From my testing, no, I wasn't able to encounter a situation where this was needed. Since we restore from the unode positions, I don't think we would need to selectively filter that above on line…
I can verify this on 4.2, but not on 4.2.1 - I suspect this is one of the recent crashes that was fixed.
@Michael_Macha - could you verify on your end that the 4.2.1 RC build [here](https://buil…
Ah, thats an interesting consideration, I hadn't thought of that. In that case, I think we would want to apply it to the multires version of the average_positions method too right?
I'll follow up on this afterwards; there's another place that apply_translations_to_shape_keys is used in the undo code, so I don't want to clutter this fix with that cleanup.
Yeah, that's fair. I was trying to think of more descriptive / "why" things to put here since this did cause a bug, but I agree that as is this is just duplicated info.
One minor comment, otherwise looks good, thanks for the cleanup!
Wouldn't it be better to do the multiplication once here at the very end?
Ah, I didn't realize that that thread had some discussion about constexpr as well as enum class. Thanks for the link. The logic makes sense, though I do still think it's worth clarifying in…