Re: Steve Langasek > Sorry, did you manage to get sensible a-c-c output? Otherwise, how did you > determine that there was a single symbol affected? The only compat_report > output I have shows *zero* symbols affected but also shows a bunch of > garbage output that makes me not trust it at all.
That was in a different sub-thread on this bug: For postgresql-16 (where libpq-dev comes from), the only symbol affected seems to be pqWaitTimed as indicated above. The good news is that the function is only used internally in libpq, and by no external user: https://codesearch.debian.net/search?q=pqWaitTimed&literal=1 https://github.com/search?q=pqWaitTimed&type=code (GitHub finds a bunch of instances, but these are all either PostgreSQL itself, or vendored copies of libpq.) So I think we can safely ignore the difference and not rename libpq5. (It has been named like that for 20 years and I really don't want to break compatibility there.) Christoph
