[ 
https://issues.apache.org/jira/browse/IVY-896?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12628932#action_12628932
 ] 

Nascif Abousalh-Neto commented on IVY-896:
------------------------------------------

Hi Xavier,

You're welcome! And the T-shirt is a great idea, fun that you remembered! :-)
I also added some Gython scripts to the GUESS site that were quite useful when 
dealing with Ivy-generated graphs: 
http://groups.google.com/group/guess-discuss/files

You can link to them in the documentation if you add this feature to Ivy. I 
will attach here another file with scripts that are Ivy specific (like removing 
the all-caller node introduced by report and repreport).

> Add GDF as a report output format
> ---------------------------------
>
>                 Key: IVY-896
>                 URL: https://issues.apache.org/jira/browse/IVY-896
>             Project: Ivy
>          Issue Type: New Feature
>          Components: Core
>    Affects Versions: 2.0.0-beta-1
>            Reporter: Nascif Abousalh-Neto
>            Priority: Minor
>             Fix For: unspecified
>
>         Attachments: gdf.xsl, j1.jpg, sp.jpg
>
>
> We have been dealing with some very large graphs here (thousands of nodes, 
> tens of thousands of edges) and the information density is so high that the 
> graph visualization tools supported by Ivy, GraphViz and yED, were not up to 
> the task.
> Then we found out about GUESS (http://graphexploration.cond.org/), which 
> takes a novel approach which combines visualization support (using JUNG, an 
> excellent library in itself) with a DSL for graph manipulation based on 
> Jython. This way a user can issue commands and run scripts that manipulate 
> the graph contents with immediate visual feedback. We have been using it and 
> has proven to be a great match for our use cases.
> GUESS claims to suport graphml but it was not able to parse the graphml 
> report generated by Ivy. So I coded an XSL that creates GDF from the Ivy 
> resolution report XML. Besides working :-) this method keeps the relevant 
> (but non-visual) attributes from the original report in the GDF as well. 
> These can be later used in GUESS to group nodes, create convex hulls, 
> calculate metrics, and so on.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to