When applying the stroke-miterlimit, cairo is using the slopes of the lines after transformations are applied to figure out when to switch to bevel. I've patched this in cairo-path-stroke-polygon.c to fix it and to make it consistent with the way cairo-path-stroke-traps.c, cairo-path-stroke-tristrip.c and cairo-path-stroke.c all do it. Finally, I'm a) not a C guy and b) don't really know the cairo code so forgive me if any of this is off the mark! I've attached an example with the bottom 2 line un-transformed, and the top 2 being squashed.
Created attachment 130881 [details] test case
Created attachment 130882 [details] Wrong - before patch
Created attachment 130883 [details] Right - after patch
Created attachment 130884 [details] [review] patch
-- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/cairo/cairo/issues/196.
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.