I haven't looked at it in a while. How easy is it to support the more complicated things like metaclasses with a common code base? Has anyone compiled a list of the trickiest things to support in a common code base? Except for the things involving strings and standard library stuff, there's a pretty good chance we are using just about anything that would be on such a list.
On the other hand, what are the chief benefits? Using 2to3 doesn't seem to be that much work, especially now that we've got all the kinks worked out. The biggest issue that I know of is that the distribution tools are not really designed to work well with separate tarballs for Python 2 and Python 3. We have to trick pip into doing the right thing by uploading identical -py32 and -py33 tarballs, and there doesn't seem to be a way to make it work at all with easy_install (https://code.google.com/p/sympy/issues/detail?id=3511). All the other Python 3 issues I know of are independent of whether we use 2to3 or a unified codebase (like the inability to make Windows installers in Python 3, see issue 1235). Aaron Meurer On Sun, Apr 14, 2013 at 7:56 PM, Matthew Brett <matthew.br...@gmail.com> wrote: > Hi, > > On Sun, Apr 14, 2013 at 5:03 PM, Aaron Meurer <asmeu...@gmail.com> wrote: >> There's also the argument that was raised the last time we talked >> about this, which is that Python 2.5 is no longer supported at all by >> the core Python, even for security updates. >> >> By the way, the App Engine supports 2.7 now, so that is less of an >> issue (though to be sure, we haven't even been successful in porting >> SymPy Live to 2.7: https://github.com/sympy/sympy-live/pull/65). >> >> I think the fact that the rest of the scientific core stack---numpy, >> scipy, IPython, probably most others---no longer support Python 2.5 >> means we should probably follow suit. So to me, at this point, the >> only question is if any of the 2.5 features are important enough that >> we should drop support immediately, or if we should get one more >> release out first. I don't think it's worth it to support it beyond >> one more release (especially at the rate we release). > > Aaron - you've said before I think that you think a common 2 / 3 > codebase is not practical at the moment. Do you still think that? > Can you say why? It's relevant to this discussion only because a > common codebase would benefit so much from Python >= 2.6. > > Cheers, > > Matthew > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to sympy+unsubscr...@googlegroups.com. > To post to this group, send email to sympy@googlegroups.com. > Visit this group at http://groups.google.com/group/sympy?hl=en-US. > For more options, visit https://groups.google.com/groups/opt_out. > > -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to sympy+unsubscr...@googlegroups.com. To post to this group, send email to sympy@googlegroups.com. Visit this group at http://groups.google.com/group/sympy?hl=en-US. For more options, visit https://groups.google.com/groups/opt_out.
