[mercury-users] module graph path/4 implementation?

[doug.auclair at logicaltypes.com]

Dear all,

Is anyone else having problems using module graph for
anything other than small problems?  My problem is
as follows:

I have 63 nodes, each node has from 2 to 8 arcs
(averaging around 6 arcs).  When I do a path/4
I get a detstack#0 overflow for any path greater
than 5 hops.  Is it because my call:

solutions((pred(Ans::out) is nondet :-
              path(Map, From, To, Ans)), Paths)

    (where Map, From, and To are ground)

is a bad way of going about things?

If not, are there better data structures and
algorithms you have used successfully to travel
transitive nodes in a directed graph?

If path/4 is inefficient, would submitting one
based on dominators (PADL 2006, pp. 73+), or
perhaps ACO (/Ant Colony Optimization/, Dorigo and
Stutzle, MIT Press, 2004), be of use here?

Sincerely,
Doug Auclair

--------------------------------------------------------------------------
mercury-users mailing list
post:  mercury-users at cs.mu.oz.au
administrative address: owner-mercury-users at cs.mu.oz.au
unsubscribe: Address: mercury-users-request at cs.mu.oz.au Message: unsubscribe
subscribe:   Address: mercury-users-request at cs.mu.oz.au Message: subscribe
--------------------------------------------------------------------------