Jeremy Siek wrote:
greene> I find the filtered_graph potentially very useful for my greene> application except for one case. In some (not rare) greene> instances I need to be able to create a filtered_graph greene> that spans two graphs.
It seems that the right approach would be to first have something that "virtually" combines the two graphs, and then simply apply filtered_graph to the result. However, off the top of my head I don't know of a good way to "virtually" combine two graphs without lots of memory overhead. I'm open to ideas :)
Well, no one said this would be easy. :) Can you comment on the applicability of the subgraph class to this problem? Originally I skimmed the document and I thought that subgraph required a consistent indexing across graphs, therefore eliminating the possibility of using it to combine independent graphs. However, going over it again I read about "local" and "global" indices. I don't think there's an interface for combining two existing graphs into a subgraph tree (BTW the graphs in my case are truly independent -- they do not share any nodes or vertices), but I wonder if the subgraph _structure_ might be appropriate. I need to look into how that all works a little more. -Dave -- "Some little people have music in them, but Fats, he was all music, and you know how big he was." -- James P. Johnson