fixes.
* This file needs deeper updates for changes done in OSL 0.6.0, see

7 years agoFix #31593: Every time I switch between edit and object mode, it crashes
Sergey Sharybin [Tue, 5 Jun 2012 08:41:53 +0000 (08:41 +0000)]
Fix #31593: Every time I switch between edit and object mode, it crashes

Crash was caused by incorrect restoring OpenGL context due to some
weird bit operations used to indicate whether stuff like color arrays
is initialized resulting in some unpredictable results on different
platforms and drivers.

7 years agoFix #31706, Crash loading old file in trunk build.
Lukas Toenne [Tue, 5 Jun 2012 08:06:15 +0000 (08:06 +0000)]
Fix #31706, Crash loading old file in trunk build.

7 years agomask switch direction now swaps handle direction too
Campbell Barton [Tue, 5 Jun 2012 07:01:43 +0000 (07:01 +0000)]
mask switch direction now swaps handle direction too

7 years agofix for possible uninitialized pointer use in mask rasterize and remove some dead...
Campbell Barton [Tue, 5 Jun 2012 06:54:18 +0000 (06:54 +0000)]
fix for possible uninitialized pointer use in mask rasterize and remove some dead code.

7 years agocorrection to own change with cmake, oiio linking.
Campbell Barton [Tue, 5 Jun 2012 06:18:31 +0000 (06:18 +0000)]
correction to own change with cmake, oiio linking.

7 years agoFix remesh output changing when input is moved relative to origin.
Nicholas Bishop [Mon, 4 Jun 2012 23:53:59 +0000 (23:53 +0000)]
Fix remesh output changing when input is moved relative to origin.

Fixes bug [#31626] Remesh modifier generates different results
depending on object origin position

Was incorrectly initializing bounding box min/max to zero, now uses

7 years agoFix NULL free warning in multires.
Nicholas Bishop [Mon, 4 Jun 2012 23:14:57 +0000 (23:14 +0000)]
Fix NULL free warning in multires.

7 years agostyle cleanup
Campbell Barton [Mon, 4 Jun 2012 22:44:58 +0000 (22:44 +0000)]
style cleanup

7 years agoonly use tiff/exr when enabled.
Campbell Barton [Mon, 4 Jun 2012 22:34:28 +0000 (22:34 +0000)]
only use tiff/exr when enabled.

7 years agoFix [#31544]: iTaSC assertion when creating armature with no joint. This degenerated...
Benoit Bolsee [Mon, 4 Jun 2012 22:29:17 +0000 (22:29 +0000)]
Fix [#31544]: iTaSC assertion when creating armature with no joint. This degenerated case can be obtained by having a single bone in the IK chain and locking all 3 axis. This case was causing an assert in the KDL library. The bug is fixed by simply not creating the IK scene in this case.

7 years agoCycles / OSL:
Thomas Dinges [Mon, 4 Jun 2012 20:50:59 +0000 (20:50 +0000)]
Cycles / OSL:
* Fixes for changes in the OSL register_closure() API.

7 years agostyle cleanup
Campbell Barton [Mon, 4 Jun 2012 20:11:09 +0000 (20:11 +0000)]
style cleanup

7 years agoCycles / OSL:
Thomas Dinges [Mon, 4 Jun 2012 19:53:50 +0000 (19:53 +0000)]
Cycles / OSL:
* Compile fixes, missing ShaderClosure argument for eval functions. (r40163)

7 years agoA few UI messages fixes.
Bastien Montagne [Mon, 4 Jun 2012 19:43:12 +0000 (19:43 +0000)]
A few UI messages fixes.

7 years agoCycles: support for image sequences in image/environment texture node.
Brecht Van Lommel [Mon, 4 Jun 2012 19:38:33 +0000 (19:38 +0000)]
Cycles: support for image sequences in image/environment texture node.

7 years agoCycles: fix spot lamp emitting light in two directions.
Brecht Van Lommel [Mon, 4 Jun 2012 19:38:10 +0000 (19:38 +0000)]
Cycles: fix spot lamp emitting light in two directions.

7 years agoCycles / OSL:
Thomas Dinges [Mon, 4 Jun 2012 19:00:13 +0000 (19:00 +0000)]
Cycles / OSL:
* __OSL__ is now defined for CPU when building with OSL.

