2025-01-30 Core Module Meeting

Meetings
1

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

Ongoing Work

Embedded Linked ID

  • Jacques and Bastien worked on initial, super-basic implementation of the ‘write real data for linked IDs’ part of the project.
    • There is a WIP PR.
    • Should also be a Blender 5.0 target, as files generated that way are likely not fully compatible with current readfile code (some asserts are expected, at the very least).

Blender 5.0

  • Blender data & file format
    • IDProperty for system-/RNA-defined storage:
      • Meeting agrees to go for the ‘hard split’ option. Makes more sense to fully separate user-defined and system-defined data, rather than try to only mitigate the issue with the ‘soft-split’ option.
      • For versionning, all IDProps should be duplicated.
      • System-defined should be read-only when accessed from ‘dictionary’ syntax (i.e. as raw IDProps). This will be usefull e.g. for versionning of such data.
      • In the future, the split could allow adding more complex generic/useful data types to RNA-defined system IDProps, like e.g. CurveMapping.
    • Support >2GB buffers:
      • Jacques has a WIP PR, needs some more work.
      • Header update is agreed on.
        • Should also reserve 4 bytes for Blender version now, to prepare for Blender 10.0!
    • Implicit Sharing Accross IDs in blendfile:
      • Not clear if we have the resources for that one.
      • Might not be a super-high priority target, as benefits are not expected to be huge. More a ‘nice to have’ one.
  • Python/RNA API
    • Remove PARM_PYFUNC_OPTIONAL: This is not always used correctly in existing code, and should not be needed anymore since named parameters are now mandatory. Campbell will make a task.
    • Campbell would also like to do some cleanup/refactor in our python modules organization.

Misc:

  • Efforts to deduplicate C and C++ BLI_math API is on-going.
  • NDof: new patch with minor improvements, exisitng code/architecture is weird. The whole NDof topic could be a good candidate for the ‘input mapping’ project discussions?

Next Meeting

The next meeting will be on Thursday 13th February, 16:00 CET/Amsterdam time (your local time: 2025-02-13T15:00:00Z).

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

3 Likes