On Tue, Mar 31, 2009 at 2:58 AM, daveloeffler <dave.loeff...@gmail.com> wrote: > > Both John Cremona and I have been having some serious problems with > "sage -t" in 3.4.1.alpha0. > > Doing "sage -testall" seems to work as it should, but doctesting a > single file fails horribly in many cases, even with files that weren't > changed at all between 3.4 and 3.4.1.alpha0: > > da...@groke:~/sage-current/devel/sage/sage/modular/modform> sage -t * > [ lots of error output ] > The following tests failed: > > sage -t "constructor.py" > sage -t "eisenstein_submodule.py" > sage -t "element.py" > sage -t "hecke_operator_on_qexp.py" > sage -t "space.py" > > None of these failed in 3.4; I did extensive development work on > modular forms based on 3.4 and everything passed. > > This seems to be something to do with trac #2129, "implement "sage -t" > for .spyx files", which was merged in 3.4.1.alpha0. It seems to have > broken doctesting for .py files at the same time.
Just out of curiosity are there any funny symlinks involved in your PATH? e.g., is sage-current a symlink to sage sage-3.4.1.alpha0? Since I just tried the above in my sage-3.4.1.alpha0 directory and it works fine, but I don't have any symlinks. IIRC, #2129 may involve looking at the PATH and SAGE_ROOT and comparing them, which might not work due to symlinks. I guess #2129 could be fixed by being more clever about detecting whether or not code is in the sage tree. Alternatively, we might have to have a different option to "sage -t" to test out of tree code. William --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-devel@googlegroups.com To unsubscribe from this group, send email to sage-devel-unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
