Many modifiers and other places use `CustomData_copy_data` to copy data
between different meshes. This function assumes that assumes that the
source and destination `CustomData` objects are "compatible" in some way.
Usually modifiers use `CustomData_copy` to create a compatible new
`CustomData` on the new mesh. The issue was that the optimization I added
for anonymous attributes broke this compatibility. It avoided copying some
attributes when they are no longer used.
This lead to attributes being copied incorrectly.
D13083 contains ideas for how this could be fixed more generally.
For now I just removed the optimization.
Differential Revision: https://developer.blender.org/D13083
Function arguments were incorrect.
Noted during @simonthommes live stream.
Reviewed By: JacquesLucke
Differential Revision: https://developer.blender.org/D13110
Issue was that the context used for dropbox handling and polling didn't
match the one used for drawing the dropbox and generating the tooltip
text (which would determine the material slot under the cursor,
requiring context). The mismatch would happen with overlapping regions.
Actually, this patch includes two fixes, each fixing the crash itself:
* Store the context from handling & polling and restore it for drawing.
* Correct the hovered region lookup for drawing to account for overlayed
regions.
Note that to properly set up context for drawing, we should also account
for the operator context, which isn't done here, see
https://developer.blender.org/T92501#1247581.
Size, position and scale of the "two-line" widget (the one to scale a
node horizontally) was not taking interface scale into account. In the
case of the report, it could happen it draws behind an output socket.
before (at 2.0 interface scale)
{F11698493}
after (at 2.0 interface scale)
{F11698501}
Maniphest Tasks: T92791
Differential Revision: https://developer.blender.org/D13088
When reading pixels for virtual passes like diffuse, that sum diffuse direct
and indirect passes, we do not need them to exist with an offset in the render
buffer.
By default, when syncing materials slots between object and its obdata,
the amount of slots in obdata is the reference.
Missing linked obdata is replaced by an empty placeholder that has no
material. In that specific case, if we have a valid object ID, we want
to update the (placeholder) obdata's material count from the object one,
and not the other way around.
Even never-shown RNA properties should have at least a description, as
this is used by API doc generation scripts.
NOTE: this is more of an opportunistic set of changes than a proper
complete fix of that loack of documentation.
* Fix systematic skipping of labels when they are the same as
the identifier (Some cases are valid, like `RGB` or `HSV` e.g.).
* Add instead heuristics checks to skip non-UI properties (non-capitalized,
or same name as identifier and Operator properties, mainly).
* Skip `bl_icon` and `icon` properties.
* Properly search for properties in all parent classes (some cases with
e.g. `Panel` would break due to intermediary utils classes, leading to
those internal UI properties not being skipped as expected).
Related to T43295.
There was a bunch of special handling to support dropping data-blocks onto
string or search-menu buttons, to change the value of these. This refactor
makes that case use the normal drop-box design, where an operator is executed
on drop that gets input properties set by the drop-box. This should also make
it easier to add support for dragging assets into these buttons.
In addition this fixes an issue: Two tooltips were shown when dragging assets
over text buttons. None should be shown, because this isn't supported.
The version patch for 0cf9794c7e was checking and setting a data name
using the macros for translation. These would access the current
preferences which can mismatch the ones currently being version patched.
See discussion in 0cf9794c7e for details.
Don't handle translation in this version patch, which is more of a
"nice-to-have" version patch, no functionality depends on it.
The geometry node evaluator now has access to the entire socket path
from the node that produces a value to the node that uses it. This allows
the evaluator to make decisions about at which points in the path the
value should be converted. Multiple conversions may be necessary under
some circumstances with nested node groups.
Differential Revision: https://developer.blender.org/D13034
This properly checks the order of point domain attributes on each
spline, and avoids the map, which makes the code easier to understand.
The assert is also added to realizing instances and the join node.
Differential Revision: https://developer.blender.org/D13071
Currently the curve to mesh node relies on the order of attributes being
the same on every spline. This is a worthwhile assumption, because it
will allow removing the attribute name storage duplication on every
spline in the future.
