Age | Commit message (Collapse) | Author |
|
|
|
Fix for bug [#31604] "Skin Modifier - Mark Root Bug?"
Essentially the same as multires customdata deletion.
|
|
Issue was caused by generated images had got no profile set (IB_PROFILE_NONE)
which confused OpenEXR saver (which makes sRGB conversion for profiles which
are not linear).
Actually this is much deeper issue which would require the whole color pipeline
workflow (which would happen at some point when currently urgent stuff is resolved),
but having correct profile set for generated images would still be helpful.
|
|
|
|
length in frames (or 1 for still images).
|
|
|
|
|
|
|
|
from Jason Wilkins (jwilkins)
only applied some parts:
* const correctness
* moved a variable into a move local scope so it is also inside a #if/endif and does not end up conditionally unused
|
|
Issue was introduced by yesterday's commit 47021 and caused by some handler's
which is getting called from ui_handler_popup frees event.
Worked around a bit by storing return value for ui_handler_popup before running
other handlers, but this only means global refactor of even handling order is
getting closer and closer.
|
|
- previous commit was so similar to the patch Shinsuke wrote, better to add him as contributor, this commit will add to generated credits.
- this commit has no functional changes.
|
|
|
|
tools...).
|
|
of linking error
|
|
|
|
|
|
libraries
could get loaded twice in some cases, causing issues with e.g. node groups.
|
|
Do not block EVT_DROP event from popup handler.
Not ideal solution sine it'll be much nicer to cleanup event handling order,
but should be acceptable for now.
|
|
(seen on Windows)
|
|
property.
|
|
just the active window.
|
|
starting from version 2.64. Unless you have a special system setup, this means the
will be in ~/.config/blender rather than ~/.blender.
When the version number is changed to 2.64, the "Copy Previous Settings" operator
in the splash will copy the settings to the new location.
XDG base directory specification:
http://standards.freedesktop.org/basedir-spec/basedir-spec-latest.html
|
|
Now loading image sequence behaves in the same way as sequencer when all images
are selected. In fact, in this case first image from selection would be used
as first frame of image sequence, so this behavior does not reflect possible
gaps in selection.
Old behavior with selecting first file only is still relevant.
|
|
|
|
domain size was < 1.0 and density+shading strength changed depended on domain size.
|
|
side,
found in the Display panel.
Patch by Simon Kirk and Irie Shinsuke, refactored to also work for non-mesh objects
and avoid globals.
|
|
check.
|
|
Detail: Smoke solver and Blender side of smoke now share the same cell length.
First reported by the Sintel artists long ago, again reported by MiikaH.
Part of my Smoke Development Project Phase III.
|
|
|
|
1. Two new boolean options have been added to the operator: "deselect"
and "toggle".
2. The previous behavior of "extend" (toggling the selection) has
been moved to the "toggle" option.
3. "extend" now only extends the selection, it never deselects.
4. "deselect" is pretty self-explanatory: it deselects (i.e. opposite
of extend).
5. The built-in keymap has been changed to use "toggle" where
"extend" was used before for this operator, to maintain the
previous behavior in the default keymap.
In short, this works towards making "extend" and "deselect" fully
consistent across all selection tools (adding to and removing from
selection, respectively), but still preserves the old behavior
as well.
(Patch reviewed by Brecht.)
|
|
Change BKE_object_minmax() to use the same logic as meshes: use the
object bounding box if available, since it will include the modifier
DM output.
|
|
Buffer overflow error in collecting split_face vertices.
|
|
Part of my Smoke Development project.
In conjunction with MiikaH and his gSoC project.
|
|
|
|
|
|
before. It doesn't make much sense to always multiply with alpha, especially
if it's a premul image that already has it multiplied in.
Patch by Irie Shinsuke.
|
|
F-Curves
It is possible to get the old behaviour (handles excluded) by bringing up the
Operator Properties (F6) while in the Graph Editor (this doesn't work elsewhere
due to the context requirements of this stuff).
|
|
|
|
|
|
|
|
* Moved all enums to immediately follow the struct(s) that they are used for/by.
Less jumping around the file to find these
* Reshuffled the order of a few structs/editor groupings. For example, now all
animation editors are grouped together, and most general/property editors are
grouped together.
|
|
* All lists of #defines now replaced with enums
* All flags are now defined using the "(1 << x)" style, which is easier to
read/maintain than the other variations
|
|
This commit restores the group colours support for F-Curves and F-Curve Groups
in the DopeSheet and Graph Editors. Currently the relevant settings for groups
are only exposed via RNA, but a followup commit will add support for
automatically setting these colours. By default, DopeSheet and Graph Editors are
set to display these colours if/when they are available.
This functionality used to be in 2.48, and is a useful mechanism for visually
distinguishing between channels for different controls when animating (if group
colours are used on the rigs too).
|
|
|
|
initializations.
|
|
garbage collection system
* if the image is not loaded (bindcode == 0) load the image
This needs to be called often. If the image is 'cleaned' by Blender, it will not help to load it after.
[ gl_load returns the errors: GL_STACK_OVERFLOW(1283) or GL_STACK_UNDERFLOW (1284) ]
Thanks Campbell for the suggestion on how to handle this (BKE_image_tag_time)
|
|
initialize name from filetype as images do.
|
|
|
|
|
|
source image instead of destination
This lead to wrong proxies generated for openexr float image sequences
|