Dear all, Here is a short report on the current progress towards *making* *sage compatible with python3*.
The last few releases have seen a good progress, monitored in https://trac.sagemath.org/ticket/26212 There remains now 414 files having failing doctests for a total of 3206 failing doctests. Among these files, *only 3 files have more than 100 failing doctests*, and 122 files have exactly one failing doctest. It is not easy to describe the remaining problems. There are lots of small issues, and at least one large issue about sorting vertices in graphs, where some work is being done right now. Another issue is related to something that would change all bases in the spaces of modular symbols, therefore changing many doctests that have hardcoded the choice of basis. This is ticket https://trac.sagemath.org/ticket/25301 for those interested in this matter. One concern now is to start trying to prevent any regression to happen (because this is already happening). Should we require people to check that they do not introduce more failing doctests with python3 than the existing ones ? Could somebody provide a buildbot so that at least the python3 build is checked ? The documentation still does not build, currently because of a regression in the manifolds folder, but mainly because of some problems with sphinx, that apparently is not seeing some of the classes. See https://trac.sagemath.org/ticket/26449 and https://trac.sagemath.org/ticket/26522 Sage is now working quite well with python3, even if not perfectly. But more work (not so easy) is needed to finish the task. Help is welcome. Frédéric -- 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.
