There is no such thing as 'bookend keys'. It would be completely detrimental to mandate keys to be placed at any specific point on all channels. The only thing that matters is the distance between…
Why is this change necessary? In which case do you expect this to return
falsein the future? Without this info, thereturn trueseems entirely unnecessary.This question is still…
This seems to miss the point: there are already methods for keyframing a property from the…
810c6279db
Merge branches 'pr-driver-var-default', 'pr-shapekey-locking', 'pr-shapekey-smoothing' and 'pr-shrinkwrap-tproj' into pending-combined
I changed the naming: init -> begin (i.e. named like original, but returns boolean); begin -> begin_post_undo.
There is no such thing as "nothing happens because of lock checks". Like I said, the whole point of those check functions not being those simple "is..." things you were suggesting is that they…
I split changes about this function into a separate commit to make them clear, and removed from_key.
Like I said, from_key was weird in that it returned null for index 0 instead of the…
I was thinking about reducing repetition by allowing to call the function on a null object without explicitly checking, but that's probably going too far, so removed this check.
You can't do this with a switch on enum. However I restructured the contents of the case branches.
- Rebased on main to update and introduce the separate refactor commit.
- Updated comments and messages.
- Renamed defval to fallback_value - note that this breaks the example file a bit.
It is NOT a functional change. It encapsulates repeated pairs of apply + free, because now operators also need to check the boolean return code and convert it into an operator status enum value.