MathGroup Archive 2008

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: SetDelayed

  • To: mathgroup at smc.vnet.net
  • Subject: [mg92892] Re: SetDelayed
  • From: Bill Rowe <readnews at sbcglobal.net>
  • Date: Fri, 17 Oct 2008 05:23:17 -0400 (EDT)

On 10/16/08 at 5:06 AM, shechtma at netvision.net.il (Itzhak) wrote:

>Why does Mathematica reply with:
>
>.."Integrand [(x-Cos[y])/(1+x^2-2*x*Cos[y])^1.5 is not numerical at
>{y}={1.5708}", to the following:

>Clear[f]
>f[x_]:=NIntegrate[(x-Cos[y])/(1+x^2-2*x*Cos[y])^1.5,{y,0,Pi}];
>Plot[Evaluate[f[x],{x,0,2}]

>How should it be written?

f[x_?NumericQ]:=NIntegrate[(x-Cos[y])/(1+x^2-2*x*Cos[y])^1.5,{y,0,Pi}]

This will prevent calling NIntegrate with something that cannot
be evaluated numerically. Note, there really is no need to
follow an expression using SetDelayed with a ";"

That is

f[x_]:=2 x

and

f[x_]:=2 x;

will evaluate exactly the same with neither producing an output
until used later.


  • Prev by Date: Import["http://.."] wackiness.... ?
  • Next by Date: Re: SetDelayed
  • Previous by thread: Re: SetDelayed
  • Next by thread: Re: SetDelayed