[sympy] Refine away Piecewise using assumptions.

2014-10-24 Thread Björn Dahlgren
Consider the following short example: http://nbviewer.ipython.org/urls/gist.githubusercontent.com/bjodah/720617780e83d7c35797/raw/c297ecf76a8e26a711b7c7868839f2dfb221bc10/gistfile1.txt refine with Q.is_true(...) works ok for refining away Piecewise from a solution (although it's a bit picky, e.g

Re: [sympy] Refine away Piecewise using assumptions.

2014-10-24 Thread Aaron Meurer
On Fri, Oct 24, 2014 at 9:45 AM, Björn Dahlgren wrote: > Consider the following short example: > > http://nbviewer.ipython.org/urls/gist.githubusercontent.com/bjodah/720617780e83d7c35797/raw/c297ecf76a8e26a711b7c7868839f2dfb221bc10/gistfile1.txt > > refine with Q.is_true(...) works ok for refining

Re: [sympy] Refine away Piecewise using assumptions.

2014-10-25 Thread Björn Dahlgren
> > I looked for Q.is_false without luck, I also looked for an Inequality > class > > but found nothing. > > Wouldn't that just be ~Q.is_true? It seems to work for me. The != > class is called Unequality (to distinguish it from > "inequalities"). > > Yes, you are absolutely right - I always

Re: [sympy] Refine away Piecewise using assumptions.

2014-10-25 Thread Aaron Meurer
Python doesn't allow "not" to return a symbolic object, so we have to use ~ (which normally means bitwise not). Aaron Meurer On Sat, Oct 25, 2014 at 8:51 AM, Björn Dahlgren wrote: > >> > I looked for Q.is_false without luck, I also looked for an Inequality >> > class >> > but found nothing. >> >