MathAction #294 Infinite recursion during integration

Topics

FrontPage
- IssueTracker
  - #294 Infinite recursion during integration <-- You are here.

The following input causes infinite recursion in rischNormalize

fricas

(1) -> integrate(simplify(D((log((x)+1)+a)/(2*((((x*(exp(exp(x)/2)))-(a*(x+exp(-x/2))))*(2/x))-3)), x)), x)

$\label{eq1}\frac{{x \ {\log \left({x + 1}\right)}}+{a \ x}}{{4 \ x \ {{{e}^{-{\frac{x}{2}}}}^{2}}\ {{e}^{\frac{{2 \ x \ {{{e}^{-{\frac{x}{2}}}}^{2}}}+ 1}{2 \ {{{e}^{-{\frac{x}{2}}}}^{2}}}}}}-{4 \ a \ {{e}^{-{\frac{x}{2}}}}}+{{\left(-{4 \ a}- 6 \right)}\ x}}$

(1)

Type: Union(Expression(Integer),...)

AFAICS goodCoef chooses to rewrite simpler kernel in terms of a more complex one. More precisely, rischNormalize is supposed to eliminate kernels which are algebraically dependent on other kernels. If a dependent kernel is found rischNormalize should eliminate it and recurse. So the recursion depth is limited by the number of kernels. However, in this example rischNormalize eliminates simpler kernel, but causes new kernels to appear.

elementary integration really should be one of the strengths... --kratt6, Sat, 03 Jun 2006 07:52:55 -0500 reply

Category: Aldor Library Compiler => Axiom Library Severity: normal => serious

Fixed in FriCAS --kratt6, Thu, 20 Dec 2007 01:17:58 -0800 reply

Status: open => fix proposed

... --gdr, Sun, 18 May 2008 13:40:08 -0700 reply

Status: fix proposed => fixed somewhere

no patch available

From wh-sandbox --alfredo, Thu, 28 Aug 2008 18:46:46 -0700 reply

http://axiom.svn.sourceforge.net/viewvc/axiom/branches/wh-sandbox/src/algebra/efstruc.spad.pamphlet?r1=263&r2=377

... --alfredo, Thu, 16 Oct 2008 12:56:27 -0700 reply

Status: fixed somewhere => fix proposed

... --test1, Thu, 08 Feb 2018 16:25:29 +0000 reply

Status: fix proposed => closed

Subject: (replying) Be Bold !!

	( 15 subscribers )