On Sun, Feb 25, 2024 at 07:01:22PM +0000, Gavin Smith wrote: > So it appears there is a problem with the counter not being reset properly > before the test runs, or sometime early on in the run. > > I'll try and do more investigation.
It appears that there was no effort made to reset the counters at all. If a parser run finished with one of the counters active, it would remain active for the next run. Then if the memory address of an element assigned to a counter happened to be reused, by pure chance, the counter would appear to refer to the new element, incorrectly. This bug (possibly only in the test suite, not in "real" runs of the program) has likely always been present (since I wrote the original code for parsetexi); the recently added test only allowed the bug to surface. I've committed a fix in commit 23da32ceb93ef7495 (2024-02-25).
