- New operator property to toggle edge panning in the keymap:
This is disabled by default to avoid edge-panning in cases where it
gets distracting, such as adding a new node. Only the explicit
translate operator(s) (GKEY or drag) have this enabled now.
- Restore the initial view rect on edge pan cancel:
The initial view rect is now stored in the edge pan operator data.
When an operator with edge panning is cancelled it can now call the
`UI_view2d_edge_pan_cancel` function to restore the original View2D
rect.
- Less delay in node editor scrolling:
Delay is useful when scrolling through long lists, such as in the
outliner, but makes node scrolling feel sluggish and unresponsive.
The lower scroll speed here makes a faster response the better option.
- Zoom influence feature:
Somewhat slower scrolling in UI-space when zoomed out. With the 0.5
zoom influence factor nodes behave as if zoom factor is halved,
otherwise it gets too fast when zoomed out. Previously scrolling would
always be constant-speed in UI space, now it's half-way between UI
space and node (view) space.
Before this patch attempting to remove a particle modifier programmatically
through Python would fail, because it deleted the modifier associated with
the currently active particle system instead of the one passed as an argument
to `bpy.types.ObjectModifiers.remove()`.
This fix adds an additional argument for the particle system to
`object_remove_particle_system`. This allows to specify which particle system
and its associated modifier shall be removed. In case of
`particle_system_remove_exec` it will remain the currently active particle
system, whereas `object_remove_particle_system` passes the particle system
of the modifier. Hence, the correct modifier will be removed.
Reviewed By: mont29
Differential Revision: https://developer.blender.org/D12234
The idea is to only allocate pixel storage only when there is an actual
data to be written to them.
This moves the code forward a better support of high-res rendering when
pixel storage is not allocated until render engine is ready to provide
pixel data.
Is expected to be no functional changes for neither users no external
engines. The only difference is that the motion and depth passes will
be displayed as transparent for until render engine provides any tile
result (at which point the pixels will be allocated and initialized to
infinite depth).
Differential Revision: https://developer.blender.org/D12195
This overlay was intended to identify the active objects in modes
like Sculpt Mode, where you don't have any extra visual indication
of what is the current and target object when switching directly
between them.
After having flash on mode transfer on the transfer mode operator,
the visual information this overlays provides is redundant. It is
still available in case some users want to use it like a way of
focusing on the active object.
Reviewed By: JulienKaspar, JacquesLucke
Differential Revision: https://developer.blender.org/D12303
We need to separate the flag telling duplicate code to not handle
remapping to new IDs etc., from the one telling the code that we are
currently duplicating a 'root' ID (i.e. not a dependency of another
duplicated ID).
This whole duplicate code/logic is still fairly unsatisfying, think it
will need further refactor, or maybe even re-design, at some point...
Add a string length argument to BLI_str_utf8_as_unicode_step to prevent
reading past the buffer bounds or the intended range since some callers
of this function take a string length to operate on part of the string.
Font drawing for example didn't respect the length argument,
potentially causing a buffer over-read with multi-byte characters
that could read past the end of the string.
The following command would read 5 bytes past the end of the input.
`BLF_draw(font_id, (char[]){252}, 1);`
In practice strings are typically null terminated so this didn't crash
reading past buffer bounds.
Nevertheless, this wasn't correct and could cause bugs in the future.
Clamping by the length now has the same behavior as a null byte.
Add test to ensure this is working as intended.
Simplify logic of speed effect frame calculation by using discrete math
where possible. Only `SEQ_SPEED_MULTIPLY` mode with animation requires
frame map to be built. Frame map building was simplified by removing
unused branches.
Functional change: Animating strip in negative range will reverse playback.
I assume this was limitation of previous system, where each frame map item
was limited to be within correct frame range. Now frame map can contain
values that point beyond usable range and they are limited by
`seq_speed_effect_target_frame_get`. This way it is possible to control
playback rate in both directions.
