Some of the files are not accurate to blender-v4.3-release and not due to any goo engine additions
Likely due to merges from beyond 4.3 that made it into 4.2 since it is an LTS and gets more support
So merging these changes to keep it in line with the release version of 4.3
Temporary directory handling had a logical error, assuming the
"session" temporary directory was owned and created by Blender
and could be recursively removed on exit.
However, it's possible creating the session sub-directory fails,
in that case the temporary directory was used for the "session".
This meant setting `C:\` as the temporary directory in the preferences
would attempt to recursively remove `C:\` on exit.
Resolve with the following changes:
- Only perform a recursive removal on the temporary directory
if a session sub-directory was created.
- If the creating the user-preferences temporary "session" sub-directory
fails fall back to the systems temporary directory and try to
create the "session" directory there.
Previously this was only done if the preference path didn't exist.
The preferences path was still used if it existed but couldn't be
written to.
Include a test to ensure this is working as expected.
Ref !144042
This change moves the tests data files and publish folder of assets
repository to the main blender.git repository as LFS files.
The goal of this change is to eliminate toil of modifying tests,
cherry-picking changes to LFS branches, adding tests as part of a
PR which brings new features or fixes.
More detailed explanation and conversation can be found in the
design task.
Ref #137215
Co-authored-by: Brecht Van Lommel <brecht@blender.org>
Pull Request: https://projects.blender.org/blender/blender/pulls/137219
Pull Request: https://projects.blender.org/blender/blender/pulls/138487
Usually we prefer to propogate attributes based on name, even for
conversion between geometry types. Builtin attributes are a common
exception though. The Curve to Points node and the Curve to Mesh node
didn't correctly handle the "custom_normal" attribute, which shouldn't
be propagated. This has only been a problem since custom normals were
converted to a generic attribute on meshes in f9b627d29c.
Pull Request: https://projects.blender.org/blender/blender/pulls/136452
This makes it possible to restore previous Blender 4.3 behavior of bump
mapping, where the large filter width was sometimes (ab)used to get a bevel
like effect on stepwise textures.
For bump from the displacement socket, filter width remains fixed at 0.1.
Ref #133991, #135841
Pull Request: https://projects.blender.org/blender/blender/pulls/136465
F-Curve interpolation uses the `FCURVE_INT_VALUES` and
`FCURVE_DISCRETE_VALUES` flags, which were not set in the keyframe
insertion function for slotted Actions. This is now resolved by making
the RNA property type part of the `FCurveDescriptor`.
Existing code has been refactored a bit, mostly to allow calling
`update_autoflags_fcurve_direct()` with just the RNA property type,
instead of passing the property itself. This avoided the need to include
pointers to RNA properties in `FCurveDescriptor`, which I think is a
slightly nicer design. It also makes it more explicit which aspect of
the property is used.
Because there's now another `std::optional<>` in the `FCurveDescriptor`,
I've also changed some `std::nullopt` to `{}` for brevity of the code,
as repeating that another time would have caused longer lines with more
rewrapping.
Pull Request: https://projects.blender.org/blender/blender/pulls/136446
Add the ability to control whether or not Hardware Ray Tracing (HWRT)
is enabled during render tests.
This is done by adding `-RT` to the end of the device name in the
Cycles device list. E.g. `HIP-RT`. This is supported with HIP, oneAPI,
and Metal.
Change in behaviour:
If you do not specify `-RT`, then HWRT will be disabled. This results
in a change in behaviour:
1. `METAL` device tests on M3 or newer Macs no longer using MetalRT
2. `ONEAPI` device tests no longer use Embree GPU.
Note: Some tests are failing on some platforms/configurations that can
now be easily tested due to this commit. This does not effect the
build bot automated testing as it does not test these configurations.
These tests have not been blocked from running, primarily to help
developers investigate and fix the issues.
Ref #123012
Pull Request: https://projects.blender.org/blender/blender/pulls/125082
Add a new RNA function `ActionSlot.users()` that returns the
data-blocks that are animated by this slot.
This covers direct assignment of the action & slot, but also use in
the NLA and in Action constraints.