* First batch of compile fixes, remove some unneeded std namespace declarations and added missing includes.

7 years ago * modified vieweroperation to not calculate based on the DO_NODE_OUTPUT
Jeroen Bakker [Mon, 4 Jun 2012 18:07:29 +0000 (18:07 +0000)]
 * modified vieweroperation to not calculate based on the DO_NODE_OUTPUT
flag of the editorbNode.

7 years agoabbreviate mask-editing to mask-edit
Campbell Barton [Mon, 4 Jun 2012 17:30:54 +0000 (17:30 +0000)]
abbreviate mask-editing to mask-edit

7 years agoinclude cleanup, also raskter wasn't building on osx
Campbell Barton [Mon, 4 Jun 2012 17:27:22 +0000 (17:27 +0000)]
include cleanup, also raskter wasn't building on osx

7 years agoCycles: spot lamp support.
Brecht Van Lommel [Mon, 4 Jun 2012 17:17:10 +0000 (17:17 +0000)]
Cycles: spot lamp support.

7 years agopicky change - rename keymap since there is no 'mask editor'
Campbell Barton [Mon, 4 Jun 2012 17:13:38 +0000 (17:13 +0000)]
picky change - rename keymap since there is no 'mask editor'

7 years agomask mode for clip editor developed by Sergey Sharybin, Pete Larabell and myself.
Campbell Barton [Mon, 4 Jun 2012 16:42:58 +0000 (16:42 +0000)]
mask mode for clip editor developed by Sergey Sharybin, Pete Larabell and myself.


note - mask editing tools need continued development, feather option is not working 100%

7 years agoCycles / OSL:
Thomas Dinges [Mon, 4 Jun 2012 16:30:15 +0000 (16:30 +0000)]
Cycles / OSL:
* Compile option to build with OSL, disabled by default.

7 years agoAntoher fix for library linking issues, must change all entries in the libmap,
Brecht Van Lommel [Mon, 4 Jun 2012 16:10:20 +0000 (16:10 +0000)]
Antoher fix for library linking issues, must change all entries in the libmap,
not just the first one.

7 years agoCycles: show frame number in render info.
Brecht Van Lommel [Mon, 4 Jun 2012 16:09:51 +0000 (16:09 +0000)]
Cycles: show frame number in render info.

7 years agoremaining mask files from tomato. these wont get svn history carried over.
Campbell Barton [Mon, 4 Jun 2012 15:49:58 +0000 (15:49 +0000)]
remaining mask files from tomato. these wont get svn history carried over.

7 years agomask header from tomato/
Campbell Barton [Mon, 4 Jun 2012 15:45:04 +0000 (15:45 +0000)]
mask header from tomato/

7 years agocopy mask file from tomato branch, sorry dont know how to do multiple of these at...
Campbell Barton [Mon, 4 Jun 2012 15:44:42 +0000 (15:44 +0000)]
copy mask file from tomato branch, sorry dont know how to do multiple of these at a time...

7 years agoraskter rasterizer by Pete Larabell, from tomato branch
Campbell Barton [Mon, 4 Jun 2012 15:38:55 +0000 (15:38 +0000)]
raskter rasterizer by Pete Larabell, from tomato branch

7 years agomask merge (initial copy of editor files)
Campbell Barton [Mon, 4 Jun 2012 15:36:16 +0000 (15:36 +0000)]
mask merge (initial copy of editor files)

7 years agoFix #31559: second 3D viewport does not update when paiting vertex/weight
Sergey Sharybin [Mon, 4 Jun 2012 15:13:36 +0000 (15:13 +0000)]
Fix #31559: second 3D viewport does not update when paiting vertex/weight

Added notification at the end of weight/vertex paint stroke so all opened
viewports would be updated (same was already done for sculpt mode).

7 years agoFix #31697: Blender crashes when using several viewer nodes in tile
Sergey Sharybin [Mon, 4 Jun 2012 13:19:17 +0000 (13:19 +0000)]
Fix #31697: Blender crashes when using several viewer nodes in tile

Issue was caused by fact that viewer node might re-size ImBuf used for
viewer node result, so if several viewer nodes are running for inputs
with different resolutions it'll result in a crash,

Now copied behavior or pre-tile compositor -- execute viewer node which
has NODE_DO_OUTPUT flag, so no several nodes would be calculated at once.

