Blender Git Commits

GLog: Lower default logging severity to INFO

Before this change messages of ERROR and above were printed.
This change makes it so LOG(INFO), LOG(WARNING), LOG(ERROR)
and LOG(FATAL) will be printed to the console by default
(without --debug-libmv and --debug-cycles).

On a user level nothing is changed because neither INFO nor
WARNING severity are used in our codebase. For developers this
change allows to use LOG(INFO) to print relevant for debugging
information. Bering able to see WARNING messages is also nice,
since those are not related to debugging, but are about some
detected "bad" state.

After this change the LOG(INFO) is really treated as a printf.
Why not to use printf to begin with? Because it is often more
annoying to print non-scalar types. Why not to use cout? Just
a convenience, so that all type of logging is handled in the
same way. When one is familiar with Glog used in the area, it
is easy to use same utilities during development. Also, it is
easy to change LOG(INFO) to VLOG(2) when development is done
and one wants to keep the log print but make it only appear
when using special verbosity flags.

The initial reason why default severity was set to maximum
possible value is because of misuse of VLOG with verbosity
level 0, which is the same as LOG(INFO). This is also why
back in the days --debug-libmv was introduced.

Now there is some redundancy between --debug-libmv, --debug-cyles
and --verbose, but changes in their meaning will cause user
level side effects.

Differential Revision: https://developer.blender.org/D10513

Libmv: Avoid use of LOG(INFO) in solver

Usage of LOG(INFO) actually went against own guidelines in the
logging.h: the INFO is for messages which are to be printed
regardless of debug/verbosity settings.

Cleanup: Mention Correct Types In Comments

Cleanup: Missing const keyword.

Fix Crash: Access Unallocated Memory.

Cryptomatte: Flexible Definition of CryptomatteLayers.

Cryptomatte layers in Blender are predefined. Other render engines
might have other naming schemes. This patch will allow creation of
cryptomatte layers with other names. This will be used by D3959 to
load cryptomatte openexr files from other render engines.

EEVEE and Cycles still use our fix naming scheme so no changes are
detectable by users.

Fix Cycles build error without Embree on Windows

Don't use ssize_t, it's not available in MSVC.

UI: Correct output attribute name in proximity node

I made an incorrect comment in patch review that "Location"
should be used instead of position. "Position" is more appropriate
here since it refers to generic data rather than an object origin.
"Position" is also the name we chose for the attribute for this reason.

Build-system: Use C-linkage for internal interface header

That way the header can be included in C++ files.

Cleanup: move some drawing code into ed_draw.c

Move some drawing code from `area.c` and `ed_util.c` into `ed_draw.c`.
This is to support the new generic slider that wil be used in T81785.

No functional changes.

Reviewed By: #animation_rigging, #user_interface, Severin, sybren

Maniphest Tasks: T81785

Differential Revision: https://developer.blender.org/D9313

Add `--open-last` CLI argument that opens the most recent file

Add a CLI argument `--open-last` that opens the most recent file,
effectively doing the same as {key Ctrl Shift O}, {key Enter} after
starting Blender.

When there are no known recent files, print a warning and do nothing,
showing the startup file instead.

Note that this does not try to be smart about restoring the last Blender
session. It just opens the file from disk, as if the user had typed
`blender $(head -n1 ~/.config/blender/2.93/config/recent-files.txt)`.

There is also no smartness when that file cannot be opened; it behaves
exactly the same as typing the most recent filename on the CLI.

Reviewed by: mont29, campbellbarton

Differential Revision: https://developer.blender.org/D10563

Cleanup: Use LISTBASE_FOREACH and LISTBASE_FOREACH_MUTABLE macro

CMake/deps: Added new macOS arm64 build dependency to instructions

macOS arm64 has extra requirements.

Harden checks in datatoc_icon binary

The goal of the change is to perform check for attempts of icons
being overwritten on canvas. The check is based on checking original
coordinate of icons against all read icons. If there are two icon
files which have same original an error will be reported. The report
includes both file names to make it easier to troubleshoot.

This change will allow to early-on catch issues which we currently
have with the release environment: official Linux builds might have
different icon from Blender compiled locally. This is because the
order in which directory listing is traversed is not defined, so
it's like a race condition between two files to win the place in
the final canvas.

There is still possible improvement in the code to move more fields
into the context structure. This is beyond of goal of this change.

Note that before committing this change icons must be brought back
to their consistent state. Otherwise the build will fail.

Differential Revision: https://developer.blender.org/D9715

Work around T86015: Crash undoing in certain scenario (disabling Global Undo).

Simply disable advanced 're-use current blend data` process when loading
a memfile step and Global Undo is disabled, since there is no way to
ensure we have a proper 'differential' state in the stack then.

NOTE: this is a quick work-around to fix the crash, not a satisfying
solution by far (pretty sure there can still be crashes if you then
re-enable Global Undo afterwards e.g.).

Fix T86087: Assert when changing compositor node value or invoking tooltip.

Regression from rBde25b79ff5c4, this flag is inconsistent currently for
embedded IDs, needs to be sorted out, see T86119.

Fix T84658: Anisotropic BSDF - most modes not using Screen Space
Reflection

Anisotropic is not really supported in Eevee, but since code looks like
it is just intended to make it behave like glossy, it should function
like it too.

Seems like the internal calling from `node_bsdf_glossy` from
`node_bsdf_anisotropic` has swapped arguments.
Also: ssr_id is available for SH_NODE_BSDF_ANISOTROPIC as well (see
`ntree_tag_bsdf_cb`), so why not use it?

Maniphest Tasks: T84658

Differential Revision: https://developer.blender.org/D10547

Fix T86116: Failure to detect click events with modifier held

Click detection logic relied on the event queue not accumulating events
as newly added events at the end of the queue adjusted
`wm->eventstate->prev{val/type}` which was then compared with events
at the front of the queue.

This made Ctrl-Click fail to detect the click as releasing Ctrl
set the previous value before the previous event was handled.

In practice this only happened in heavy scenes,
updating 10 fps or lower. It also made automated tests fail that
accumulate events before handling them.

WM: prevent model keymap handling overwriting previous event values

`wmEvent.prevtype` & `wmEvent.prevval` were overwritten
when used in modal keymaps.

Now they are restored to their original values.

WM: set previous values for events in the event queue

This makes event's prevval and prevtype usable for events in the queue.

Previously they were unused, except as a hack for modal keymap handling.

This is needed to fix T86116, where the `wm->eventstate->prev{val/type}`
are set to values from events that have not been processed.