Chris Yate <chrisy...@gmail.com> writes: > On Sat, 1 Oct 2016 at 12:50 Chris Yate <chrisy...@gmail.com> wrote: > >> On Sat, 1 Oct 2016 at 12:27 Phil Holmes <m...@philholmes.net> wrote: >> >> Have you narrowed the problem down to a specific new release? - e.g. it >> happens in 2.19.32 but not 2.19.31? >> >> >> Right. Very good point. >> >> I've started again with Gub, with the platforms I don't care about >> commented out and have run "make bootstrap". Perhaps this will work better >> than last time. In the meantime, a binary search of 2.19 to find when this >> error starts happening. >> >> Chris >> >> > Hi Phil, > > The culprit is 2.19.21.
Ah, that's likely to be not the real culprit but just the version where assertions started to be enabled by default on Windows (and other) builds. A number of assertions might reasonably be replaced by "programming errors": the main point of an assertion is that LilyPond does not know how to safely continue on an assertion failure. A "programming error" still indicates a "Huh?" moment but does not mean that continuing might be fatal. Typical assertions are that some pointer or data structure is valid before accessing them. Typical programming errors are unusual values. Of course, either warrant fixing if they occur. -- David Kastrup _______________________________________________ lilypond-devel mailing list lilypond-devel@gnu.org https://lists.gnu.org/mailman/listinfo/lilypond-devel