Should be pretty ok because calculation of several viewers doesn't actually
make sense because there's only one buffer they might use and it's getting
re-calculated when changing active viewer node.

7 years agoRemoved some old cruft - commented out select_actionchannel_by_name() lines
Joshua Leung [Mon, 4 Jun 2012 13:07:41 +0000 (13:07 +0000)]
Removed some old cruft - commented out select_actionchannel_by_name() lines

7 years agoAttempted fix for borderselect crash that Jeremy (@Mango) is getting on a
Joshua Leung [Mon, 4 Jun 2012 12:29:37 +0000 (12:29 +0000)]
Attempted fix for borderselect crash that Jeremy (@Mango) is getting on a
particular file in Pose Mode

7 years agoFix #31172: applying boolean removes uv maps
Sergey Sharybin [Mon, 4 Jun 2012 12:10:38 +0000 (12:10 +0000)]
Fix #31172: applying boolean removes uv maps

The issue was caused by CDDM_tessfaces_to_faces not dealing with CD layers.

There was already function BKE_mesh_convert_mfaces_to_mpolys which converted
mfaces to mpolys with converting all CD layers. Made it a bit more general so
it might work with given arrays of faces/polys and re-used it from CDDM module.

Checked with UV and sculpt data from Blender 2.61 and it loaded nice, so
hopefully there's no regressions in loading older files.

7 years agoFix for recent library linking fix, this triggered another issue that was
Brecht Van Lommel [Mon, 4 Jun 2012 12:01:49 +0000 (12:01 +0000)]
Fix for recent library linking fix, this triggered another issue that was
hidden by libraries getting loaded twice, hopefully it's working ok now,
tested with various .blend files.

7 years agoFix for relative blur size for non-fast gaussian blur methods.
Sergey Sharybin [Mon, 4 Jun 2012 10:51:00 +0000 (10:51 +0000)]
Fix for relative blur size for non-fast gaussian blur methods.

Call base blur's initExecution from GaussianBokehBlurOperation and GaussianXBlurOperation

7 years agoFix #31352: Image sequence texture - can't load image after changing image source
Sergey Sharybin [Mon, 4 Jun 2012 10:41:18 +0000 (10:41 +0000)]
Fix #31352: Image sequence texture - can't load image after changing image source

Added new flag to Imageuser which indicates whether user frame calculation is needed.
This flag is getting set in BKE_image_signal and handled in actual image usage areas
where both image user and current frame is known.

7 years agoCorrection to recent changes in paint module -- missed NULL check for active object
Sergey Sharybin [Mon, 4 Jun 2012 07:44:57 +0000 (07:44 +0000)]
Correction to recent changes in paint module -- missed NULL check for active object

7 years agoTag unused variables
Sergey Sharybin [Mon, 4 Jun 2012 07:43:36 +0000 (07:43 +0000)]
Tag unused variables

7 years agoFix #31093: Brush's "Paint curve presets" in Image paint in UV\Image Editor doesn...
Sergey Sharybin [Mon, 4 Jun 2012 07:29:45 +0000 (07:29 +0000)]
Fix #31093: Brush's "Paint curve presets" in Image paint in UV\Image Editor doesn't work if weight paint mode is active

Some operators like curve presets, color sample and some more were using object's
mode to distinguish in which mode user is currently painting. Such approach fails
in cases when there's paint mode active in 3D viewport and Image Editor.

Changed logic here to use some context's state like active space which helps
distinguishing current paint mode more accurate.

Ported all areas which uses paint_get_active() to new paint_get_active_from_context().
There're still some calls to paint_get_active(), but that shouldn't be harmful due
to that places indeed have object's mode as priority when getting paint mode.

