Thanks Matthew. The matplotlib import alone didn't seem to be the fix although I can't quite figure out what is going on. I'll test out some of the methods you suggested when I get a chance later today. Cheers!
On Wednesday, December 12, 2018 at 4:37:05 AM UTC+11, Matthew Woehlke wrote: > > On 10/12/2018 21.21, Wilfred Tyler Gee wrote: > > This appears to be due to the importing of the `pyplot` (i.e. `from > > matplotlib import pyplot as plt`) module in a non-interactive > environment. > > I will try to create a minimal example and then report it. > > A core dump sounds like the crash is in C[++] code rather than Python > code (which is also why -P is useless; you'd need to run in *gdb*, not > pdb). Presumably matplotlib has some native code, and something is going > wrong. > > I often see issues like this due to library conflicts. It is probably > not "a non-interactive environment" that is causing your problem, but > something else that was previously imported. > > If possible, I would try running your build under valgrind: > > $ type -P sphinx-build > /path/to/sphinx-build > $ valgrind python /path/to/sphinx-build . > > If it's a library conflict, you *probably* won't get a very useful stack > trace, but sometimes just seeing what libraries show up in the trace is > enough of a clue to guess what is going wrong. > > (You can also replace `valgrind` with `gdb --args`... in that case, keep > an eye on what libraries get loaded, looking for anything suspicious, > especially loading different versions of the "same" library, or loading > libraries that should go together from different places.) > > -- > Matthew > -- You received this message because you are subscribed to the Google Groups "sphinx-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/sphinx-users. For more options, visit https://groups.google.com/d/optout.
