Very small vertices (length <= 2 or 3 iu) are frequent in filled areas, and
create false positive detections because one cannot calculate a meaningful
orientation of these vertices.
Mesa does not support adaptive swapping but will sometimes report that
it does. This prevents Mesa drivers from testing (and segfaulting) the
limits of their call
Fixes https://gitlab.com/kicad/code/kicad/issues/10069
(cherry picked from commit 100cf757fe)
1) Unify metrics between stroke and outline fonts
2) Unify handling of rotation and mirroring
3) Bug fixes in collision handling
4) Use of VECTOR2I (instead of VECTOR2D) for world-coordinates
5) Generate outline font overbar with rounded ends and fix rotation
bugs
6) Generate wxEVT_CHOICE events from FONT_CHOICE::SetFontSelection
7) Change text-item PNS SOLIDs to use compound shapes
A long-standing issue with multiple workarounds
(https://gitlab.com/kicad/code/kicad/-/issues/1877) prevented eeschema
printouts with multiple images or images in both the worksheet and the
schematic.
We worked around this by ordering the bitmap printing to be the first
step in the printing process. This partially worked but didn't get
everything correct. The primary difficulty in debugging this was that
the wxPrintPreview that we generate looked correct. Only the actual
printout was incorrect.
The fix is to explicitly clip bitmap drawing to the bitmap area,
regardless of how large the bitmap is.
(cherry picked from commit 69469b254a)
The Alt key is not usable by programs under Windows and Alt+Tab is
reserved on both MSW and Linux. Instead, we use the Shift modifier to
determine which preset to open
SHAPE_LINE_CHAIN will, by default, remove duplicate points when
appending. We don't want to do this when constructing our polygons in
symbol editor, so we need to explicitly call the routine without
deduplication
Fixes https://gitlab.com/kicad/code/kicad/issues/10289
(cherry picked from commit 62f8603353)