I'd prefer """Handler for the absolute value."""
(see PEP 257 [1]) over """handler for the absolute value""" but that's certainly not important. After all, most docstrings in sympy are not conforming to all these conventions. You could still leave the old information (before you made the docstrings consistent) there if you want. I'm not sure about putting all the logic in __init__.py. Isn't it supposed to have only the imports? Vinzent [1] http://www.python.org/dev/peps/pep-0257/ 2009/8/13 Fabian Pedregosa <fab...@fseoane.net> > > Vinzent Steinberg wrote: > > Again, a few comments. > > > > > args = map(refine, expr.args, [assumptions]*len(expr.args)) > > > > I'd use a list comprehension here (faster iirc and imho more readable) > > something like: > > > > args = [refine(a, assumptions) for a in expr.args] > > certainly, it also looks more natural. Fixed that and normalized > docstrings. It's in my branch. > > > > > The docstrings in refine.py are somewhat inconsistently formatted, but I > > don't care too much. > > > > All in all it looks very good. :) > > > > Vinzent > > > > 2009/8/13 Fabian Pedregosa <fab...@fseoane.net <mailto: > fab...@fseoane.net>> > > > > Ondrej Certik wrote: > > > On Wed, Aug 12, 2009 at 11:28 AM, Ondrej Certik<ond...@certik.cz > > <mailto:ond...@certik.cz>> wrote: > > >> On Wed, Aug 12, 2009 at 8:00 AM, Vinzent > > >> Steinberg<vinzent.steinb...@googlemail.com > > <mailto:vinzent.steinb...@googlemail.com>> wrote: > > >>> 2009/8/12 Fabian Pedregosa <fab...@fseoane.net > > <mailto:fab...@fseoane.net>> > > >>>> Ondrej Certik wrote: > > >>>>> This still fails in the latest branch: > > >>>>> > > >>>>> bin/test sympy/concrete/tests/test_sums_products.py -k > > >>>>> test_composite_sums > > >>>>> > > >>>>> I just sent you a login info to my amd64 box offlist, please > > debug it. > > >>>> > > >>>> This can be fixed by: > > >>>> > > >>>> diff --git a/sympy/core/assumptions.py > b/sympy/core/assumptions.py > > >>>> index e8a9fd6..8220550 100644 > > >>>> --- a/sympy/core/assumptions.py > > >>>> +++ b/sympy/core/assumptions.py > > >>>> @@ -101,7 +101,6 @@ class AssumeMeths(object): > > >>>> 'complex -> commutative', > > >>>> > > >>>> 'odd == integer & !even', > > >>>> - 'even == integer & !odd', > > >>>> > > >>>> 'real == negative | zero | positive', > > >>>> > > >>>> sadly this makes other tests to fail. > > >>> Any idea why this is necessary? > > >>> > > >>>> I propose pushing patches 1-4 (after fixing Vinzent's > > comments), work on > > >>>> performance and then start removing the old system. > > >>> +1 > > >>> As long as we don't drop the old system, we don't depend on > > functionality > > >>> and performance of the new one. > > >> Please prepare patch series, that can go in, I will look at it. > > > > > > Any progress? Scipy09 is coming, so every day counts. > > > > yes, I attach the reworked patches. You can also pull from my master > > branch: > > > > git pull http://fseoane.net/git/sympy.git master > > > > or see them online: > > > > http://fseoane.net/cgi-bin/gitweb.cgi?p=sympy.git;a=summary > > > > > > Thanks, > > > > > > > > > > Ondrej > > > > > > > > > > > > > > > > > > > > > > > > > --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "sympy-patches" group. To post to this group, send email to sympy-patches@googlegroups.com To unsubscribe from this group, send email to sympy-patches+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sympy-patches?hl=en -~----------~----~----~----~------~----~------~--~---
