Decimate used `is_quad_convex_v3` to check if faces should be converted
to quads, adjust the epsilon since small faces were being skipped.
Ref: !132342
The "Blender" menu contains seven items in "System / "Redraw Timer"
that are for troubleshooting and testing. Six of the items take far
less than a second, but one takes a long time. "Animation Play" times
how long it takes to play through your current animation ten times. But
there is no way to guess this and some users have run it accidentally
by finding it in menu search. It gives no feedback at all, and there
are no hints on how long it will take. Once it plays through your
animation once you might guess that it runs forever. This PR shows
the test name and where it is in its ten runs. On platforms that
support it this also shows an app progress bar (on the taskbar icon for
Windows).
Pull Request: https://projects.blender.org/blender/blender/pulls/132648
In a recent version of Blender, Cycles and EEVEE gained support for
true sphere lights.
This commit adds tests for these sphere lights inside of volumes.
Ref: blender/blender#123012
Ref: blender/blender-test-data!32
- Using a main function allows the scripts to be imported without
executing logic.
- Declaring `__all__` lets tools such as "vulture" detect unused code.
Accessing the internal extensions command would warn that the processes
stdout wasn't closed on completion. While it's harmless at the moment,
ensure it's closed using a context manager.
Rendering a scene with py-drivers would crash after dropping an
extension onto Blender's window then canceling before the update
completed.
This was caused by context.temp_override pushing arguments into the
context but failing to pop them in the case of an error - which would
happen when the popup was closed while the update ran.
ContextTempOverride objects were not freeing the reference to their
arguments when the context managers __exit__ didn't run.
This would happen if __enter__ failed with an unsupported configuration
or if object was created and not used.
Resolve by making ContextTempOverride a container type with GC support.
This also allows the ContextTempOverride to be reused since it's
arguments are now cleared when it's deallocated instead of __exit__.
When hovering over the "Editor Type" button it always shows Shift-F1 as
the shortcut no matter what the current Editor Type is. We already have
specific code for "SCREEN_OT_space_type_set_or_cycle" for the items on
the popup menu, but it doesn't handle the case of hovering the menu
itself. This adds just adds a line to handle this.
Pull Request: https://projects.blender.org/blender/blender/pulls/132612
- Pass large std::function by const ref
- Shift the mesh material assignment helpers inside main namespace
- Use MaterialFaceGroups type alias in more places
- Remove unused bmain argument from import_mesh_skel_bindings
Pull Request: https://projects.blender.org/blender/blender/pulls/132586
Nodes should have the same name in the Add Node menu as their actual
name, but the Spiral node is called "Curve Spiral". Since there is no
ambiguity with another primitive node called Spiral, it can be renamed
to that.
This also fixes an issue with translation, where the node could not be
translated since its name was never extracted to the translation
files.
Pull Request: https://projects.blender.org/blender/blender/pulls/122289
Some recently added versioning code incorrectly assumed that the
noise texture node would always have its storage struct allocated.
Just create the storage struct in versioning where necessary, to
avoid relying on defaults not changing in the future when the
struct is created as the node tree is initialized.
Pull Request: https://projects.blender.org/blender/blender/pulls/132555
The issue was that the propagation of referenced anonymous attributes treated
geometry outputs of the foreach zone as "normal". That means that every
anonymous attributes referenced by the input socket would also be referenced by
the output socket.
However, just like in the repeat zone, this so called "propagate relation" needs
some special behavior, because anonymous attributes references created inside a
zone have to remain inside that zone. Instead, the output node creates a new
anonymous attribute reference that is used outside of the zone.
Pull Request: https://projects.blender.org/blender/blender/pulls/132560
Symptoms were broken `keyframe_type` & `options` (e.g. falsely assuming
'NEEDED' option when this was not provided at all).
The parameter order was flipped (leading to false interpretations for
the enum (flag) values).