```python
>>> D.actions['SuzanneAction'].slots['OBSuzanne'].users()
[bpy.data.objects['Suzanne']]
```
This was implemented as a function, and not a collection property,
because Blender's bookkeeping of the slot users can be marked 'dirty'.
In that case the slot user list needs to be rebuilt, which happens for
all Actions and all their slots simultaneously. This was considered
too broad a data-changing action to 'hide' inside a getter of a
property. Also it needs a `bmain` pointer, which is not available in
getters, but is available in functions.
Pull Request: https://projects.blender.org/blender/blender/pulls/135734
`ColorManagedDisplaySettings`, `ColorManagedViewSettings`,
`ColorManagedInputColorspaceSettings` were affected since these are used
from multiple places (Scene, File output nodes, ..).
Similar to how we are getting the path for `ImageFormatSettings` (which
pretty much has the distinction logic already), we can do so for the
ColorManage Settings as well (piggbacking on the already correct path to
the `ImageFormatSettings`.
With this, we can also remove these "known failures" from the test
introduced in 4032b853c3.
Pull Request: https://projects.blender.org/blender/blender/pulls/135238
This applies to modifiers, constraints and shape keys.
Any driver on such data was not removed with it,
leaving invalid drivers on objects.
With this patch, the drivers are removed, but animation is left untouched.
This is because animation might be reused somewhere else and we don't
want to introduce potential side effects there.
This also adds unit tests for the fixed cases
Pull Request: https://projects.blender.org/blender/blender/pulls/134511
This adds support for Overlay tests.
There are some differences with how we handle tests for other engines:
- The renders are captured using `bpy.ops.render.opengl()`, but this
won't work on our GPU build bots.
- A single blend file can run multiple tests by outputting a txt list
with the test names.
- Each overlay test blend file requires a matching script file with
the same name inside `tests/python/overlay/`.
- To reproduce a specific test state you can run
`blender "(...)/tests/data/overlay/<test>.blend" -P "(...)/tests/python/overlay/<test>.py" -- --test <test-number>`.
Note:
The current test permutations are WIP, so reference images are not
committed to the data repo for now.
Pull Request: https://projects.blender.org/blender/blender/pulls/133879
Expose the convenience function `blender::animrig::action_fcurve_ensure()`
to RNA as `Action.fcurve_ensure_for_datablock(...)`.
The function requires that the Action is already assigned to the
data-block. It then takes care of slot assignment / creation, as well as
the creation of a layer and a keyframe strip.
This function call:
```python
fcurve = action.fcurve_ensure_for_datablock(ob_cube, "location", index=2)
```
effectively performs this logic:
```python
# Ensure the slot exists and is assigned:
slot = ob_cube.animation_data.slot
if not slot:
slot = find_slot_for_keying(action)
if not slot:
slot = action.slots.new(ob_cube.name)
ob_cube.animation_data.slot = slot
# Ensure a layer exists:
if action.layers:
layer = action.layers[0]
else:
layer = action.layers.new("Layer")
# Ensure a keyframe strip exists:
if layer.strips:
strip = layer.strips[0]
else:
strip = layer.strips.new('KEYFRAME')
# Ensure the channelbag exists:
channelbag = strip.channelbag(slot, ensure=True)
# Ensure the F-Curve exists:
fcurve = channelbag.fcurves.find("location", index=1)
if not fcurve:
fcurve = channelbag.fcurves.new("location", index=1)
```
Here `find_slot_for_keying()` represents the logic that's also used when
creating keys via the user interface or the `bpy_struct.keyframe_insert()`
function.
Pull Request: https://projects.blender.org/blender/blender/pulls/134686
In a recent commit (1), a bug was fixed where the generated texture
coordinates on volumes in Cycles with OptiX OSL was incorrect.
Due to this fix, OptiX OSL now passes render tests in scenes that use
that feature, so this commit enables those tests on that platform.
(1) 0177537c9a
Pull Request: https://projects.blender.org/blender/blender/pulls/134629
`environment_mirror_ball.blend` and `image.blend` are both
failling because of mipmap/anisotropic filtering differences
on apple silicon.
The volume tests has issues with lightprobe baking.
Blocking them until we have a workaround.
The derivatives of the normal were simply not computed.