Mostly fixes T89120 apart from offset handling.
Reviewed By: mano-wii
Differential Revision: https://developer.blender.org/D11939
Prefetch needs to avoid rendering scene strips, because
- Rendering in background needs own dependency graph, which fails to
initialize from evaluated data.
- This locks UI and can make it unresponsive for long time periods.
In T88237 prefetch failed to avoid scene strip, because of effect strip
was attached to scene strip.
Ensure, that no effect that is attached to scene strip either directly
or indirectly would be rendered.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D11247
When splitting strips, first they are duplicated and then offsets
adjusted. This can fail on cross transitions, because some strips don't
overlap with split frame.
All strips, that relate to each other must be duplicated to ensure
correct relations after splitting, so solution is to delete non
overlapping strips from left or right side respectively.
Since cross transition don't have to overlap with source strips,
splitting such strips would lead to effect being deleted, which
could cause crash when iterating over strips in python. Therefore
splitting of such strips is now forbidden and will generate error.
Splitting of transition will also generate error solely because such
operation is illogical.
Reviewed By: sergey
Differential Revision: https://developer.blender.org/D12121
When all strips are selected and overlap is caused, this causes VSE to
hang in infinite loop, because such situation should never happen.
To prevent infinite loop, ensure, that strip overlap is not tested
against single overlapping strip itself.
Prevent overlap that can not be handled because of issue described above
by moving overlapping strip between channels.
Reviewed By: campbellbarton
Differential Revision: D12209
`CurveMapping.evaluate` function expectes `CurveMapping` to be
initialized, while this wasn't documented.
I don't see any reason for not initializing `CurveMapping` on demand.
Initialization was added in rBf16047c2df1e8be56bf76524f9eb1fa5ecde2176
Reviewed By: campbellbarton
Differential Revision: https://developer.blender.org/D12145
Changes the threshold comparison from absolute to relative.
Removes threshold for MLoopCol comparison.
Adds a compare relative threshold function.
Reviewed By: JacquesLucke
Differential Revision: https://developer.blender.org/D12273
`MeshBufferCache` is a struct representing a list of buffers.
As such, `GPUIndexBuf **tris_per_mat` is out of place as it does not
represent one of the buffers in the list.
In fact this member should be close to `GPUBatch **surface_per_mat` as
they are related.
The code for dependencies between buffer and batch had to be reworked
as it relies on the member's position.
Differential Revision: https://developer.blender.org/D12227
When using a Cryptomatte node and selecting 2 views in Multi-View,
its output values are doubled. When selecting 3 tripled and so on.
This causes incorrect compositing results for all the views.
The node creates an input operation for each rendered cryptomatte
pass. In Multi-View, passes are rendered for each view but compositor
is executed per view and should only create operations for those
corresponding to the current view being executed. Otherwise duplicated
operations add up later in cryptomatte operation.
Reviewed By: jbakker
Maniphest Tasks: T89998
Differential Revision: https://developer.blender.org/D12216
Adds full frame implementation to Channel Key, Chroma Key, Color Key,
Color Spill, Cryptomatte, Difference Key, Distance Key, Keying,
Keying Screen and Luminance Key nodes. The other nodes
in "Matte" sub-menu are submitted separately.
No functional changes.
Part of T88150.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D12220
Adds full frame implementation to these nodes operations.
When enabling "extend bounds" node option, tiled implementation
result is slightly different because it's using `TranslateOperation`
with bilinear sampling for centering.
Full frame always uses nearest to don't lose image quality.
It has the disadvantage of causing image jiggling on backdrop
when switching size values as it's not pixel perfect.
This is fixed by rounding to even.
No functional changes.
Part of T88150.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D12167
Adds full frame implementation to "Displace", "Crop", "Flip",
"Plane Track Deform", "Corner Pin", "Movie Distortion",
"Lens Distortion" and "Map UV" nodes.
