Ok, At the moment, my code is not boostified, but I've uploaded it to http://groups.yahoo.com/group/boost/files/equiv_rel.zip It has a simpler interface than either your implementation or the incremental components implementation, but may be less powerful/general. I'm using union-by-size, rather than -by-rank. Is there a reason to use rank instead?
Hope this is useful, Jeffrey On Sun, 19 Jan 2003 15:31:59 -0500, Jeremy Siek wrote: > Hi Jeffrey, > > I think so. > > union/find is used in the Boost Graph Library, and there is an > implementation of it, in boost/pending/disjoint_sets.hpp. The reason it is > in pending is that I've never bothered to write docs and submit it to > boost. Anyways, it would be interesting to compare the interface and > implementation to what you've got. And if you're interested in championing > the addition of union/find to boost I'd be behind you. > > Cheers, > Jeremy > > On Thu, 16 Jan 2003, Jeffrey Yasskin wrote: > jyassk> I have a small class that uses the union/find algorithm to > jyassk> implement a dynamic equivalence relation. With some work, it could > jyassk> probably also be used to iterate through the equivalent partition. > jyassk> > jyassk> Would this be a good thing to add to boost? > jyassk> > jyassk> Jeffrey Yasskin _______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
