2025-04-10 Core Module Meeting

Everyone is welcome on the Google Meet linked below every other Thursday, alternating between 10:00 CET and 16:00 CET.

Present:

The meeting covers the core module topics (DNA/RNA, ID management, Depsgraph, Blendfile I/O, etc.).

Links

Updates on Blender 5.0 Projects for 4.5 Release

Blendfile Compression Enabled by Default in 5.0

More as a note for now, final decision is still pending further testing. Meeting has no objection about it so far.

ID Name Size Increase

Meeting agrees on the general idea of increasing ID names fized size to 258 bytes.

The initial PR currently also changes the name buffer sizes for other types like bones.

For internal-sub-ID data names (bones, modifiers, nodes etc.), it might be worth first trying to move them to dynamic strings, while also keeping current static fixed size strings writing in blendfiles for forward compatibility.

Bastien will ensure that we have a proper design task for this topic.

Brush Asset Library Question

For sculpt brush use of geometry nodes that’s in development, Hans raises the question of how to add local data-blocks to the node group. The meeting reiterated that the right approach is to copy the data-block into the brush’s library when it’s assigned. Drag and drop also needs to be handled. Eventually there should be some operator to update an assigned ID since they’re no longer the same, but that can be left for the future.

Next Meeting

The next meeting will be on Thursday 24th April, 10:00 CEST/Amsterdam time (your local time: 2025-04-24T08:00:00Z).

The provisional meeting agenda will be linked in the Core Meetings Overview thread before the meeting.

3 Likes

Here is the promised proof of concept for dynamically sized name buffers: #137310 - Constraints: support dynamically sized buffer for constraint names - blender - Blender Projects

Overall, it seemed to work quite well. I had to add a few functions that would likely also be useful when the same thing is done for other names. Obviously, it takes a little bit of effort to resolve all potential issues when turning an inline buffer into a pointer, but nothing was super unexpected yet.

1 Like