7 years agopatch [#31644] Py Console: Indent and unindent independent of cursor (aligned) [Tab...
Campbell Barton [Mon, 4 Jun 2012 07:24:19 +0000 (07:24 +0000)]
patch [#31644] Py Console: Indent and unindent independent of cursor (aligned) [Tab / Shift + Tab / Ctrl + Tab]
from Sebastian Nell (codemanx)

7 years agoTheme Colors for NLA Strips and DopeSheet Summary
Joshua Leung [Mon, 4 Jun 2012 01:17:37 +0000 (01:17 +0000)]
Theme Colors for NLA Strips and DopeSheet Summary

* NLA Strip colors are now themable
* Changed the "Active Action"/"Summary" colors to be a bit more muted. The new
colors are now closer to those for keyframes, though they are still different
enough to be clearly distinguishable.
* Removed some colors wihch don't seem to be used (from NLA theme colors)
* Added function to get theme colors + alpha as floats

7 years agoCycles / OSL:
Thomas Dinges [Sun, 3 Jun 2012 18:04:52 +0000 (18:04 +0000)]
Cycles / OSL:
* Update the Oren Nayar shader to incorporate updates / fixes from r41968 and r44330.

7 years agofix [#31674] BMesh Corrupt Stack around BMVert array "verts" - ONLY ON DEBUG VERSION
Campbell Barton [Sun, 3 Jun 2012 17:16:20 +0000 (17:16 +0000)]
fix [#31674] BMesh Corrupt Stack around BMVert array "verts" - ONLY ON DEBUG VERSION

7 years agorename context menu 'Python Documentation' to 'Online Python Reference'
Campbell Barton [Sun, 3 Jun 2012 16:47:19 +0000 (16:47 +0000)]
rename context menu 'Python Documentation' to 'Online Python Reference'

7 years agominor optimizations to compositor, avoid indirections when operating array members...
Campbell Barton [Sun, 3 Jun 2012 16:23:47 +0000 (16:23 +0000)]
minor optimizations to compositor, avoid indirections when operating array members multiple times

7 years agoBugfix for [#31692] Blur Node (Flat) X size affects Y size, Y size does nothing
Thomas Dinges [Sun, 3 Jun 2012 15:44:11 +0000 (15:44 +0000)]
Bugfix for [#31692] Blur Node (Flat) X size affects Y size, Y size does nothing

* GaussianYBlurOperation accessed sizex instead of sizey.

7 years agoMinor UI message fix
Bastien Montagne [Sun, 3 Jun 2012 13:38:27 +0000 (13:38 +0000)]
Minor UI message fix

7 years agoFix bug [#31588]: iTaSC does not handle armature scaling correctly. iTaSC solver...
Benoit Bolsee [Sun, 3 Jun 2012 12:06:42 +0000 (12:06 +0000)]
Fix bug [#31588]: iTaSC does not handle armature scaling correctly. iTaSC solver operates in world reference, therefore armature scale is used to build the ik scene. But the scaling was not taken out when applying the pose at the end of the simulation.

7 years agoFix Eigen3 unaligned array in iTaSC. The unaligned array cause crash when SSE2 is...
Benoit Bolsee [Sun, 3 Jun 2012 12:00:41 +0000 (12:00 +0000)]
Fix Eigen3 unaligned array in iTaSC. The unaligned array cause crash when SSE2 is enabled. The source of unaligned array were: matrix passing by value in std::make_pair() and offset of matrix member in structure.

7 years agocode cleanup: replace some non utf8 chars
Campbell Barton [Sun, 3 Jun 2012 11:16:13 +0000 (11:16 +0000)]
code cleanup: replace some non utf8 chars

7 years agouse fnmatch for reference globing rather then regex - regex overly complicates simple...
Campbell Barton [Sun, 3 Jun 2012 11:01:18 +0000 (11:01 +0000)]
use fnmatch for reference globing rather then regex - regex overly complicates simple cases and wasnt being taken advantage of.

7 years agoquiet warnings
Campbell Barton [Sun, 3 Jun 2012 10:33:28 +0000 (10:33 +0000)]
quiet warnings

7 years agoBugfix: Bone Groups now displayed with correct icon in Outliner
Joshua Leung [Sun, 3 Jun 2012 10:23:48 +0000 (10:23 +0000)]
Bugfix: Bone Groups now displayed with correct icon in Outliner

7 years agoCycles: different fix for object info random range issue.
Brecht Van Lommel [Sun, 3 Jun 2012 09:50:17 +0000 (09:50 +0000)]
Cycles: different fix for object info random range issue.