The other nodes in "Distort" sub-menu are implemented
separately in other commits.
No functional changes.
Part of T88150.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D12166
Adds full frame implementation to "Rotate", "Transform" and
"Stabilize2D" nodes.
To avoid sampling twice when concatenating scale and rotate
operations, a `TransformOperation` is implemented with all
the functionality.
The nodes have no functional changes.
Part of T88150.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D12165
Current sampling methods have off by one issues on full frame:
- Bilinear sampling do not fully sample bottom and left image border,
creating edges.
- Single elem buffers are not sampled at all when they should be
at least on the borders to smooth edges.
- EWA filtering is partially implemented on `ReadBufferOperation`, it
needs to be moved to `MemoryBuffer` on full frame.
In order to not affect tiled implementation, this commit creates
specific sampling methods for full frame needs.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D12164
Adds full frame implementation to all nodes in "Converter" sub-menu
except "ID Mask" which is implemented separately.
No functional changes.
Part of T88150.
Reviewed By: jbakker
Differential Revision: https://developer.blender.org/D12095
Authored by Henrik Dick (weasel)
Reviewed By YimingWu (NicksBest), Antonio Vazquez (antoniov)
Differential Revision: https://developer.blender.org/D12284
selected pass
Caused by {rBebaa3fcedd23}.
Seems this above commit assumed an ImageUser's multi_index is only used
for Multiview/Stereo? This is not the case, multi_index also stores the
index for layer/pass combination.
If we call both BKE_image_multilayer_index and BKE_image_multiview_index
(even though this is not appropriate/needed for multilayer images?), we
might end up overwriting multi_index again.
note: looking at this I was also wondering why we update the ImageUser
in image-buffer-aquiring funnctions [and not from the UI, e.g.
template_image_layers, but that is a whole different story I guess, see
comment in T90772 as well]
note2: this could also use a utility function (this is not the only
place where this is done), this is fo a cleanup commit.
Maniphest Tasks: T90772
Differential Revision: https://developer.blender.org/D12267
This was working differently in 2.79, tried tracking this down and it
seems this was wrong since the 2.8 beginning in {rB7907dfc40018}.
This would not only crash without an active scene camera, but would also
result in different tracks from different camera's constraints could not
be selected.
So select id depends on corresponding camera, remove the dependency on
scene camera completely.
Maniphest Tasks: T90651
Differential Revision: https://developer.blender.org/D12230
Compressing blendfiles can help save a lot of disk space, but the slowdown
while loading and saving is a major annoyance.
Currently Blender uses Zlib (aka gzip aka Deflate) for compression, but there
are now several more modern algorithms that outperform it in every way.
In this patch, I decided for Zstandard aka Zstd for several reasons:
- It is widely supported, both in other programs and libraries as well as in
general-purpose compression utilities on Unix
- It is extremely flexible - spanning several orders of magnitude of
compression speeds depending on the level setting.
- It is pretty much on the Pareto frontier for all of its configurations
(meaning that no other algorithm is both faster and more efficient).
One downside of course is that older versions of Blender will not be able to
read these files, but one can always just re-save them without compression or
decompress the file manually with an external tool.
The implementation here saves additional metadata into the compressed file in
order to allow for efficient seeking when loading. This is standard-compliant
and will be ignored by other tools that support Zstd.
If the metadata is not present (e.g. because you manually compressed a .blend
file with another tool), Blender will fall back to sequential reading.
Saving is multithreaded to improve performance. Loading is currently not
multithreaded since it's not easy to predict the access patterns of the
loading code when seeking is supported.
In the future, we might want to look into making this more predictable or
disabling seeking for the main .blend file, which would then allow for
multiple background threads that decompress data ahead of time.
The compression level was chosen to get sizes comparable to previous versions
at much higher speeds. In the future, this could be exposed as an option.
Reviewed By: campbellbarton, brecht, mont29
Differential Revision: https://developer.blender.org/D5799
