Skip to content

fix integer overflows when computing crossing matrices

Coming out of the discussion on #1488 (closed).

We should perhaps discuss whether an API break is worth it for this academic fix – we don’t have evidence that any graph that experienced these integer overflows is actually processable after fixing them. On the other hand, maybe we should rip the band aid off and do a more pervasive size_t migration for Graphviz 9.

Merge request reports