The offsetted normals are computed by perturbating the barycentric
coordinates. At triangle boundaries, the normals are extrapolated,
so discontinuities might be visible.
Currently only supported on triangles.
Pull Request: https://projects.blender.org/blender/blender/pulls/133769
Use sub-pixel differentials for bump mapping helps with reducing
artifacts when objects are moving or when textures have high frequency
details.
Currently we scale it by 0.1 because it seems to work good in practice,
we can adjust the value in the future if it turns out to be impractical.
Ref: #122892
Pull Request: https://projects.blender.org/blender/blender/pulls/133991
This allows Python scripts to easily determine what Slot a Channelbag is
for. This is particularly important because we're trying to discourage
the use of Slot handles in the Python APIs, and before this the only
way to identify which Slot a Channelbag was for was via the Channelbag's
`slot_handle` property.
Pull Request: https://projects.blender.org/blender/blender/pulls/134053
Previously it was possible to make the type prefix of a Slot's identifier get
out-of-sync with its actual target ID type, by setting the identifier via
Python.
This PR changes `Slot.identifier` assignment to ensure that the type prefix is
set to match `target_id_type`. This now makes it impossible for the identifier
prefix and `target_id_type` to get out of sync, since this API previously was
the only way to accomplish that.
When the prefix that the user attempts to set doesn't match the `target_id_type`
of the Slot, a warning is issued telling the user about the mismatch and that
the identifier has been set with the correct prefix instead.
Pull Request: https://projects.blender.org/blender/blender/pulls/133983
Most of the old Animato properties on an Action (e.g. FCurve list, Channel
Groups) already act as proxies for the data for the first slot in the first
strip of the first layer. (Say that three times fast!) However, this was not yet
the case for `Action.id_root`.
This PR changes `Action.id_root` to act as a proxy for the first Slot's
`target_id_type` property, both for reading and writing.
If the Action has no Slots, then reading always returns 'UNSPECIFIED', and
writing will create a Slot and set its `target_id_type`.
Note that the ability to write to the first Slot's `target_id_type` via
`Action.id_root` conflicts with `target_id_type` supposedly only being writable
when it's still 'UNSPECIFIED' (#133883). Although that's certainly a little
weird, practically speaking this doesn't break anything for now, and is a
temporary kludge to keep `id_root` working until we can remove it in Blender
5.0. `id_root` will be removed entirely in 5.0, resolving this inconsistency.
Pull Request: https://projects.blender.org/blender/blender/pulls/133823
Similar to how brush assets are created and managed this
PR allows to export pose assets into a different library.
Because of this there is a limitation to this where each
asset is stored in a separate blend file.
This may be lifted in the future as there are planned changes in
the design phase: #122061
### Create Asset
Now available in the 3D viewport in the "Pose" menu: "Create Pose Asset".
The button in the Dope Sheet will now call this new operator as well.
Clicking either of those will open a popup in which you can:
* Choose the name of the asset, which library and catalog it goes into.
* Clicking "Create" will create a pose asset on disk in the given library.
It is possible to create files into an outside library or add it in the current file.
The latter option does a lot less since it basically just creates the
action and tags it as an asset.
If no Asset Shelf **AND** no Asset Browser is visible anywhere in Blender,
the Asset Shelf will be shown on the 3D viewport from which
the operator was called.
### Adjust Pose Asset
Right clicking a pose asset that has been created in the way described
before will have options to overwrite it.
Only the active object will be considered for updating a pose asset
Available Options (the latter 3 under the "Modify Pose Asset" submenu):
* Adjust Pose Asset: From the selected bones, update ONLY channels that
are also present in the asset. This is the default.
* Replace: Will completely replace the data in the Pose Asset from
the current selection
* Add: Adds the current selection to the Pose Asset. Any already existing
channels have their values updated
* Remove: Remove selected bones from the pose asset
Currently this refreshes the thumbnail. In the case of custom
thumbnails it might not be something want
### Deleting an existing Pose Asset
Right click on a Pose Asset and hit "Delete Pose Asset". Works in the shelf
and in the asset library. Doing so will pop up a confirmation dialog,
if confirming, the asset is gone forever. Deleting a local asset is basically the
same as clearing the asset. This is a bit confusing because you get
two options that basically do the same thing sometimes,
but "Delete" works in other cases as well.
