MathGroup Archive 2009

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

Search the Archive

Re: patterns

  • To: mathgroup at smc.vnet.net
  • Subject: [mg101941] Re: patterns
  • From: nma <marxer at mec.li>
  • Date: Fri, 24 Jul 2009 06:10:03 -0400 (EDT)
  • References: <h41f5m$rin$1@smc.vnet.net> <h46p6l$e7b$1@smc.vnet.net>

On 23 Jul., 09:57, Albert Retey <a... at gmx-topmail.de> wrote:
> Hi,
>
> > This works fine without Verbatim or HoldPattern:
>
> > In[1]:= result = {11.2,
> >         {dUp[1]->0., dUp[2]->0., dUp[3]->0., dUp[4]->2.1,dUp[5]=
->0.,
> >          dUm[1]->0., dUm[2]->2.6,dUm[3]->0.5,dUm[4]->0., dUm[=
5]->0.,
> >          dLp[1]->0., dLp[2]->0., dLp[3]->0., dLp[4]->1.4,dLp[=
5]->0.,
> >          dLm[1]->0., dLm[2]->3.1,dLm[3]->1.5,dLm[4]->0., dLm[=
5]->0.,
> >          a[1]->6.75, a[2]->1.25, c[1]->1.65, c[2]->0.15}};
>
> > In[2]:= Cases[result[[2]], Rule_[(a|c)[_],_] ]
>
> > Out[2]= {a[1]->6.75, a[2]->1.25, c[1]->1.65, c[2]->0.15}
>
> and this does, too:
>
> Cases[result[[2]], _[(a | c)[_], _]]
>
> note that the way you used Rule, you are creating a named part of the
> pattern (whose name is Rule) that you don't ever use. So its name does
> not matter and no harm is done by using Rule as that name. If you'd use
> the name Rule in a possible RHS of the rule, you might create surprising
> results, though...
>
> The difference to other solutions is that it will also match different
> occurrences of a or c, not only rules, e.g. something like {c[1],5.6}
> which may or may not be intended...
>
> hth,
>
> albert

Hello

If you want to exclude something like {c[1],5.6} you can use:

   Cases[result[[2]], Pattern[dummy, (a | c)[_] -> _]]
   or short
   Cases[result[[2]], dummy : ((a | c)[_] -> _)]

The syntax of Cases is:
   Cases[{e1, ...}, pattern]
   or
   Cases[{e1, ...}, pattern -> rhs]

Therfore you can't use just
   Cases[result[[2]], (a | c)[_] -> _]

because this would fit the "Cases[{e1, ...}, pattern -> rhs]" syntax.

Best Regards
Norbert Marxer





  • Prev by Date: Re: Re: best way to save 3D plot as .eps for latex document?
  • Next by Date: Re: Re: Thoughts on a Wolfram|Alpha package for
  • Previous by thread: Re: patterns
  • Next by thread: Thoughts on a Wolfram|Alpha package for