BLI_assert(positions.size() == factors.size()
Maybe use std::pair
instead of the custom struct
? Don't know if it's something we tend to avoid.
Does BMesh not use the mask data? If so, the gathering should be removed.
Is passing this in as a lambda possible? It would save about ~100 lines where this seems to be the only real difference
nit: I think the shadowing here is potentially more harmful to readability than beneficial, I'd personally use running_sum
, sum
, acccum
, or something along those lines.
Not sure I know enough about this area to make a fix, seems like the imapaint.paint
runtime data hasn't been initialized yet when this versioning code is being called.
Relevant stacktrace
_BLI_assert_abort() (/home/skim/Projects/blender-git/blender/source/blender/blenlib/intern/BLI_assert.c:51)
paint_brush_set_default_reference(Paint * paint, const bool…
Broke in 7b0ea0f1b4309, FYI: @JulianEisel
I think that's an option, certainly. To me this case seemed to be something we should explicitly handle instead of conditionally asserting based on the state of the node data. I don't have a…
This is fixed as of the recent 4.3 builds with the brush refactor, I'm closing this as it is unlikely we will backport a similar change for earlier versions given the scope of changes needed..