Mailing-List: contact river-dev-help@incubator.apache.org; run by ezmlm
Precedence: bulk
Reply-To: river-dev@incubator.apache.org
Received-SPF: neutral (athena.apache.org: local policy)
Message-ID: <4D3AA499.2000001@zeus.net.au>
Date: Sat, 22 Jan 2011 19:34:17 +1000
From: Peter Firmstone <jini@zeus.net.au>
User-Agent: Thunderbird 2.0.0.14 (X11/20080531)
MIME-Version: 1.0
To: river-dev@incubator.apache.org
Subject: Re: Dependency Analsys
References: <4D395ABD.1020401@zeus.net.au>
	<AANLkTi=GVVs6q6YRuOvVyYp3ZcMqkrJMg6jZaEKE4ia-@mail.gmail.com>
	<4D3A9DC5.1030200@zeus.net.au>
 <AANLkTi=8_SqtR7Do229hWpv9k4qJkk1Lr+2BMObsg96m@mail.gmail.com>
In-Reply-To: <AANLkTi=8_SqtR7Do229hWpv9k4qJkk1Lr+2BMObsg96m@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Ok, sounds like a good experiment, the interesting thing about River 
packages, is they tend to originate from multiple jar files.

The code's been in River since the last release, since ClassDep depends 
on it, I've created a jar locally that allows you to run classdepend 
from the command line.  I think I'll upload the changes so anyone 
interested can play around with it too.

The programmable API is more powerful than the command line api, perhaps 
classdepend needs a dynamic graphical api.  JTree?

Cheers,

Peter.

Dan Creswell wrote:
> How about we trim it down by considering only inter-package dependencies or
> similar to start with?
>
> Then maybe separate graphs for each package?
>
> On 22 January 2011 09:05, Peter Firmstone <jini@zeus.net.au> wrote:
>
>   
>> Thanks Dan & Patricia for your suggestions,
>>
>> I'm still struggling with the graph output, it's a lot of data, perhaps too
>> much, the PNG turned out to be 30MB, for all dependency links in River,
>> originating from PersistentOutriggerImpl.
>>
>> I've attached the text file used to generate the graph.
>>
>> I'll have to get back to you.
>>
>> Cheers,
>>
>> Peter.
>>
>>