matchdeclare matching functions or an alternative?

Sorry about being vague,

Thank you both. Both your advice is appreciated. However, I think Stavros
Macrakis captured exactly what I was after!



On Sat, Feb 23, 2013 at 3:39 PM, Stavros Macrakis <macrakis at alum.mit.edu>wrote:

> Your request isn't very clear, but I suspect that what you're looking for
> is something like
>
> matchdeclare(ff_xx, lambda([ex],not freeof('xx,ex)));
>
> which will match any subexpression containing the variable xx.
>
> You will have to be very careful in writing rules using a match like this.
>  In particular, you want to make sure that the xx's somehow disappear from
> your expression, otherwise you will have an infinite loop.
>
> Here's a little (meaningless) test:
>
>
> defrule(ff_xx_rule,ff_xx,subst('yy,'xx,ff_xx)^2+diff(ff_xx,'xx));
> apply1(xx^3/3,ff_xx_rule) => ((yy^6/9+yy^2)^2+2*yy)^2+2
>
>
> Is that what you had in mind?
>
>          -s
>
>
> On Fri, Feb 22, 2013 at 7:25 PM, Mike Valenzuela <mickle.mouse at gmail.com>wrote:
>
>> Hello,
>>
>> I am wondering if there is a way to use matchdeclare to do the following:
>>
>> matchdeclare(ff(xx), isfunction(ff,xx))
>> Where ff is supposed to be a function of xx.
>>
>> If it is possible, how?
>> If not what do you suggest as an alternative?
>> I am looking for write a simplification rule which allows an operator to
>> "propagate" through implicit/explicit operators. By implicit I mean that
>> ff(xx) could be something as simple as xx+1. By explicit I mean something
>> like sin(xx). This is for implementing a custom "chain-rule."
>>
>> Thanks in advance
>>
>>
>> _______________________________________________
>> Maxima mailing list
>> Maxima at math.utexas.edu
>> http://www.math.utexas.edu/mailman/listinfo/maxima
>>
>>
>