Replace a linear search through all objects for every single selection
result with a Map from select ID to object base. This makes the runtime
linear instead of quadratic, resulting in a 100x performance improvement
from 6s to 60ms when selecting about half of 60k objects in a scene.
Pull Request: https://projects.blender.org/blender/blender/pulls/137832
Just small tweaks that should have been addressed when the editor
border theme setting was introduced.
* Fix: Match Top Bar and Status Bar headers with editor border.
* Fix: Properties Editor navigation bar and tabs outline tweak for
readability.
* Use light color for editor outline/active.
Pull Request: https://projects.blender.org/blender/blender/pulls/137955
No functional changes.
This patch converts the `ePose_SelectSame_Mode` enum to an enum class
called `SelectRelatedMode`.
This is done in preparation of adding more modes to the operator.
Pull Request: https://projects.blender.org/blender/blender/pulls/137952
In forward path tracing, when we pass volume bounding meshes, we
accumulate `volume_bounds_bounce`. We should match this behaviour in NEE
instead of accumulating `transparent_bounce`.
Pull Request: https://projects.blender.org/blender/blender/pulls/137556
The object was not properly tagged for update when syncing object motion
blur transforms. This caused issue with the camera in volume test after
recent scene update refactoring in 86b67a20d6, but potentially also
other ones.
Pull Request: https://projects.blender.org/blender/blender/pulls/137930
Previously when joining strokes, only point order is reversed as needed
without taking care of bezier handle positions, this can cause bezier
typed strokes to be joined with reversed handles. This fix swaps left/
right handle attributes together with points in `reverse_points_of` call.
Pull Request: https://projects.blender.org/blender/blender/pulls/137533
Blender translates the "Dynamic" setting (`RBO_FLAG_DISABLED`) into a
bullet _kinematic state_ (see `RB_body_set_kinematic_state` /
`CF_KINEMATIC_OBJECT`). This is different from the _activation state_
(done through e.g. `RB_body_set_activation_state`) and also different
from the active/passive type. So we have (at least) three levels to
control this.
And it seems like even though a bullet body is fully intended to be
kinematic (aka animated), it still accumulates forces from the
`RB_body_apply_central_force` call. I have not tracked this down
further, but it _seems_ logical to only apply forces on a body with
the "Dynamic" setting.
Tested this with activation/deactivation (with seems intact), so in my
limited tests I wasnt able to break it
Pull Request: https://projects.blender.org/blender/blender/pulls/137901
The `USE_EXPERIMENTAL_TESTS` variable was not exposed as an option, so
the two tests that use the option to check whether tests should be run
had to be manually enabled by changing `tests/python/CMakeLists.txt`.
This commit renames the variable to `WITH_TESTS_EXPERIMENTAL`, defaults
the option to `OFF`, and marks it as an advanced option.
Pull Request: https://projects.blender.org/blender/blender/pulls/133831
According to the OptiX denoising guide lines, renders should be in the
0..10000 range. As a result, clamping was applied to renders
prior to GPU denoising.
However this lead to a issue where denoising of bright images would
behave differently between CPU OIDN and GPU OIDN as this same clamping
was also applied to the GPU OIDN denoiser.
According to Attila Áfra, a developer working on OIDN, this clamping
isn't neccesary for OIDN. So this commit removes the clamping from
OIDN.
Pull Request: https://projects.blender.org/blender/blender/pulls/137879
Adjust keymap poll function of markers keymap. i.e. when no marker
exists, return false from the keymap so marker operations are not
invoked from key press. Also check whether marker region is visible.
If its is hidden, we can skip the markers keymap.
Pull Request: https://projects.blender.org/blender/blender/pulls/137803
Add the new Manifold solver to the list, and make it the default. Though
it doesn't handle non-manifold sculpt meshes, it makes sense to make it
the default because it's so much faster than the other options, making
the trim tool much more compelling.
Also, instead of joining the object's mesh and the trim mesh as a BMesh,
then running the boolean operation (which requires four conversions
between mesh and BMesh for the non-float solvers), use the same API
as the mesh boolean geometry node. The "join" case is handled
separately now.
Pull Request: https://projects.blender.org/blender/blender/pulls/137915
The object to world transform was applied to the result (which was
meant to be in world space), rather than the inverse. However, the
processing of transforms is more complicated than necessary. Instead
of passing around a separate "target transform" that's meant to be used
inverted after the boolean operation, just make the input transforms
transform the input meshes into the desired transform space of the
output (object-local space for the modifier).
Pull Request: https://projects.blender.org/blender/blender/pulls/137919
Saves a tiny amount of time computing the loose vertices cache
after a boolean operation. Since the boolean output is just created
from faces, there are no loose vertices (there are also no loose
edges because of the call to `mesh_calc_edges`, which itself
tags the loose edge cache).
Object conversion operator used to fail silently when encountered
unsupported object types. Now it will show a warning message instead
so users will have a visual feedback when nothing is happening.
Pull Request: https://projects.blender.org/blender/blender/pulls/134827
pprop pointer created before initialization of op->customdata i.e.
`text_open_init()`. This triggers crash due to accessing member of
nullptr object `pprop->prop`
Pull Request: https://projects.blender.org/blender/blender/pulls/137880
Null check the operator member as this is documented to be null
and is checked elsewhere in the handler logic.
Co-authored-by: Kabinet0 <m3kabin@gmail.com>
Ref !137506
Adding the additional check is more of a "technical" correction,
for the callback to return 0 the sort implementation would have
to compare an element in the array with itself.
Co-authored-by: Zheng-Ze <zheng-ze@noreply.localhost>
Ref !137723
615100acda added support to
`NODE_OT_add_image` to open multiple images, This makes the NW
operator redundant, also `NODE_OT_add_image` can open multiple
images at drag-n-drop. The only difference is that this operator adds
the new image nodes as collapsed nodes.
Pull Request: https://projects.blender.org/blender/blender/pulls/128443
