On Fri, May 15, 2015 at 6:09 PM, Ondřej Čertík <[email protected]> wrote: > Hi Lars, > > On Fri, May 15, 2015 at 4:56 AM, The Lartians <[email protected]> wrote: >> Hello, >> >> I'm a physics student from Göttingen and currently working on scientific >> simulation software in c++. I am currently using sympy in my python wrappers >> as it allows a convenient way for users to define and view simulation >> functions (they get compiled on-the-fly and are passed as function pointers >> to the simulation). However I've run into a few issues on the way. For >> example, simplify on small imaginary numbers yields 0 (which in my case can >> have a huge impact on the simulation). Additionally, integrals of piecewise >> functions often return bad results (if they evaluate at all). These bugs >> make it impossible for me to use sympy in good conscience with scientific >> projects. >> >> I was therefore wondering if a "reliable" version of sympy exists, meaning a >> version where all code known to contain bugs or produce bad results is >> excluded until fixed. Otherwise I would probably create a branch. > > Thanks for reporting the bugs below. The "reliable" version of sympy > exists, and it is the latest git master or a release. We try to make > sure that we never return a wrong result, but rather no result if we > cannot do it. > > As you said, if you found a result that is wrong, then indeed that > needs to be fixed as soon as possible. Your bugs are still young, 3 > days and 5 days, so I think it's not that bad that we didn't manage to > fix it yet. :) As far as I know, we didn't know about them before. > > If you know how to fix them, then please send us a pull request. If > you only fix it in your branch, then nobody else can benefit from your > fixes. > > If you find more problems, please report them all, and please do not > hesitate to write to this list with the high priority ones (i.e. > returning wrong results). > > Thanks again for letting us know about these problems, and if you know > how to fix any of them, please let us know.
I found what the problem was in https://github.com/sympy/sympy/issues/9398, and posted a possible solution: https://github.com/sympy/sympy/pull/9403 Now we just need to make sure that all tests pass and that it didn't break anything, but it fixes the problem. Ondrej -- 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 [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/sympy. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CADDwiVDCBSPD7gMzqHY%3DmPuD%2B_E0FFouxWX5tbSNAggSEBt7hw%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
