Currently, the compositor template ID creates a new node tree. This is
inconsistent with other node editors where the button copies the node
tree.
Also use "New" in geometry nodes operator description instead of "Copy"
Pull Request: https://projects.blender.org/blender/blender/pulls/146222
This patch makes a distinction between passes accessed through nodes
like the Render Layers node, and inputs accessed through nodes like the
Group Input node. Previously, the Group Input node returned the passes
of the active scene and view layer accordingly to the name, while now,
it only returns the Image (Combined) pass and the result will be zeros.
Pull Request: https://projects.blender.org/blender/blender/pulls/146831
The Node Wrangler addon has a _Reset Nodes_ operator that can remove the input
node of a node zone. This crashes in reference set updates because the code
expects valid input/output node pairs in each zone.
The fix is two-fold:
1. Finding zones for the runtime now returns an empty result to ensure no
invalid node pointers are being accessed. This should not happen in practice,
all operators should make sure zone relationships are not broken.
2. The node wrangler addon is updated to ignore all zone types, including the
newer repeat, closure, and for-each-element zones. The type filtering was
outdated and now uses the `bl_idname` consistently.
Pull Request: https://projects.blender.org/blender/blender/pulls/147028
Currenty, when switching from one scene to another, the mode
of the active objects (before & after) is left untouched.
This becomes a workflow problem when trying to work across
different scenes. E.g. when opening the new "Storyboarding"
template and scrubbing to the next scene strip, the Grease Pencil
object is now in Object Mode, when in the previous shot it was
in Draw Mode.
To solve this, when syncing scenes try to change the mode of
the next active object to the mode of the previous active object
if the type of the object is the same.
Pull Request: https://projects.blender.org/blender/blender/pulls/146981
Fix: correctly update Actions when renaming/re-typing a custom property
When renaming a custom property on a data-block, only update the
animation in a slot for that data-block. Previously all F-Curves were
updated, even when they were meant for other data-blocks.
This also handles the cases where the type of a custom property is
changed, and not just its name.
This is part of #146586
Pull Request: https://projects.blender.org/blender/blender/pulls/146979
Regression in [0] removed checks for indices referencing themselves
which need to be kept but can still be used as targets.
Restore this logic as well as fixing another problem (#147022)
where auto-merge would not merge into the nearest vertex, this
was especially noticeable then the threshold was set to a large value
but would happen at smaller values too.
[0]: bdae3e28a2
A long overdue pass on editor descriptions/tooltips.
Some editors used a variation of the editor name as description, while
others didn't even have one. In some editors their functionality
changed so use this opportunity to reflect this (UV for example,
no longer depends on "mesh editmode").
See PR for details.
Pull Request: https://projects.blender.org/blender/blender/pulls/146968
To support setting the custom `CurveMapping` to a well defined preset,
there exist a number of operators that are hardcoded to apply a
particular preset to a particular curve.
This commit begins to replace this functionality and make it part of the
template itself, allowing the preset to be applied to any curve. For
now, it only supports either positive or negative slopes, primarily for
the brush usecases.
The `brush.curve_preset` and `brush.sculpt_curves_falloff_preset`
operators are unneeded after this change and have been removed.
Notably, these preset controls have not been added elsewhere, they can
be added on a case by case basis in future commits by interested
modules.
Pull Request: https://projects.blender.org/blender/blender/pulls/146847
Based on extensive testing, this gives matching HDR brighness across most
application, with both PNG and AVIF. Video remains at 100 nits as that
appears to tbe convention there.
This is implemented by adding two modified PQ and HLG color spaces to the
OCIO config on startup, and for the specific cases of image save and loaded
these will replace the regular PQ and HLG color spaces.
This was chosen rather than adding them as color spaces in the OCIO config,
so that it can work for any config with appropriate interop IDs, including
the ACES config. Additionally, it would be unclear how to make this work with
view + display transforms, we wouldn't want to burden the users with having
to pick a different display depending if they are saving images or video.
Ref #145855, #144911
Pull Request: https://projects.blender.org/blender/blender/pulls/146888
The "id" attribute was built-in on curves (point domain), mesh (point domain)
and instances, but not on e.g. point clouds. This mismatch causes issues because
of the special rules regarding built-in attribute propagation. Furthermore, an
implication of this was that the id attribute had to be an integer attribute on
a specific domain, which is not always ideal.