7 years agoPart-Bugfix, Part-Feature Completion: 'Armature' Option for Mask Modifier
Joshua Leung [Sun, 3 Jun 2012 07:49:54 +0000 (07:49 +0000)]
Part-Bugfix, Part-Feature Completion: 'Armature' Option for Mask Modifier
finally works

This commit finally hooks up the Mask Modifier's "Armature" option with the
relevant depsgraph updates on bone selection. Hence, this feature finally works
as it was originally intended - that is, bone selections can be used to control
which parts of the mesh that the mask modifier is applied to are displayed,
giving riggers more freedom to experiment with rigs that don't necessarily
feature overbearing/cluttering widgets.

Regarding the implementation ("has_viz_deps" flag):
This feature is just the "tip of the iceberg" of a number of related set of
rigging/visual animation tools I've had in mind for a while now (dating back to
the introduction of this modifier). Key considerations
- Not all rigs will use this, so we don't want an extra (depsgraph-flush +
search) recalc cost for those that don't use this.
- There are some planned features which will also use this

7 years agoMask Modifier - Bugfix and dead-code cleanup
Joshua Leung [Sun, 3 Jun 2012 01:05:20 +0000 (01:05 +0000)]
Mask Modifier - Bugfix and dead-code cleanup

'Armature' mode for the Mask Modifier was not working at all anymore even when
the selection <-> depsgraph recalc issue was patched to work (this latter fix is
coming in another commit). It appears that this probably happened during one or
more of the refactors which may have taken place around here over the years
since I first introduced it.

This commit does two things:
* Removed the unused/redundant "vgroupHash"
* Fixed the incorrect assumption used for determining if the vertex actually
belonged to a vgroup corresponding to a selected bone.

7 years agoCycles / OSL Textures:
Thomas Dinges [Sun, 3 Jun 2012 00:02:24 +0000 (00:02 +0000)]
Cycles / OSL Textures:
* Ported my checker texture to OSL. :)

7 years agoCycles / OSL Textures:
Thomas Dinges [Sat, 2 Jun 2012 22:57:26 +0000 (22:57 +0000)]
Cycles / OSL Textures:
* More fixes, replaced all remaining size variables with scale ones.
* Remove nonzero check in OSL textures, not needed anymore, was there to prevent division by zero for the size variable.

7 years agoCompile fix - "spf" was undefined
Joshua Leung [Sat, 2 Jun 2012 22:47:33 +0000 (22:47 +0000)]
Compile fix - "spf" was undefined

7 years agoCloth collisions: Revert 47335.
Daniel Genrich [Sat, 2 Jun 2012 22:01:04 +0000 (22:01 +0000)]
Cloth collisions: Revert 47335.

7 years agoCycles / OSL:
Thomas Dinges [Sat, 2 Jun 2012 21:34:25 +0000 (21:34 +0000)]
Cycles / OSL:
* More fixes for r41599, removed clouds and distorted noise textures and ported the Noise texture to OSL.
ToDo: Color output is still commented, needs a closer look.

* Some more fixes (comments, uninitialized variables)

7 years agoCycles / Object Info Node:
Thomas Dinges [Sat, 2 Jun 2012 21:04:16 +0000 (21:04 +0000)]
Cycles / Object Info Node:
* The random output had a range from -0.5 to 0.5, added an offset of 0.5 to bring it into the 0.0 / 1.0 range, which is in alignment with Shader Node Guidelines.
Patch by Agustin Benavidez, thanks!

7 years agoFixing most (if not all) remaining issues with nodes UI i18n.
Bastien Montagne [Sat, 2 Jun 2012 19:58:12 +0000 (19:58 +0000)]
Fixing most (if not all) remaining issues with nodes UI i18n.

Note that I had to script-tag all sokets' names, as they are currently completely unknown from bpy.types (and hence unreachable for our po generating scripts).

7 years agoCycles / OSL:
Thomas Dinges [Sat, 2 Jun 2012 19:25:12 +0000 (19:25 +0000)]
Cycles / OSL:
* Forgot to rename Gradient OSL file.

7 years agoCycles / OSL:
Thomas Dinges [Sat, 2 Jun 2012 19:14:14 +0000 (19:14 +0000)]
Cycles / OSL:
* More OSL fixes for r41599, removed marble and wood texture and ported Wave Texture to OSL.

7 years agoCycles / Wave Texture:
Thomas Dinges [Sat, 2 Jun 2012 17:44:28 +0000 (17:44 +0000)]
Cycles / Wave Texture:
* Found an issue in the shader code, it used the x coordinate twice.

