I have a concrete and reproducible example where doing more doctests in a file breaks other, unrelated, doctests, see https://trac.sagemath.org/ticket/26856#comment:8
I don't even know where to start debugging this... On Thursday, November 29, 2018 at 12:58:34 PM UTC, E. Madison Bray wrote: > > On Thu, Nov 29, 2018 at 12:19 PM Jeroen Demeyer <j.de...@ugent.be > <javascript:>> wrote: > > > > On 2018-11-29 10:38, E. Madison Bray wrote: > > > But I sometimes get failures in tests depending on which order the > > > *files* were run in. That's what I'm talking about. > > > > But that's really unusual (and unrelated to --randorder). > > I guess I misunderstood what --randorder does. I thought it would > just randomize the order in which files were tested but apparently it > randomizes doctests within a file (and if so, what's the point of > that?) I think from docstring to docstring at least there should be > better test isolation. > > > Do you have > > *any* concrete example of such a failure? > > Not at the moment but I've had it before: For example, a test might > pass if I ran `./sage -t -a`, but fail if I just ran `./sage -t > src/sage/some_subpackage`. Usually when this has come up I didn't > bother to investigate further. > -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To post to this group, send email to sage-devel@googlegroups.com. Visit this group at https://groups.google.com/group/sage-devel. For more options, visit https://groups.google.com/d/optout.
