All regressions have been resolved now, and it seems to me like it's going to automatically migrate, but I'm not sure whether we missed the window to do so without a freeze exception. As far as I can see, the freeze date was February 19. I think the pikepdf upload that resolved the (false) regressions was February 20.
If possible, consider this to be a freeze exception request (if needed). Hopefully the above is adequate justification. 10.6.2 is a relatively small change from 10.5.0 in terms of functionality, but it adds a number of new interfaces to the API and, most importantly, adds some documentation and preprocessor symbols designed to ease the transition to qpdf 11.0, which will switch shared pointer implementations that may require source changes in some rare cases. I released qpdf 10.6 two weeks before freeze to ensure it would make the cut, but then there were delays because of pikepdf test failures that weren't real problems, just reliance on previously incorrect functionality from qpdf. Over 10.5, 10.6 adds a number of fixes to character encoding issues (which is what caused the pikepdf test failures) and also makes it possible to use the C API to do a number of things that could previously only be done with the C++ API. Bottom line: this is a much better version to be in an LTS release than 10.5.0. ** Summary changed: - qpdf 10.6.1 not syncing? + qpdf 10.6.2 not syncing (freeze exception?) -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1960924 Title: qpdf 10.6.2 not syncing (freeze exception?) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/qpdf/+bug/1960924/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
