[Boost Graph] Two-graphs common spanning tree problem: generic algorithm (with concept checking, of course)
Hi all, my name is Michele Caini, I write from Italy, Florence, and this is my first mail on this list. I'd like to propose a new generic implementation of the two-graphs common three algorithm, proposed first by Grimbleby and then modified by Schach, that made a revolution on the first idea basing his algorithm on the solution for the single spanning three problem proposed and developed by Mint, Read and Tarjan. This algorithm is mainly used in symbolic circuit analysis, and I develop a generic solution for my symbolic solutor: QSapecNG, Qt-based frontend for symbolic circuit solution. If Boost Graph library mantainers are interested on publishing this algorithm with the next release of library, I'm happy to submit it, and to work on to optimize the code. Thanks to all, Michele ps: sorry for my English. :-)
On Sun, 31 Jan 2010, Michele Caini wrote:
Hi all, my name is Michele Caini, I write from Italy, Florence, and this is my first mail on this list.
I'd like to propose a new generic implementation of the two-graphs common three algorithm, proposed first by Grimbleby and then modified by Schach, that made a revolution on the first idea basing his algorithm on the solution for the single spanning three problem proposed and developed by Mint, Read and Tarjan.
This algorithm is mainly used in symbolic circuit analysis, and I develop a generic solution for my symbolic solutor: QSapecNG, Qt-based frontend for symbolic circuit solution. If Boost Graph library mantainers are interested on publishing this algorithm with the next release of library, I'm happy to submit it, and to work on to optimize the code.
We are always interested in new algorithms for BGL. The code would need to be changed to be as generic as possible and Boost-compliant in naming conventions and such, but I can make suggestions and review code for that. You would need to have the code available under the Boost Software License to put it into BGL. If you have that, please attach the right copyright and license statements to your code and send it to the mailing list (I think actually the Boost developer list <boost@lists.boost.org> rather than boost-users). Thank you for your offer. -- Jeremiah Willcock
Il 03/02/2010 19:48, Jeremiah Willcock ha scritto:
On Sun, 31 Jan 2010, Michele Caini wrote:
[...]
I'd like to propose a new generic implementation of the two-graphs common three algorithm, proposed first by Grimbleby and then modified by Schach, that made a revolution on the first idea basing his algorithm on the solution for the single spanning three problem proposed and developed by Mint, Read and Tarjan.
[...] If Boost Graph library mantainers are interested on publishing this algorithm with the next release of library, I'm happy to submit it, and to work on to optimize the code.
We are always interested in new algorithms for BGL. [...]
-- Jeremiah Willcock
Hi, Jeremiah. I'm also developing the old alternative, the grimbleby version of the algorithm, so we'll can propose all of them (for comparision, of course). I'll write on boost developers list as soon as the algorithms will be developed, don't worry about. Thanks a lot, Michele
participants (2)
-
Jeremiah Willcock -
Michele Caini