|
Boost Users : |
Subject: Re: [Boost-users] Default weight map doesn't work for weighted grid graph in A-star search
From: Jeremiah Willcock (jewillco_at_[hidden])
Date: 2010-07-16 00:30:30
On Thu, 15 Jul 2010, W.P. McNeill wrote:
>
>
> On Thu, Jul 15, 2010 at 6:50 PM, Jeremiah Willcock <jewillco_at_[hidden]> wrote:
> On Thu, 15 Jul 2010, W.P. McNeill wrote:
>
>
>
> On Thu, Jul 15, 2010 at 5:45 PM, Jeremiah Willcock <jewillco_at_[hidden]> wrote:
> Â Â Â Â Â Â I just verified that everything builds and works with version 64058 on
> http://svn.boost.org/svn/boost/trunk. Â Is that the 1.44
> Â Â Â Â Â Â branch?
>
>
> No -- that's the trunk, and some of those changes will not get in until 1.45 (but BGL should be the same). Â What
> you need to test against is
> https://svn.boost.org/svn/boost/branches/release.
>
>    I also put my implicit graph example into a single source to make it easier to include in the examples
> directory. Â
>
>
> Good. Â I'll check it over again and most likely put it in soon.
>
>
> The A* maze solver example and the implicit graph example both build and run correctly using the Boost release
> tree
> (https://svn.boost.org/svn/boost/branches/release) revision 64062.
>
>
> I added both of your examples into the Boost trunk in r64063 (with a small change in r64064). Â Please check and see if you are fine
> with them. Â I put implicit_graph.cpp in as-is (except for changing a quote to a normal ASCII quote); I tweaked astar_maze.cpp a bit
> (changing the distance type to double and adding missing braces around an initializer) to fix compiler warnings. Â Thank you for
> your contributions to BGL.
>
>
> I picked up all your changes.  I'm surprised that astar_maze.cpp has warnings since I'm building it with -Wall -Werror g++ flags.
>
> The astar_maze.cpp in revision 64064 looks good.  Can you pick up the latest implicit_graph.cpp on github (commit 916fe4e05c0b75dfd48d).
> Â This fixes the non-ASCII quote, changes stringification for edges and (most importantly) removes a comment from the code which is
> incorrect and might be misleading to people using it as an example. Â After this change I'm going to declare this example complete.
>
> Thanks again for all your help.
That new version should be in r64065.
-- Jeremiah Willcock
Boost-users list run by williamkempf at hotmail.com, kalb at libertysoft.com, bjorn.karlsson at readsoft.com, gregod at cs.rpi.edu, wekempf at cox.net