For historic reference: the `keyframe_type` was introduced in
570fbba3f3 (but in a way that would have broken the existing API),
that was tried to be corrected in 46aaa53998 (but that commit only
flipped the order in the parameter definitions, not in the actual
function `rna_FKeyframe_points_insert`).
So to resolve, correct the order in the function declaration.
Pull Request: https://projects.blender.org/blender/blender/pulls/132595
We were already flipping the minima & maxima the right way, but the
checkboxes for which axis gets limited to a min/max were not updated. So
when a max X actually gets a min X, those checkboxes should be set
accordingly (now done).
Pull Request: https://projects.blender.org/blender/blender/pulls/132265
The issue was that the range for the time offset operator
was so large, that the 0.1 increment seemingly had no effect.
The solution is to add functionality to the slider to specify the increment step.
Since the time offset operator deals in frames it made sense to snap to whole numbers.
Also fixes#132187
For that the increment step has been chosen to allow for 10 steps in the slider range.
Pull Request: https://projects.blender.org/blender/blender/pulls/132554
Only draw relationship lines between a constrained object/bone and its
target when there is actually a valid target. Previously Blender would
always draw a line, which would go to the world origin when the
constraint has no target. This was visually rather noisy and potentially
even misleading when there is actually an object at the origin.
Fixes#131477
Pull Request: https://projects.blender.org/blender/blender/pulls/132592
With the recent sculpt refactor project prior to 4.3, the Sculpt undo
system was changed such that an individual `SculptUndoStep` would store
a single type on the `StepData` struct instead of storing individual
nodes that each had their own type.
This works for most operators supported by Sculpt Mode, however, the
`Apply Base` multires operator was dependent on needing to store both
Geometry and Position data.
To restore old functionality, this commit removes the need for storing
this Position data by instead forcing the object space Sculpt Mode
multires data to be flushed to the tangent space MDisp data prior to
saving the current geometry state so that it can be restored correctly.
This has the following benefits:
* We can continue to assume that Geometry steps represent full-mesh
changes and should require no further specialized processing.
* This better aligns with the future state of trying to flush this
multires data on a per-stroke basis instead of needing to wait unti
lthe user either quits or exits sculpt mode.
Pull Request: https://projects.blender.org/blender/blender/pulls/132569
Update the OptiX denoiser model type to the newer and more frequently
updated `OPTIX_DENOISER_MODEL_KIND_AOV` denoiser.
The output of the new denoiser type is typically higher quality.
And the old denoiser is no longer being updated.
Original change was proposed by Patrick Mours from NVIDIA.
Pull Request: https://projects.blender.org/blender/blender/pulls/132542
Simplify the `${CONSTRAINT}_get_tarmat` functions, mostly by reducing
nesting via early returns.
Code like this:
```cpp
void blabla_get_tarmat(...) {
if (VALID_CONS_TARGET(ct)) {
// ... do useful stuff.
}
else if (ct) {
unit_m4(ct->matrix);
}
}
```
now looks like:
```cpp
void blabla_get_tarmat(...) {
if (!VALID_CONS_TARGET(ct)) {
unit_ct_matrix_nullsafe(ct);
return;
}
// ... do useful stuff.
}
```
This is mostly done as preparation for an upcoming functional change.
No functional changes.
Pull Request: https://projects.blender.org/blender/blender/pulls/132591
While in theory it would be good to have calls to super classes'
`__del__()` destructors in subclasses, matching the ones to
`__init__()`, several limitations of current CPython implementation do
not make it a practical requirement.
So remove `__del__` from examples, and add a note summarizing the
current problems with using it (aka `tp_finalize` in C++ code).
Also see !132476 for some discussion about that topic.
Falloff/cavity mask checkboxes seems disabled themselves when unticked.
The layout was split so that only the options below the checkbox are inactive.
UI looks and acts the same when checkboxes are ticked.
Pull Request: https://projects.blender.org/blender/blender/pulls/132507
