The problem is always the same : some Sage graphs are created as instances of BipartiteGraph (*), and even though some methods may not need to modify it they copy it and work on the copy, or well....
The point is : it must be the 5th bug report I see whose answer is "You used a BipartiteGraph without knowing, and the answer is that it is not the graph class you should use" Hence a good workaround --> cast as Graph. What do you think of *never* returning BipartiteGraph instances through the graphs.* methods ? I think that's one think the user should explicitely ask for, and not get without knowing. Nathann (*) Mainly it wants to *keep* the graph bipartite. And it's even worse than that : if the graph stays bipartite but the 2-coloring changes during operations like edge removal/addition, it will scream too. -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org