This patch turns the id attribute into a normal generic attribute, except that
some nodes internally modify this attribute in special ways. This may cause some
compatibility breakage in rare cases, but that can generally be easily fixed by
either removing the id attribute or setting it explicitly on the right domain. I
can't think if a feasible way to avoid this unfortunately.
The internal special cases for the id attributes are generally skipped unless
the attribute is an integer attribute on the point/instance domain.
Pull Request: https://projects.blender.org/blender/blender/pulls/146941
This adds three corner types: `Round` `Sharp` and `Flat`.
These control how the corner of `Grease Pencil` line strokes are rendered.
- The `Round` type draws circular arcs, and is what `Grease Pencil`
currently supports and is default.
- The `Flat` type cuts off the tip of the corner.
- The `Sharp` type allows for sharp corners to be created.
If the angle is sharper than `Miter Limit` then the tip will be cut like `Flat`
These three types match the main types of `line joins` present in `SVG`
files.
This data is stored in one `Point` attribute called `miter_angle`.
This stores both the `Corner Type`, the `Miter Limit` and defaults to
the `Round` type.
This PR adds:
- Rendering of the corner types.
- An operator for setting the `Corner Type` and `Miter Angle` attribute.
- Corner types to the `Outline` operator and modifier.
Part of #145380.
Pull Request: https://projects.blender.org/blender/blender/pulls/143688
Built-in nodes already used the functionality to hide the input labels in a few
situations. However, this functionality was not exposed for node groups before.
It required some refactoring to get the semantics of the flag right. The tricky
aspect is that "Hide Label" would be an incorrect name, because the label is
still shown under various circumstances. Instead, the flag merely indicates that
drawing code may skip drawing the label.
This adds a new "Optional Label" input for node group inputs. Other names are
possible like "Requires Label" which would be the inverse.
Overall the implementation is pretty straight forward after
1f489ea31a,
469a70dba9 and
f79896f5b9.
Pull Request: https://projects.blender.org/blender/blender/pulls/146939
In 95259228d9 an optimization was made to
simplify geometry nodes socket names on panels when they contain the
exact group title as prefix, but this leads to issue where `Opening` can
be truncated as `ing` which is not desired. Now will check if a space
character is present after the prefix, then do the truncation.
Note that this implementation still doesn't truncate `CamelCase` or any
CJK name strings because they don't use the space character as
separator.
Pull Request: https://projects.blender.org/blender/blender/pulls/146973
This implements the Menu Switch node in shader nodes. It's the same node that is
used in Geometry Nodes and the Compositor.
The Menu Switch node is purely handled during preprocessing and thus builds on
top of #141936. Hence, it's input has to be a single value, just like the
iteration count for repeat zones. This limitation can be lifted in the future,
but currently there is no way to produce a non-single menu value in shader
nodes. This will become possible if other Switch nodes are added though.
Pull Request: https://projects.blender.org/blender/blender/pulls/146896
Caused by a bad conversion to StringRef in 950a2bb8a3.
Instead pass around StringRef since that's all that's needed by the UI code
anyway. And make the argument to the node socket drawing RNA function
clearer, previously it relied on the char pointer being first in StringRefNull.
Pull Request: https://projects.blender.org/blender/blender/pulls/146985
Since 1122a05cb6 tool settings could return None, but we do not check this
consistently in scripts. Now always return some tool settings, since it's difficult
to verify and easy to forget proper null checks in e.g. operator poll functions.
The sequencer UI code was updated to continue showing tool settings only when
there is a sequencer scene.
Co-authored-by: Sergey Sharybin <sergey@blender.org>
Pull Request: https://projects.blender.org/blender/blender/pulls/146166
Following to the uiLayout refactor, this converts each
layout resolve function as virtuals methods, so `x_`, `y_`,
`w_` and `h_` properties can become protected.
No user visible changes expected.
Pull Request: https://projects.blender.org/blender/blender/pulls/146185
Current approach for computing the curve tangent approximations using
the average of the sum of the normalized segment direction vectors
suffers from catastrophic cancellation as the angle between the segments
approach 0°, with increasingly large error in resulting tangent.
