This simplifies running built-in IO tests with:
ctest -R bf_io_
Also use "bf_io_" prefix for the libraries since it was already used
by some and it's a useful hint the libraries are used for IO.
There should be no Subpixel AA if the user has deselected
"Anti-Aliasing". Render target FT_LOAD_TARGET_MONO is strongly
hinted to pixel grid and shouldn't be messed with.
Pull Request: https://projects.blender.org/blender/blender/pulls/112759
After discussion with module members, it was agreed that "Unassigned"
was better because the word is already used in the asset browser, and
because it sounds more natural.
The operator uses the relatively new "GEOMETRY_OT" prefix,
which wasn't handled in the function that finds a keymap for an
operator. To support this, choose the 3D View keymap depending
on the current mode for the geometry object types.
In the data-block selectors in the node editor and the header,
apply a few rules to increase the separation between tool and
modifier node groups.
1. The modifier accepts non-asset groups or assets marked
as modifiers (created in the modifier context). Tool node groups
cannot be used by modifiers.
2. The tool context can only edit node group assets marked as tools
(created in the tool context).
When switching contexts, the node group is cleared if the state would
be invalid. The tool node group is still stored in the editor's "root node
tree" property. Since it isn't retrieved from the context, the pin button
is useless, so it isn't displayed in the tool mode.
See #101778, #111526.
Pull Request: https://projects.blender.org/blender/blender/pulls/112389
This change makes it so the list interface in the properties panels looks
closer to things like shape keys, vertex groups and so on: there are two
buttons to add selected objects to the collection and remove active item
from the collection, as well as the "extra" drop down menu.
The add operator adds selected objects to the light linking collection
using the Include policy. For the light linking it means that the objects
are added as receivers that receive the light, and for the shadow linking
it means that objects are added as blockers which cast shadow from the
light.
The communication of the active list element is done via context property
similar to how it was done before. The difference is that these properties
are set on a parent of the list layout, which makes it so they are inherited
by the layout hierarchy needed to place the Remove button.
Pull Request: https://projects.blender.org/blender/blender/pulls/112713
Part of #107742.
There used to be a small margin between items since the layout feels a
bit crammed otherwise. But this meant the mouse could be between items,
with no item highlighted or reacting to interactions. This was
especially annoying when dragging over items for drag and drop: in
between items dropping wasn't possible, and the drag-tooltip would
disappear, causing notable flickering during motions over the tree view.
The view item is now slightly enlarged to keep a look that is not too
crammed, and still remove the space between items. Item highlights are
still drawn with a smaller height (matching the normal widget height),
since anything else looked odd to me.
This now feels quite consistent with similar UIs (e.g. File Browser list
view or the Outliner), even though we give the items a bit more space.
The "mask expand" operator from 0083c96125 is not exposed in
the UI and is redundant with the addition of the more general "expand"
operator from 82e7032477. The latter commit even mentioned
that the first mask expand operator was made obsolete.
Pull Request: https://projects.blender.org/blender/blender/pulls/112726
Padding buttons are supposed to be invisible but were drawn like
regular menu items.
For the most part it doesn't make a difference, since most themes
use transparent menu items anyway.
But e.g. the "White" theme is using opaque menu items, which
would causes the padding button to overlap at the corner when
the roundness of the menu background is very high.
Pull Request: https://projects.blender.org/blender/blender/pulls/112731
Inlined sockets in the same vertical space are no longer supported.
This removes `input_output` socket declarations, the inlining feature in
node drawing, and the `Both` option for node group interface sockets.
Versioning code splits existing node group sockets into individual
sockets again. Unfortunately some links may get lost in versioning files
using the feature, because of an unnoticed bug: Socket identifiers have
to be unique in the node group items list but inlined input/output
sockets have the same identifier. This still works for most situations
because uniqueness is only required within input/output lists. Creating
proper unique identifiers will discard any link from the previous output
socket. This cannot easily be fixed without `after_linking` versioning
code, which should be avoided.
Pull Request: https://projects.blender.org/blender/blender/pulls/112560
webp 1.3 changed the filenames on windows to include a `lib` prefix
(ie libwebp.lib rather than webp.lib) now this is a common thing
on linux and cmake has a `CMAKE_FIND_LIBRARY_PREFIXES` variable that
has a list of prefixes to look for during a `find_library` call.
