On Wed, Apr 24, 2013 at 12:05 PM, Louis Suárez-Potts <lui...@gmail.com> wrote:
> Hi
>
> On 2013-04-24, at 11:42 , Rob Weir <robw...@apache.org> wrote:
>
>> On Wed, Apr 24, 2013 at 3:28 AM, Andre Fischer <awf....@gmail.com> wrote:
>>> On 23.04.2013 22:58, Kay Schenk wrote:
>>>>
>>>> In my search to find out more about resource files (.src), I came across
>>>> the old Tools information on the web server, and specifically the project
>>>> dependency chart:
>>>>
>>>> http://www.openoffice.org/tools/project_dependencies.png
>>>>
>>>> Is this still valid?
>>>>
>>> No, it is not.  Some new modules are missing, for example apr, apr-util,
>>> coinmp, serf.
>>> We should create a new one.  This one has probably been created with
>>> graphviz.  Does anybody know of a better graphh visualization tool that
>>> produces prettier and/or better readable output?
>>>
>>
>> I use Gephi, and it is quite good:
>>
>> https://gephi.org/
>>
>> If we can get the raw data into a CSV file showing the dependencies,
>> that will import nicely into Gephi.
>
> Thanks! I hadn't been aware of this before. It's quite nice. I'm curious if 
> they'd (or we'd) be interested in devising an extension that could utilize it 
> as we wish. Yes, it's easy enough now, as is, but for those of us even lazier 
> than thou….
>

It is possible.  But if we wanted to add support I'd target the
GraphML format, which is understood by Gephi and other graph analysis
packages:

http://graphml.graphdrawing.org/

A simple project for a student could be an extension that exports
GraphML from a spreadsheet.

-Rob

>
> Louis
>>
>> For example:
>>
>> A,B
>> B,C
>> B,D
>>
>> This shows A depends on B, and B depends on C and D
>>
>> You can also repeat edges to give them greater weight.
>>
>> When graphing you can then scale the nodes so the ones that have the
>> highest in-degree are larges, i.e., the ones that most other modules
>> depend upon, etc.  You can also color code.  Of course, all this can
>> be done by graphviz as well.  The beauty of Gephi is it does these
>> calculations automatically,
>>
>> Regards,
>>
>> -Rob
>>
>>
>>> -Andre
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
>>> For additional commands, e-mail: dev-h...@openoffice.apache.org
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
>> For additional commands, e-mail: dev-h...@openoffice.apache.org
>>
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
> For additional commands, e-mail: dev-h...@openoffice.apache.org
>

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org
For additional commands, e-mail: dev-h...@openoffice.apache.org

Reply via email to