Hmm... I'd forgotten about radexpand.
expr: sqrt(1-x^2)/(2-2*x^2) $
scanmap(factor,expr),radexpand:all;
gets you
-%i/(2*sqrt(x-1)*sqrt(x+1))
The %i might be a bit inconvenient, but is nonetheless correct.... I
wonder if we have any code that will "notice" that the result is real iff
-1<x<1 and therefore rewrite it as say
-1/(2*sqrt(1-x)*sqrt(x+1))
which is syntactically real.
-s
On Wed, Feb 1, 2012 at 05:40, Barton Willis <willisb at unk.edu> wrote:
> Maxima calls inte to do integrate((sqrt(1-x^2)*(x^2+1))/(2-2*x^2),x,0,1).
> If I recall,
> inte is one of the last functions integrate tries. Some other method (not
> sure which one)
> handles integrate((x^2+1)/(2*sqrt(1-x^2)),x,0,1).
>
> integrate((sqrt(1-x^2)*(x^2+1))/(2-2*x^2),x,0,1) is in the testsuite.
>
> --Barton
>
>
> _______________________________________________
> Maxima mailing list
> Maxima at math.utexas.edu
> http://www.math.utexas.edu/mailman/listinfo/maxima
>