Improving current behavior by conditionally computing the tangent using
the cross product to rotate the difference around the orthogonal axis
show a significant improvement for small angles. For segments in an axis
aligned plane, additional errors from computing the cross product is
avoided as terms cancel out. When the rotation axis is not axis aligned,
the method still shows a constant improvement for angles smaller
than ~45°.
Changes improves behavior for cases such as #145558.
For more information see https://projects.blender.org/blender/blender/pulls/146332.
Pull Request: https://projects.blender.org/blender/blender/pulls/146332
Attribute writer creation can fail in cases where a built-in attribute exists
and prevents arbitrary name/type/domain combinations. In those cases the
attribute should just be skipped.
Pull Request: https://projects.blender.org/blender/blender/pulls/146976
Hardocde the position of the essentials assets in the menu,
for consistent alphabetical order and a clearer indication of
the different between the nodes and builtin "Array" modifiers
(the old one is now called "Array (Legacy)").
Pull Request: https://projects.blender.org/blender/blender/pulls/146887
OpenGL has the concept of streaming buffers these buffers are marked to be
rewritten and used a small number of times. Vulkan (being low-level API)
doesn't have this concept. When performing font rendering Blender uses
streaming buffers and can slow down Vulkan as GPU barriers are added
between uploading and using the buffer.
Using a different approach could reduce the GPU barriers. The overall
idea is:
Altering render graph nodes
During font rendering the streaming buffer is rewritten from start with
the data to render the next part. This could only cover a part of the fully
allocated buffer. And would introduce a barrier before and after rewritting
the next part.
The allocated buffer on the GPU can fit more data but that data needs to be
passed along the first update to reduce the barriers. Allowing access to an
existing node in the render graph would allow to change the initial upload to
upload more data, without additional barriers.
VKStreamingBuffer
A new buffer type is introduced that will keep track of the streaming buffer
on the current render graph. A streaming buffer can be shared between multiple
threads and requires state manager to be done per context.
Pull Request: https://projects.blender.org/blender/blender/pulls/146956
7becc38a3c introduced new text rendering.
In the refactoring the vertex buffer was replaced by a more shallow
storage buffer. However the refactoring removed one optimization that
the vulkan backend uses, namely the actual amount of bytes that is being
used by the draw call. This resulted in overly large data transfers.
For example in the text editor the texts are rendered one glyph at a
time. But the storage buffer would always upload the data for 1024
glyphs.
This PR allows the usage size of a storage buffer to be set to ensure
that the data transfers are limited.
The implementation wasn't added to OpenGL as it draws incorrectly.
Issue detected during the research of !146956 That PR will implement
better data streaming inside the Vulkan backend and also requires
to know the actual usage size of the buffer to detect what data can
be grouped together.
Pull Request: https://projects.blender.org/blender/blender/pulls/146958
The current filter was still using the legacy Grease Pencil
type (now Annotations).
Now this uses the correct `FILTER_ID_GP` and uses the
other filter for Annotations.
Pull Request: https://projects.blender.org/blender/blender/pulls/146978
Deletion of NTree IDs through custom code path using
`blender::bke::node_tree_free_tree` is fully by-passing standard ID
management code, and was missed when ID::runtime was made an allocated
pointer.
We should really spend some time at some point to get these custom temp
node IDs handling code under control, there should be little to no need
for all this custom allocation and deletion code anymore nowadays.
Caused by 866fcd0a09
To search for strips, we do so on an `Editing` object or a `MetaStrip`
object.
Code from 866fcd0a09 got the `MetaStack` but then tried to use this
directly as a `MetaStrip`.
We should be getting its `parent_strip` though it seems.
For reference, this is what was done prior to 866fcd0a09
`sequences_object = ed.meta_stack[-1]`
Pull Request: https://projects.blender.org/blender/blender/pulls/146967
This adds a new operator which can join multiple nodes together. Currently, it
only supports joining Group Input nodes. However, in the future it could be
extended to join e.g. Bake and Capture Attribute nodes.
This uses the recently freed up ctrl+J shortcut for this functionality, which
feels natural to me.
The implementation is fairly straight forward. The main tricky aspect is
sometimes the nodes can't be joined when that would result in two sockets being
linked to each other twice. In this case, the a separate Group Input node is
kept.