However, the join geometry node broke this order, since it just created
new attributes without any regard to the order. To fix this, I added a
"reorder" step after all the merged attributes have been created, the
types have been joined, etc. It should be possible to change this code
so that the attributes are added with the right order in the first
place, but I would like to do that after refactoring spline attribute
storage, and not for 3.0.
Differential Revision: https://developer.blender.org/D13074
We run into float precision issues here, clamp the number of octaves to
one less, which has little to no visual difference. This was empirically
determined to work up to 16 before, but with additional inputs like
roughness only 15 appears to work.
Also adds misisng clamp for the geometry nodes implementation.
Not sure why this bug was only discovered by such an elaborate steps
and why it took so long to be discovered. The root of the issue is
that in the 956c539e59 the typical flow of tag+flush+evaluate was
violated and tagging for visibility change happened after flush.
Since rBb67fe05d4bea2d3c9efbd127e9d9dc3a897e89e6 collections
have a geometry component that depends on all the geometries
inside the collection. Contrary to what I originally thought
`ID_RECALC_COPY_ON_WRITE` does not trigger a collection geometry
update. This makes sense because a collection may change in ways
that do not require a geometry update.
Instead, we have to trigger the geometry update manually now by
passing `ID_RECALC_GEOMETRY` when appropriate.
Currently, colored links overlay only supports standard sockets defined
by Blender. Some add-ons like Animation Nodes defines custom sockets for
everything and hence doesn't get colored sockets.
This patch uses the draw color from the socket type info to draw links
in order to support custom sockets.
Differential Revision: https://developer.blender.org/D13044
Reviewed By: Hans Goudey
The geometry node port of voronoi_smooth_f1 function has a
division by zero when smoothness is set to zero.
Using a safe_divide within the function causes issues
and was noted in the original patch D12725.
Solution in this case is to clamp zero smoothness to FLT_EPSILON.
Reviewed By: JacquesLucke
Maniphest Tasks: T92736
Differential Revision: https://developer.blender.org/D13069
Need to make sure images needed for hair shaders are loaded
before running the shader.
The naming is a bit misleading, but this is an internal API
and we can change it easily. Submitting minimal patch needed
to fix logic in the code to make it safer to review for 3.0.
Differential Revision: https://developer.blender.org/D13067
Calling it with a None argument, or no arguments, or with a property
that is missing UI data for some reason would fail. There is no
particular reason why ensuring those things don't happen is helpful,
so just add null checks for safety.
Differential Revision: https://developer.blender.org/D13024
Some compositor tests (e.g. `compositor_color_test`) broke
because of rB0c3b215e7d5456878b155d13440864f49ad1f230.
The issue was a heap-use-after-free bug caused by a missing
call to `MEM_CacheLimiter_unmanage`.
Operators such as setting the object mode failed after calling
WM_OT_open_mainfile from Python.
Keep the window after loading a file outside the main event loop.
Previously, `ImageTile->ok` and `ImageUser->ok` were used to indicate
whether an image failed to load. There were three possible values
which (probably) had the following meanings:
* `0`: There was an error while loading the image. Don't try to load again.
* `1`: Default value. Try to load the image.
* `2`: The image was loaded successfully.
This image-wide flag did not make sense unfortunately, because loading
may work for some frames of an image sequence but not for others.
Remember than an image data block can also contain a movie.
The purpose of the `->ok` flag was to serve as an optimization to avoid
trying to load a file over and over again when there is an error (e.g. the
file does not exist or is invalid). To get the optimization back, the patch
is changing `MovieCache` so that it can also cache failed load attempts.
As a consequence, `ibuf` is allowed to be `NULL` in a few more places.
I added the appropriate null checks.
This also solves issues when image sequences are used with the
Image Texture node in Geometry nodes (also see D12827).
Differential Revision: https://developer.blender.org/D12957
