The attached test case does not render properly. To see the bug, counters need to be implemented (http://bugzilla.opendarwin.org/show_bug.cgi?id=4980), but I have not yet committed that patch, so this bug is not yet present in the nightlies. I am filing it anyway in anticipation.
Created attachment 10769 [details] Test case
I think http://dbaron.org/css2.1/tests/imported-20050702/t1204-multiple-01-c.xht might be the same problem but ignoring the increment instead of the reset.
We seem to pass http://dbaron.org/css2.1/tests/imported-20050702/t1204-multiple-00-c.xht Does that mean this is fixed? if so, please close.
We still fail: css2.1/t1204-reset-00-c-o.html css2.1/t1204-reset-01-c-o.html css2.1/t1204-reset-02-c-o.html css2.1/t1204-increment-00-c-o.html css2.1/t1204-increment-01-c-o.html css2.1/t1204-increment-02-c-o.html At least the first 3 are probably related to this bug.
By using Safari 15.4, I am only able to reproduce on following test case. https://dbaron.org/css2.1/tests/imported-20050702/t1204-reset-00-c-o.xht and it is showing as Firefox Nightly 102 on macOS 12.4. Other test cases listed in Comment 4, now seems to pass.
<rdar://problem/94162231>