commons-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Simone Tripodi (JIRA)" <>
Subject [jira] [Commented] (SANDBOX-335) [Graph] Graph Coloring: Backtracking algorithm
Date Fri, 01 Jul 2011 18:02:28 GMT


Simone Tripodi commented on SANDBOX-335:

Hi Marco!
thanks for yet another patch! :) I didn't see the algorithm implementation but I have anyway
few comments about code style:

 * I don't see the advantage of passing more generic graph instances in GraphUtils/Tests methods
when in the body are built only specialized Vertices/Edges, using the ones one trunk is more
canonical - and that's why we use generics here
 * I suggest you renaming methods in {{greedyColoring}} and {{backTrackingColoring}}, shortcuts
on method names are not friendly;
 * I don't see the reason why you modified the {{ColoredVertices}} class applying the Class
Adapter instead the Object Adapter as was in the origin; in that way users can potentially
modify stored data and that's not good at all.

Can you provide the patch applying that modifications?
Thanks in advance!!!

> [Graph] Graph Coloring: Backtracking algorithm
> ----------------------------------------------
>                 Key: SANDBOX-335
>                 URL:
>             Project: Commons Sandbox
>          Issue Type: Improvement
>          Components: Graph
>            Reporter: Marco Speranza
>            Priority: Minor
>         Attachments: GraphColoring-BackTrackingAlgo.patch
> Here is an little enhancement for graph coloring problem. I implemented an bruteforce
algorithm that uses the backtracking technique to find the exact solution for the m-coloring
problem (m is the maximum number of usable colors).  

This message is automatically generated by JIRA.
For more information on JIRA, see:


View raw message