Out of cycle dataset releases

From PCGen Wiki
Revision as of 02:21, 30 July 2008 by Eddyanthony (talk | contribs) (New page: As of PCGen version 5.14.0 packages can be created which can be easily installed via the Tools/Install Data menu. This means we can now develop datasets out of cycle which can be downloade...)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

As of PCGen version 5.14.0 packages can be created which can be easily installed via the Tools/Install Data menu. This means we can now develop datasets out of cycle which can be downloaded separately for use with the current production release. The standards and processes for developing these sets are detailed below.

For the file naming convention for these sets is as follows:

<pcgenminversion>_<pubname>_<setname>_<subset>_<version>

Example:
5141_paizo_riseoftherunelords_advset5_1.0
5140_paradigmconcepts_forgedinmagic_1.2

If this would make the name really long we can abbreviate to something sensible. For the version number (of the set) the first release is 1.0, bug fixes will increment the right decimal (1.1) and significant FREQs increment the left decimal (2.0). When a new release replaces an older version we will remove the older version unless the target PCGen version has changed in which case we'll leave the older version available for those who don't want to upgrade.

In SVN we'll create a directory in the content directory called outofcycledatasets, this is in the trunk but outside of the release directory. The directory for each set will be named with the convention above minus the release number at the end, for example:

5141_paizo_riseoftherunelords_advset5
5140_paradigmconcepts_forgedinmagic

Each datasets directory should have the structure of the release as well as the install.lst file, the install.lst file should have the release version number in it somewhere.

We should try and avoid having duplicate copies of the set, if the set does not need to use code from the trunk that should not be a problem, the set can stay in outofcycledatasets until near the end of the cycle and then they can be moved into the alpha data folder and updated so that it will be released with the new production release.