Article:
  Configuration Management in Java EE Applications Using Subversion
Subject:   Application which does this
Date:   2006-05-04 08:29:54
From:   michal.dobisek
Nice article, it's almost exactly describing what we are doing in our Polarion for SVN product (http://polarion.com/) - on the conceptual level. We use Subversion as the only source for the data and we do synchronize the Lucene index with changes in SVN to get fast searching. We also use some caching to achieve a good performance.


We don't use a database at all. The advantage of the single-source approach is that we are able to correctly handle even the changes on done externally directly in the Subversion - this is usefull for many usecases like


  • Live Documents - we can store the tracker items in Word or Excel documents inside the SVN and people can work with these either via our web-ui or directly by mofifying the documents in SVN using Word/Excel

  • Distributed work - people can work with the tracker using the stand alone FastTrack Eclipse plugin (for free http://polarion.org/), which is accessing the SVN directly. Server is watching the changes in SVN and gets ynced with them (but people can still uses FastTrack without Polarion server to cooperate on tracker in the same way as they do on sources - if the Polarion Server is present, they will get some additional features like email notifications, server-side building, etc).


Michal Dobisek
Polarion Software