Filip Konvička wrote:
Just for your info: taking the algorithms from Gecode to Mozart is almost
impossible (unless you completely rewrite). This is due to the fact that
Mozart's CPI is quite limited compared to Gecode (no incremental
propagation, only idempotent propagators supported, no staging, eager
fixpoint computation, etc).
What a shameless advertisement! ;-)
I did not think that the *code* could be used, just the algorithms (the
papers on which the code is based are surely available somewhere, I've
seen some pdfs). Yet another option is that Mozart could use an external
Gecode-based constraint solver. :-)
FYI, our Colombian friends just started a project that aims at
integrating Gecode in Mozart. If this project succeeds, the
implementation of the constraint system should be moved out of Mozart,
and taken over by Gecode.
Cheers,
raph
_________________________________________________________________________________
mozart-hackers mailing list [email protected]
http://www.mozart-oz.org/mailman/listinfo/mozart-hackers
