finding subterms in a bigger term



On 9/18/07, Richard Fateman <fateman at cs.berkeley.edu> wrote:
>
> As it happens, this is done by    factcomb().  Try it.
>

Factcomb gives surprising (though correct) results in many cases.  Sometimes
a judicious combination of minfactorial and factcomb gives simple-looking
results, sometimes not.  Of course, what is "simple" is in the user's eye.

     factcomb((x+1)!/x!) => (x+1)!/x!
           (I would have expected x+1, which you can get with minfactorial)

     factcomb(  x^2 * (x-1)! ) => (x+1)! - x!
          (I would have expected x*x!;
           minfactorial of input leaves it unchanged;
           of output gets (x+1)*x!-x! )

     factcomb(  x!/x^2 ) => (x-2)! * (x-1)! / ((x-1)! + (x-2)!)
          (I would have expected (x-1)!/x , which no combination of
           minfactorial and factcomb gets)

This may be some well-defined quasi-canonical form, for all I know.
In Andre's case, it blows up the expression quite a bit, apparently not in a
useful way.

           -s