This commit adds some first building blocks for the two operators to work modally based on mouse input. To make their function easier, two hotkeys are introduced, Ctrl+B for bevel and I for inset.
TODO:
After discussion with Campbell, we would like to add scale-style line indicators for the operators. This is already done for transform operators but a new interface for mesh operations may have to be written using pieces from that code since, strictly speaking bevel and inset are not exactly "transform" operators.
Also, a better input method for inset is needed and more options exposed. The method implemented right now uses mouse move for thickness and ctrl-mouse move for depth. These are calculated using the distance of the selection center in screen space and the mouse position. While that may work and prevents abrupt changes in values when switching from thickness tweak mode to depth tweak mode, it limits the magnitude of values that can be put into the tool especially in small or large scale.
Alternatives until a better method is written include:
* use relative offset (works but may give strange results)
* tweak manually after the operation.
Have to use a ugly hack, as for pose bones, rotscale transform matrix is not always the same as translate one... :/
Adresses feature request [#30979] snapping: "align rotation with the snapping target" and pose-mode.
Was a bmesh todo, main issue was with shape keys, now disabled any changes to
the shape key data layer, and disabled the option altogether when editing
non-basis shape keys.
when a single face in a loop is hidden, loop cut will subdivide edges on both sides.
creating an edge between the hidden faces.
without this workaround the edge ends up hidden and selected.
added a check in BM_mesh_validate() for hidden/selected elements.
- edgebisect bmesh operator used uninialized beauty field.
- BLI_join_dirfile could read from before the string bounds when passed an empty dir string.
- pransform could use an uninitialized projected coordinate (unlikely but possible)
- RNA_property_path_from_ID_check would compare against an uninitialized pointer when the path wasn't found.
also have bmesh walker use BM_edge_other_vert() utility function.
- access to a meshs editmesh before the pointer was checked to be a mesh.
- uninitialized memory use in transform (not a problem practically but nice to quiet the error in valgrind).
This option might be useful in some cases like keeping UV coordinated in kind of
"correct" state so sliding edge loop will also slide edges in UV. But for other
layers like shapekeys and vertex colors it's arguable it's actually needed.
The way it's currently done isn't actually acceptable because it doesn't take
flags like UVCALC_TRANSFORM_CORRECT into account and also makes extra unwanted
things like interpolating shapekeys which leads to propagating changes into
basis keys and also results with slide applied twice for relative keys because
offset is also applying on loading edit mesh.
Also discussed with Brecht who agreed with marking this things as TODO and
disabling for now.
This should fix issue reported in #30387: Edge Slide propogates to Basis Shape Key
* optimize BM_face_other_edge_loop to do about half as many iterations for quad heavy meshes, with ngons the gain is much more since searching around the entire ngon when the edge already stores its loop is silly.
... also nicer in cases where edge has no face users it avoids a loop on all face corners.
