On Thu, Mar 5, 2009 at 12:32 PM, Barton Willis <willisb at unk.edu> wrote:
> In the context of hypergeometic functions, I've been thinking about
> this problem too. For every identity in Abramowitz and Stegun and
> Gradsteyn-Rhyzik do we want to invent a name for the rule? What would
> that look like and how useful would it be? How could we organize this
> so that it is useful and easy to use?
Hmm, I like the idea of making a look-up table of identities,
identified by some abbreviation like "A&S 1.2.3" (I think just
the number is not enough).
I've tinkered with some macros to make rules more
readable e.g. foo(x, y) --> bar(y) - bar(x) when (bazp(x) and quuxp(y))
(stuff following "when" expand into matchdeclare's and the
"-->" expands into a rule definition).
So we could have stuff like
identities["A&S 1.2.3"] : foo(x, y) --> bar(y) - bar(x) when (bazp(x)
and quuxp(y));
and then
expand_foo (e) := apply1 (e, identities["A&S 1.2.3"], identities["A&S 1.2.7"]);
There's a certain amount of wishful thinking here,
but maybe this can provide some kind of inspiration.
Robert Dodier