Re: Issue 1201 in sympy: better tan(x**2).diff(x)

Comment #9 on issue 1201 by asmeu...@gmail.com: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 The suggestion fo comment 2 is the same as https://code.google.com/p/sympy/issues/detail?id=1696#c7. This is easy to fix. -- You received this message because

Re: Issue 1201 in sympy: better tan(x**2).diff(x)

Issue 1201: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 This issue is now blocking issue sympy:1696. See http://code.google.com/p/sympy/issues/detail?id=1696 -- You received this message because you are listed in the owner or CC fields of this issue, or because

Re: Issue 1201 in sympy: better tan(x**2).diff(x)

Comment #8 on issue 1201 by smi...@gmail.com: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 With the FU routines of https://github.com/sympy/sympy/pull/1737 you can selectively rewrite something like 1/cos(x)**-2 TR16(1/cos(x)**2) tan(x)**2 + 1 -- You

Re: Issue 1201 in sympy: better tan(x**2).diff(x)

Updates: Cc: -mattpap Labels: Integration Comment #6 on issue 1201 by asmeurer: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 I think functions like tan(x) should have tan(x).diff(x) return sec(x)**2 or cos(x)**-2, and have a special

Issue 1201 in sympy: better tan(x**2).diff(x)

Comment #5 on issue 1201 by tan2tan2: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 Passing the result to trigsimp() yields the desired behavior. Maybe this is good enough for OP? In [8]: trigsimp(diff(tan(x**2),x)) Out[8]: 2*x 2 / 2\ cos

Issue 1201 in sympy: better tan(x**2).diff(x)

Issue 1201: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 New issue report by Vinzent.Steinberg: Failed example: tan(x**2).diff(x) Expected: 2*x*cos(x**2)**(-2) Got: 2*x*(1 + tan(x**2)**2) [20:11] ondrej well, tan(x).diff(x) [20:11] ondrej

Issue 1201 in sympy: better tan(x**2).diff(x)

Issue 1201: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 Comment #1 by ondrej.certik: The fix is trivial: $ git diff diff --git a/sympy/functions/elementary/trigonometric.py b/sympy/functions/eleme index cfff053..eab527c 100644 ---

Issue 1201 in sympy: better tan(x**2).diff(x)

Issue 1201: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 Comment #2 by mattpap: This change does break the integrator because Risch algorithm requires certain derivatives in a special form (derivative of tan(x) is a nice example). The solution is to add

Issue 1201 in sympy: better tan(x**2).diff(x)

Issue 1201: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 Comment #3 by ondrej.certik: Yes, we can discuss what derivative for tan(x) is simpler. But I think the risch integrator should not depend on the choice, right? we should check Mathematica and other

Issue 1201 in sympy: better tan(x**2).diff(x)

Issue 1201: better tan(x**2).diff(x) http://code.google.com/p/sympy/issues/detail?id=1201 Comment #4 by mattpap: But I think the risch integrator should not depend on the choice, right? As I said, the algorithm works this way. Similar assumption is done e.g. in limits / series code, which