On Wed, Nov 14, 2012 at 7:35 PM, Alan Watson <[email protected]> wrote:

> There may well be better ways to say this. Alternatively, I think one
> could say that the leafs are compared by eqv? (since structure has been
> unfolded).
>

Almost.   Not all leaf-less graphs are equal? to each other.     I think
that all leafless graphs consisting only of pairs are,  but once you start
interspersing vectors with pairs, this should affect equal?-ity.   Consider
"paths" which are sequences of operations like car, cdr, (vector-ref <node>
0), (vector-ref <node> 1), ....   For equal? graphs, a path walked in one
should be walkable in the other.   If you arrive at a leaf, then the leaves
are string-equal?, bytevector-equal?, or just eqv?

This is my mental model, but someone please correct me if I'm mistaken.
_______________________________________________
Scheme-reports mailing list
[email protected]
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports

Reply via email to