Cool. My motivation was that the URL loading method would track the latest changes very closely... maybe that was a little bit too close :)
So a common complaint about using device classification is that you have to constantly track and maintain new devices. If the client is configured to load via URL, that removes this burden from the user. Whenever the client is instantiated, it grabs the latest version matched data. Also, if you load from URL, you dont need the devicemap-data jar dependency, but it might be safe to include as a fallback. This stuff should probably be documented... I actually made a task to fix this cause I had an uneasy feeling putting the SVN url in the code: https://issues.apache.org/jira/browse/DMAP-41 Also, here is the branch for 1.0.1 changes: http://svn.apache.org/viewvc/incubator/devicemap/branches/devicemap-java-client-1.0.1/ I already updated trunk to 1.1.0-SNAPSHOT. ________________________________ From: Bertrand Delacretaz <[email protected]> To: "[email protected]" <[email protected]> Sent: Friday, July 18, 2014 8:18 AM Subject: Re: URLResource reading from svn.apache.org by default? On Fri, Jul 18, 2014 at 1:33 PM, Reza <[email protected]> wrote: > ...What we can do is this, keep 1.0.0 for voting only. If and when that > passes, > we can patch the url and immediately releases a 1.0.1 and that release will > be made available to the public... Works for me! -Bertrand
