Re: [q-lang-users] Strong examples of Q equational programming wanted

[Greg B. <gr...@sl...>]

Albert Graef wrote:
> Have you actually tried the Prolog solution? 

    Yes.  But I haven't tested it extensively, so there could be an
error lurking in there.  And the previous program isn't quite as
functional as the Q version, since it doesn't automatically know how to
normalize integer arithmetic, but a couple of extra lines could cure
that...

    simp(X*Y,Z) :- integer(X),integer(Y), Z is X*Y, !.
    simp(X+Y,Z) :- integer(X),integer(Y), Z is X+Y, !.
 
> It looks to me as if those rules only apply to the toplevel terms, so
> some recursive descent would be needed. (In Q this is implicit because
> of the evaluation strategy.)

    Hmm, I don't know if I see what you are saying.  The simplification
routine is interleaved with the recursive differentiation relation.
Every time we reconstruct a term, we see if there is a simplification
that applies.  And I don't think any of the current reconstructions
require more than one simplification to get rid of the extra 1 and 0
terms.  Of course, if we tried to implement a fancier CAS this might not
hold. 

Maybe you could provide an example of what you mean?

Greg Buchholz