Regression: Deleting a material slot in 3.4 assigns faces to the material below instead of the material above #103051

New Issue

unizen · 2022-12-09T01:17:44+01:00

unizen commented

2022-12-09 01:17:44 +01:00

Operating system: Windows 10

Version:
Broken: 3.4
Worked: 3.3 and earlier

Description:

Deleting a material slot will re-assign the faces previously assigned to that material to the material in the slot below, instead of the slot above, as in previous versions of Blender.

I see no good reason for this change, and it will likely confuse anyone who is used to combining materials the old way. So I'm assuming it is a bug and not an intended change.

Steps to reproduce:

On the default cube, create 3 material slots with 3 different materials.
Set a different viewport color for each material for easier visualization.
In edit mode, select all faces and assign to the material in the middle slot.
In object mode, remove the middle material slot.
Notice that the faces of the mesh have been re-assigned to the material in the bottom slot instead of the top slot

(doing these same steps in 3.3 or earlier you will notice the faces have been assigned to the material in the top slot)

Operating system: Windows 10 Version: Broken: 3.4 Worked: 3.3 and earlier Description: Deleting a material slot will re-assign the faces previously assigned to that material to the material in the slot below, instead of the slot above, as in previous versions of Blender. I see no good reason for this change, and it will likely confuse anyone who is used to combining materials the old way. So I'm assuming it is a bug and not an intended change. Steps to reproduce: On the default cube, create 3 material slots with 3 different materials. Set a different viewport color for each material for easier visualization. In edit mode, select all faces and assign to the material in the middle slot. In object mode, remove the middle material slot. Notice that the faces of the mesh have been re-assigned to the material in the bottom slot instead of the top slot (doing these same steps in 3.3 or earlier you will notice the faces have been assigned to the material in the top slot)

unizen commented

2022-12-09 01:17:44 +01:00

Added subscriber: @unizen

Omar Emara commented

2022-12-09 10:09:22 +01:00

Added subscriber: @OmarEmaraDev

Omar Emara commented

2022-12-09 10:09:22 +01:00

Changed status from 'Needs Triage' to: 'Confirmed'

Omar Emara commented

2022-12-09 10:09:22 +01:00

Looking into it.

Omar Emara changed title from ~~[3.4] Deleting a material slot assigns faces to the material below instead of the material above~~ to Regression: Deleting a material slot in 3.4 assigns faces to the material below instead of the material above

2022-12-09 10:09:47 +01:00

Omar Emara commented

2022-12-09 10:14:21 +01:00

Added subscriber: @HooglyBoogly

Omar Emara commented

2022-12-09 10:14:21 +01:00

I haven't bisected, but this is most likely caused by f1c0249f34.
Maybe just >= instead of >?

diff --git a/source/blender/blenkernel/intern/mesh.cc b/source/blender/blenkernel/intern/mesh.cc
index deaa9aec3f7..c8589faf16a 100644
--- a/source/blender/blenkernel/intern/mesh.cc
+++ b/source/blender/blenkernel/intern/mesh.cc
@@ -1353,7 +1353,7 @@ void BKE_mesh_material_index_remove(Mesh *me, short index)
   }
   MutableVArraySpan<int> indices_span(material_indices.varray);
   for (const int i : indices_span.index_range()) {
-    if (indices_span[i] > 0 && indices_span[i] > index) {
+    if (indices_span[i] > 0 && indices_span[i] >= index) {
       indices_span[i]--;
     }
   }

I haven't bisected, but this is most likely caused by f1c0249f34. Maybe just >= instead of >? ``` diff --git a/source/blender/blenkernel/intern/mesh.cc b/source/blender/blenkernel/intern/mesh.cc index deaa9aec3f7..c8589faf16a 100644 --- a/source/blender/blenkernel/intern/mesh.cc +++ b/source/blender/blenkernel/intern/mesh.cc @@ -1353,7 +1353,7 @@ void BKE_mesh_material_index_remove(Mesh *me, short index) } MutableVArraySpan<int> indices_span(material_indices.varray); for (const int i : indices_span.index_range()) { - if (indices_span[i] > 0 && indices_span[i] > index) { + if (indices_span[i] > 0 && indices_span[i] >= index) { indices_span[i]--; } } ```

Hans Goudey self-assigned this 2022-12-09 16:48:58 +01:00

Hans Goudey commented

2022-12-09 16:48:58 +01:00

Agh, such a small thing! Thanks for looking into it @OmarEmaraDev and thanks for the report @unizen.