Hitachi Vantara Pentaho Community Forums
Results 1 to 6 of 6

Thread: How are you handling prpt's in subversion?

  1. #1
    Join Date
    Apr 2007
    Posts
    2,009

    Default How are you handling prpt's in subversion?

    Hi,

    How is everyone handling prpt's in subversion/cvs etc? do you just accept that they're binary and therefore you wont be able to do any diffing?

    I was wondering about extracting the prpt into a subfolder of my source directory, and comitting both the prpt and the extract. The danger is that they get out of date of course! i.e. i commit a prpt without extracting it first...

    Thanks,
    Dan

  2. #2
    Join Date
    Mar 2008
    Posts
    364

    Default

    Good point...
    I unpack the reports into subfolders in a "src" folder that I check in.
    - I have the prpt's in a folder /design
    - I have an ant-target that extracts all subfolders of that folder into
    /src, which is then checked in.
    - When I check out /src i have an ant-target that repacks every exploded subfolder to /design.
    The problem is that you have to run design->src before you synchronize with the repository in order not to loose your own changes. Merging minor changes in the query or formatting of elements work ok. Merging big layout changes in xml does not work from my experience

  3. #3
    Join Date
    Mar 2008
    Posts
    101

    Default

    Good morning,
    I also have the same concern. When you say unpack the prpt, what command are you actually running to do that?
    Thanks,
    Peter

  4. #4
    Join Date
    Apr 2007
    Posts
    2,009

    Default

    Just use your favourite zip tool - it's just a bog standard zip format.

    Thanks for the replies - I like the src/ subdirectory idea.

  5. #5
    Join Date
    Mar 2008
    Posts
    364

    Default

    I am just using ant unzip and zip task to do it. Also some stuff from ant-contrib for looping and finding/repacking folders. And finally xmltask to give the reports the correct target datasource when I deploy to a target system and can't use a jndi name that is common to all databases.

    But I have to say the process is far from final and there are only two people working on the reports at the moment so I can't really say if it will be running smooth in the future.

    As more people are commiting reports I would also like to implement some ant driven junit-tests that run the reports with just the engine-core and some representative parameter-data, that the respective report-author has to provide with his submission. That way reports could be stress-tested against a certain target database automatically.

  6. #6
    Join Date
    Apr 2007
    Posts
    2,009

    Default

    mgie - your idea of ant junit tests for reports sounds very interesting.I'm sure that will be of interest to a lot of the community!

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Privacy Policy | Legal Notices | Safe Harbor Privacy Policy

Copyright © 2005 - 2019 Hitachi Vantara Corporation. All Rights Reserved.