[PATCH] limits of fresnel_c and fresnel_s



On 09/07/2013 12:00 AM, Stavros Macrakis wrote:
>  integrate(cos(x^2),x) gives a result in terms of erf which seems 
perfectly reasonable. Maxima doesn't guarantee any particular form of an 
integral.

Yeah, it's not unreasonable.
More important are the missing limits.

>
>  -s
>
>
>  On Fri, Sep 6, 2013 at 4:14 PM, John Lapeyre 
<lapeyre.math122a at gmail.com> wrote:
>
>  Hello,
>
>  I am attaching a patch for the limits of fresnel_c
>  and fresnel_s.
>
>  Also, we have
>  (%i14) diff(fresnel_c(x),x);
>  (%o14) cos(%pi*x^2/2)
>  (%i15) diff(fresnel_s(x),x);
>  (%o15) sin(%pi*x^2/2)
>
>  But integrating cos(x^2) and sin(x^2) give complicated expressions,
>  when they should probably return the fresnel integrals... unless there
>  is a switch I don't know about.
>
>  Tracing gives this:
>
>  1: (DIFFDIV ((%COS SIMP) ((MEXPT SIMP) $X 2)) $X)
>  1: DIFFDIV returned NIL
>
>
>  -- John
>
>
>  _______________________________________________
>  Maxima mailing list
>  Maxima at math.utexas.edu
>  http://www.math.utexas.edu/mailman/listinfo/maxima
>
>