kleonc
kleonc
~Kinda reverts #67831 as now the checks within the group calls are done again by pointers~ _Edit_: #67831 was actually reverted in #69252 so now this PR just improves the...
Fixes #79096. | Before| After | |--------|--------| |||
`Transform2D::inverse`/`Transform3D::inverse` which were used in `Animation::subtract_variant` work only for rotation + translation, don't work for scale, skew, etc. This resulted in e.g. `PropertyTweener` incorrectly calculating the delta for affine transforms:...
Seen a user confused by `Transform3D transformC = transformA.InterpolateWith(transformB, 0.0f);` resulting in a `transformC` being way different than `transformA` (specifically `transformC.Scale` not matching `transformA.Scale`). Turns out there's a bug in...
Fixes #95574. To me the previous behavior seems bugged (hence marking it as such) but I'm not fully sure about it. @markdibarry Not sure if that's the behavior you suggested...
Fixes #95667. For the y-sorted subtree's root modulation is obtained once again after gathering y-sorted items, just like e.g. transform. In the similar manner it needs to be cancelled out....
### Tested versions v4.3.stable.official [77dcf97d8] and earlier (not a regression) ### System information N/A ### Issue description Title. In action:  ### Steps to reproduce 1. Create scene with single...
After reading the code I concluded that `uv_draw_ofs` is not affected by the `uv_draw_zoom`, hence should be multpliplied by it as well when snapping. Also simplified some code when reading....
Makes the transforms used for Y-sorting canvas items match the transforms used for such items in `RendererCanvasCull::_cull_canvas_item`, aka it accounts for physics interpolation and transform snapping. Fixes #92982. This is...