Re: [boost] Re: BGL: what constitutes an undirected graph?

From: Jeremy Siek (jsiek_at_[hidden])
Date: 2001-02-01 21:58:46

On Fri, 2 Feb 2001 hankel_o_fung_at_[hidden] wrote:
hankel> --- In boost_at_y..., Jeremy Siek <jsiek_at_l...> wrote:
hankel> > 
hankel> > I'm guessing, but perhaps you are looking for some kind of guarantee
hankel> > like the following for undirected graphs:
hankel> > 
hankel> > for (tie(ui, ui_end) = out_edges(u, g); ui != ui_end; ++ui)
hankel> >   assert(source(*ui, g) == u);
hankel> > 
hankel> Yes, that's what I mean. Sorry for not writing concisely.
hankel> 
hankel> > This certainly should be made explicit.
hankel> So, what is the current status? Is the above property of out_edges
hankel> already a requirement (though undocumented) of BGL? Or is it just
hankel> a coincidence that the current implementations in BGL assume the
hankel> above?

I would have to say that the current status is that the above property is
already a requirement, and that this qualifies as a documentation bug.
I've already updated the descriptions in IncidenceGraph and
BidirectionalGraph, and added some text to the docs for adjacency_list.

Thanks for emailing about this!

Cheers,
Jeremy

----------------------------------------------------------------------
 Jeremy Siek                        www: http://www.lsc.nd.edu/~jsiek/
 Ph.D. Candidate                  email: jsiek_at_[hidden]
 Univ. of Notre Dame         work phone: [phone]
----------------------------------------------------------------------