I currently don't see a way around that.
Part of design #131840
Pull Request: https://projects.blender.org/blender/blender/pulls/132747
The white noise test is failing on macOS EEVEE Metal.
This seems to be a precision issue related to the texture coordinates
being snapped on a whole number.
This commit adds a small offset to the texture coordinates prior to the
snapping to try and avoid this issue.
Ref: blender/blender-test-data!53
Pull Request: https://projects.blender.org/blender/blender/pulls/134010
Rename `ActionKeyframeStrip.channels()` to `.channelbag()`, and change
its first parameter from `slot_handle` to `slot`.
This is to be consistent with `ActionKeyframeStrip.channelbags`, which is
the array of channelbags in the keyframe strip. Having a function that's
singluar makes sense for finding a single element in the array.
The change from using the slot handle to using the slot is to be consistent
with `.channelbags.new(slot)`. Furthermore, the Python API should be using
slot handles as little as possible (they're basically meaningless numbers).
Using the slots directly is preferred. If that's not possible, it is
recommended to use the slot identifier (`slot.identifier`) instead, as that
can be used to look up the slot (`action.slots[slot_identifier]`).
This breaks the glTF add-on, which will be fixed in !133915.
Pull Request: https://projects.blender.org/blender/blender/pulls/133868
Instead of converting to a string immediately, only convert to a string
when necessary, like for trace logging. This reduces the size of the
`USDPrimReader` base class by 24 bytes, makes several Vectors and Maps
smaller at runtime, and reduces some unnecessary string allocations and
conversions in various other places.
The recently added python Hook `get_prim_path` dictionary will now
contain `SdfPath` objects rather than strings; removing the string
conversion cost when calling other USD APIs like `GetPrimAtPath` which
take in `SdfPath` objects to begin with.
This also makes the code a bit more self explanatory since the type
makes it clear what kind of path you are dealing with.
Pull Request: https://projects.blender.org/blender/blender/pulls/133954
Previously, when use of the legacy Action APIs (specifically the `fcurves` and
`groups` properties) caused a new Slot to be created, it would be named "Slot".
This PR changes things so that Slots created that way are named "Legacy Slot",
just like Slots from upgraded legacy Actions.
The rationale is that Slots created in this way are expected (by the code that
created them) to be used as if the Action were still a legacy Action, and it's
good to reflect that in the name of the data. This also makes it clearer to
users when e.g. scripts and addons they're using may not yet have been updated
to fully work with Slotted Actions.
For consistency, this PR also names Layers created in the same way "Legacy
Layer", the same as Layers from upgraded legacy Actions.
Pull Request: https://projects.blender.org/blender/blender/pulls/133888
When loading old blend files, versioned Actions can end up having a slot with an
'UNSPECIFIED' `target_id_type`. Assigning the slot to an ID will then set the
slot's `target_id_type` to match the type of the ID, but there was previously no
way to directly set it via Python if needed.
This PR changes the writability of `target_id_type` to match the extent of its
mutability when assigning a slot to an IDs. Which is to say, it can be set if
it's 'UNSPECIFIED', but not otherwise.
RNA doesn't have a good way to represent this, so we accomplish this with a
custom set function that simply ignores the write if the slot's `target_id_type`
isn't 'UNSPECIFIED'. This isn't ideal, but is the least-bad solution at the
moment.
Pull Request: https://projects.blender.org/blender/blender/pulls/133883
In order to better interop with the broader Alembic/USD ecosystem, align
the crease values we export with what we believe is expected by native
OpenSubdiv, a 0-10 range.
On import we will translate the native OpenSubdiv range back into
Blender's 0-1 range.
To account for SubD assets produced by Blender before this change, a
compat check is put in place for both Alembic and USD to use the old
methodology when encountering such files. The compat check makes use
of the Blender version we place inside the format's metadata fields. Old
assets loaded into a new Blender will look ok. New assets loaded into an
old Blender would need to be reworked.
Pull Request: https://projects.blender.org/blender/blender/pulls/132582
codywinch
Campbell Barton