This sounds reasonable, I didn't see a real use for defining TupleSets that meant direct failure before (I just assumed that that would be a modeling error).
It won't be in 3.0.0, but it will be added after 3.0.0 is released. Cheers, Mikael On Tue, Mar 10, 2009 at 9:14 AM, Malcolm Ryan <[email protected]> wrote: > I use a TupleSet to represent the edges of a graph to constrain paths. > It is certainly possible to have a graph with no edges. When I go > looking for paths in that graph, I want failure. What I _don't_ want > is for the code to crash. > > Malcolm > > On 10/03/2009, at 6:16 PM, Mikael Zayenz Lagerkvist wrote: > >> On Tue, Mar 10, 2009 at 2:20 AM, Malcolm Ryan <[email protected] >> > wrote: >>> On 09/03/2009, at 8:55 PM, Christian Schulte wrote: >>>> The attempt to finalize an empty tuple set will also throw an >>>> exception but >>>> most likely only in 3.0.1 and not 3.0.0. >>> >>> Is that the intended behaviour? Having an empty tuple set seems >>> perfectly legitimate to me. >> >> An empty TupleSet is not a meaningful object to use for a constraint, >> since it would mean direct failure. >> >> Cheers, >> Mikael >> >> -- >> Mikael Zayenz Lagerkvist, http://www.ict.kth.se/~zayenz/ > > > _______________________________________________ > Gecode users mailing list > [email protected] > https://www.gecode.org/mailman/listinfo/gecode-users > -- Mikael Zayenz Lagerkvist, http://www.ict.kth.se/~zayenz/ _______________________________________________ Gecode users mailing list [email protected] https://www.gecode.org/mailman/listinfo/gecode-users