The selected nodes are merged into the active node (in case the active node is
part of the selection, otherwise there is a fallback).
Pull Request: https://projects.blender.org/blender/blender/pulls/146894
When writing .blend files, Blender traditionally wrote raw runtime-pointers into
the file. Since these pointers are different whenever Blender is restarted or
the file is loaded again, the written .blend file will be very different every
time. The file always changing is a problem with tools that use change-detection
on .blend files such as:
* Change detection during undo in Blender. When a serialized data-block is the
same in two consecutive undo steps, it's known that the data didn't change and
the data-block does not have to be reloaded and can potentially skip depsgraph
evaluation. Also see #141262.
* BAT: https://projects.blender.org/studio/flamenco/issues/104437
* Generally using .blend files in version control. The diffs can be smaller when
pointers aren't changing all the time and have a lower memory footprint.
This PR makes pointers in .blend files across multiples saves much more
consistent, improving the situation for the cases above. Although there is still
other data that changes on almost every save currently; that needs to be
addressed separately.
The basic design for pointer stability in blend files stable pointers, described
in #127706, is fairly straight forward. This patch implements a slightly
modified variant of that design. When reading .blend files, Blender already does
not care if the stored pointer values are actual pointer values, or just some
arbitrary identifiers. Therefore, we mainly just have to change the write-file
code. This also implies that this change is fully forward and backward
compatible.
The main non-obvious aspect of this patch is how to actually do the remapping of
runtime pointers to stable identifiers. In theory, having a single integer that
increments for every newly detected pointer works. But in practice that leads to
many changes in the .blend file because one pointer is added or removed
somewhere, all subsequent pointers will be different too. So some kind of
scoping is required to make sure that one small change does not affect
everything else.
This PR starts a new scope pointer identifier scope whenever a new ID data-block
starts. At first I thought it would be good to have separate maps for id-local
and global pointers. However, that's tricky currently, because at write-time, we
don't always have enough information to know if a specific pointer is local or
global. I worked on #146136 to improve the situation but the problem is bigger
than that since we also have various void pointers in DNA structs. Fortunately,
the solution implemented now also works fine with a single global map.
Implicit sharing in undo steps also had to be changed slightly to work with the
stable address identifiers instead of raw pointers.
There's also new code to find all pointers in DNA structs in the first place.
This is done once when writing starts. Then whenever a struct is written, a copy
is made, all pointers are replaced and the modified struct is written to the
.blend file. There is an optimization for the case when a struct does not
contain any pointers because then the copy can be skipped.
For checking the diff between two saved .blend files, I recommend enabling
`GENERATE_DEBUG_BLEND_FILE` and then diffing the text version of the .blend
files.
Co-authored-by: Hans Goudey <hans@blender.org>
Pull Request: https://projects.blender.org/blender/blender/pulls/127729
Each name has to be unique within a group, so when renaming an idproperty, one
has to make sure that the parent group does not contain duplicates afterwards.
This patch raises a `NameError` when setting the name to one that exists
already. Alternatively, one could delete the already-existing property, but that
seems unexpected and the user should rather do that explicitly.
This also adds a new unit test for this case.
Pull Request: https://projects.blender.org/blender/blender/pulls/146892
The grid layout for UI lists wasn't used in practice from all we can
tell. It was badly maintained for a long time (bugs went unnoticed). I
think it was added for an earlier version of the asset UI design.
This was planned for removal in 5.0, see blender/blender#110461.
Usages in bundled scripts were already removed in efa8d942b8.
Pull Request: https://projects.blender.org/blender/blender/pulls/146656
When drag is initiated, store selected keyblocks inside pointer array
(see: `create_drag_data`). Later iterate over these pointers inside
`on_drop()` function to place them one by one at `drop_index`
Resolves#143929
See PR description for video
Pull Request: https://projects.blender.org/blender/blender/pulls/144431
Unlike other Outliner views, this will not open up elements to find
objects "underneath" (since Data API view lazy loads elements iirc), but
given that you already have that object visible somewhere in the DataAPI
view (so e.g. Objects expanded) we could actually do the centering/
scrolling by getting an ID from the `TreeElement` (in case of an
`TSE_RNA_STRUCT`).
This came up in #145904
Pull Request: https://projects.blender.org/blender/blender/pulls/145920