`CMAKE_FIND_LIBRARY_PREFIXES` gets set during the call to the
`project` method in the main CMakeLists of a project. Now for windows
`lib` is *not* a common prefix by CMake, and it doesn't add "lib" to
CMAKE_FIND_LIBRARY_PREFIXES during that call.
so find library doesn't look for it, the libs are not found and an
unhappy time is had by all. Now the most obvious solution would be to
pass `-DCMAKE_FIND_LIBRARY_PREFIXES=lib` to CMake to sidestep this
however, the `project` call will set the variable overwriting
anything you passed through the CLI.
So the fix here is to have `find_library` counter-intuitively look
for both `libwebp` and `webp`
The last webp update changed the filenames of the webp libraries
on windows causing oiio not to find them and oiio silently build
without webp support, which only came to light after all of
blender was build and a test failed.
This change makes the OIIO validate and error out if certain
dependencies are not found at configure time so these mistakes
are caught early.
- Use `StringRef` over `StringRefNull`, since it handles `nullptr` in
its constructor and gives more flexibility about the string's source.
- Change `const char *` to `StringRef` as well, for improved ergonomics
and efficiency.
- Move `add_interface_socket_from_node` to a .cc file. It's not small
or performance sensitive enough to be inlined.
- Avoid allocating empty strings for item descriptions.
- Allocate with `BLI_strdupn` to avoid unnecessary `strlen` calls.
- Move a bit more code into a proper namespace.
Pull Request: https://projects.blender.org/blender/blender/pulls/112721
`Restore item` button is not resetting `active` property when keyitem is registered
as `inctive` by default. To fix this, clear `inactive` flag then use default keyitem's
flag to correctly restore the keyitem.
Pull Request: https://projects.blender.org/blender/blender/pulls/112497
Support the property settings that were implemented some time ago.
As an aside, change the defaults for min and max to integers: integer
values can be implicitly cast to float, but the reverse results in an
error. Also, pointer properties don't support a default value.
Pull Request: https://projects.blender.org/blender/blender/pulls/112712
AgX and Filmic have different looks in the configuration, but the looks
share the same UI name and purpose. So we can switch to the
corresponding look when switching view transforms.
Thanks to Yiming Wu for helping track down the problem.
Pull Request: https://projects.blender.org/blender/blender/pulls/112686
The goal is to make the evaluation of repeat zones more efficient by making
use of the lazy-function evaluation system. Performance is improved in two ways:
* Unnecessary nodes are not evaluated anymore. E.g. if a repeat zone outputs two
geometries but only one of those is actually used, the other one will not be
computed anymore.
* Support evaluating different iteration indices at the same time on different threads.
It is possible that some uses of repeat zones become slower with this refactor,
especially when each iteration does very little work and there are a lot of iterations.
The old implementation was not optimized for this use case either but now there
is a bit more overhead constant overhead per iteration than before.
On the bright side, this change can result in some very significant speedups when
some computations can be skipped. See #112421 for some example setups.
There is one todo comment for adding back-links for socket usages. Properly linking
those up can result in better (shorter) life-times for anonymous attributes. Even without
that, performance is already better than before.
The implementation reuses the existing lazy-function graph system for the repeat
zone, by building a dynamically sized graph based on the number of iterations.
Building a lazy-function graph makes it possible to use the lazyness and multi-threading
features of the lazy-function graph executor. This is much easier than reimplementing
this behavior for repeat zones specifically (hence there was only single-threaded eager
execution before).
Pull Request: https://projects.blender.org/blender/blender/pulls/112421
Without this it was necessary to convert pasted text into spaces
for them to display properly.
Since tabs are a valid part of a string, it's incorrect to assume
tabs can always be expanded to spaces.
Pasted text and console output now display tabs properly,
the tab still uses spaces for indenting though.
Adds the ability to set text cursor position in the Console editor
by clicking with the mouse and many other expected operations such as
cut, select all & deselect on text entry.
Ref !108626
Using SDL's initialization logic, this is mainly a change for XDG
as LIBDECOR already required a configure event before accessing
the underlying XDG window.
While I didn't notice functional changes with this change window
flickering on startup remains an issue with some compositors
(KDE & river). Debugging these issues is simpler when both windowing
decoration systems work in a similar way & window configuration
is guaranteed to have run before the window is returned.
(via #xdg_surface_ack_configure).
Campbell Barton
Pratik Borhade
Hans Goudey
Harley Acheson
Sergey Sharybin
Julian Eisel
Leon Schittek
Philipp Oeser
Ray Molenkamp
Lukas Tönne
Alexander Gavrilov
Brecht Van Lommel
Jason Fielder
Jacques Lucke
Jeroen Bakker