Re: [boost] Graph: change in breadth_first_search()

From: Andrew Myers (andrew_at_[hidden])
Date: 2001-01-18 01:20:18

Thanks for the warning! - let us know when this reaches the next boost release

By the way, my vertex to vertex shortest path code is working a treat. Throwing
an exception in the visitor when reaching the desired vertex is reached has
proved an effective means of terminating the algorithm. I then use the
predecessors vector to reconstruct the path.

Jeremy Siek wrote:
> 
> Hmm, we still need an event-point for when the vertex is pop'ed from the
> queue (the old position of discover_vertex()). We'll name this
> examine_vertex().
> 
> So Andrew, you'll be able to use examine_vertex() instead of
> discover_vertex().
> 
> On Wed, 17 Jan 2001, Jeremy Siek wrote:
> jsiek>
> jsiek> CHANGE NOTICE: The point at which breadth_first_search()
> jsiek> calls the visitor's discover_vertex() will be changing so
> jsiek> that it agrees with the typical definition in graph
> jsiek> literature, which is that discover means when the vertex
> jsiek> turns from white to gray. My apologies for not getting this
> jsiek> right in the first place. The change has been checked into
> jsiek> boost cvs.
> jsiek>
> jsiek> I know this affects Andrew Myers, sorry! You'll probably
> jsiek> want to go back to stopping your search at the
> jsiek> finish_vertex() event.
> jsiek>
> jsiek> Jeremy
> 
> ----------------------------------------------------------------------
>  Jeremy Siek                        www: http://www.lsc.nd.edu/~jsiek/
>  Ph.D. Candidate                  email: jsiek_at_[hidden]
>  Univ. of Notre Dame         work phone: [phone]
> ----------------------------------------------------------------------

-- 
  Andrew Myers                                                      I-SiTE 
  Senior Software Engineer                                3D Laser Imaging
  Phone: [phone]                  
  Fax  : [phone]                         
  Email: andrew_at_[hidden]                   http://www.isite3d.com.au