7 years agoCycles / OSL Fixes:
Thomas Dinges [Sat, 2 Jun 2012 17:10:59 +0000 (17:10 +0000)]
Cycles / OSL Fixes:
* OSL textures were never updated after the refactor in november 2011.

* Remove stucci texture and renamed blend to gradient.

7 years agoexperenental manual linking from the UI. realize this is an issue which is not agreed...
Campbell Barton [Fri, 1 Jun 2012 20:38:40 +0000 (20:38 +0000)]
experenental manual linking from the UI. realize this is an issue which is not agreed on so probably this will be disabled for release.

the data is stored here so more dev can commit:


7 years agocode cleanup: rename bool --> do_tint, confused qtcreator
Campbell Barton [Fri, 1 Jun 2012 20:28:53 +0000 (20:28 +0000)]
code cleanup: rename bool --> do_tint, confused qtcreator

7 years agoCommenting out unused var, prevents my debug Blender to build... ;)
Bastien Montagne [Fri, 1 Jun 2012 17:47:05 +0000 (17:47 +0000)]
Commenting out unused var, prevents my debug Blender to build... ;)

7 years agoMake "Match Movie Length" into an operator (must be called inside a context having...
Bastien Montagne [Fri, 1 Jun 2012 17:28:09 +0000 (17:28 +0000)]
Make "Match Movie Length" into an operator (must be called inside a context having either a "texture" Tex, an "area" SpaceImage, or both "edit_image" Image and "edit_image_user" ImageUser).

Thanks to Campbell who pointed me to uiLayoutSetContextPointer() func!

7 years agoFollowup fix Bugfix [#31629]: Cloth simulation collisions used still too high repulsions.
Daniel Genrich [Fri, 1 Jun 2012 16:50:12 +0000 (16:50 +0000)]
Followup fix Bugfix [#31629]: Cloth simulation collisions used still too high repulsions.

7 years agoSmoke Bugfix: Velocity of moving objects was too high after scaling patch.
Daniel Genrich [Fri, 1 Jun 2012 15:49:39 +0000 (15:49 +0000)]
Smoke Bugfix: Velocity of moving objects was too high after scaling patch.

Spoted by MiikaH.

7 years agoFix #31673, File Output node forgets about socket names. Was checking for wrong file...
Lukas Toenne [Fri, 1 Jun 2012 15:16:58 +0000 (15:16 +0000)]
Fix #31673, File Output node forgets about socket names. Was checking for wrong file version for do_versions.

7 years agofix for crash when loading a file while rendering.
Campbell Barton [Fri, 1 Jun 2012 15:05:30 +0000 (15:05 +0000)]
fix for crash when loading a file while rendering.

7 years agoTweaks for Pasting Keyframes in DopeSheet/Graph Editors
Joshua Leung [Fri, 1 Jun 2012 15:00:28 +0000 (15:00 +0000)]
Tweaks for Pasting Keyframes in DopeSheet/Graph Editors

In response to [#31670], I've reviewed the way that the Paste Keyframes tool for
the DopeSheet and Graph Editors works. Previously, it required you to always
select the F-Curves to paste the keyframes into before allowing you to paste
keyframes. This was because it is quite difficult to infer which ID-block's set
of curves is intended if more than one ID-block has similar curves (e.g. a scene
with two materials, and both have their diffuse color animated). The underlying
assumption and intention of the feature here was that the copy+paste were only
being used by animators to copy animation between similar curves, to transfer
and offset animation across block boundaries.

However, it turns out that many people were by far more familiar with the
simpler copy/paste paradigm from everywhere else (i.e. instead of trying to use
duplicate to copy keyframes around within their respective F-Curves).
Furthermore, in most cases there is only going to be a single character being
animated at a time (vs multiple), which means that most of the time the matching
problem is much simpler.

Hence, the Paste now works as follows:
- If there are selected F-Curves, we limit the paste-matching to only consider
those in the selected F-Curves. This makes it possible to still explicitly
specify where to paste.
- In the more general case (no prior selections), pasting will try to match
anything relevant it finds.

- Check on whether the strictest matching level needs adjustments to limit the
number of false positives
- Testing and feedback of the new behaviour needed <--- ANIMATORS! PLEASE TEST