- Lucene index is now written to disk. One additional tweak (noted in the source code) to this would be for the server to have a properties file to specify where the lucene index should be written or the index might fail to get created if the folder permissions are not correct.
- I started updating search, but haven't tested it yet. Most of the code is there, it just needs to be tested and checked in.
- I created another eAIRS parameter tool (based on the first) that implements CyberintegratorToolTupelo interface. This only required minor code changes from the existing tool, but the new tool is provided access to the Tupelo context and can write metadata, if needed. The interface differences between CyberintegratorToolTupelo and CyberintegratorToolFile are minor.
- Built and deployed eAIRS locally for testing since we were having issues with correct versions of interfaces being built. It turns out that you have to be careful building with Maven because if you have old versions of a project already built, it might include the old jar instead of the new one (build order matters). Wiping out local copies of the projects forces the build order to be done properly.
- Attended Monday Staff meeting via skype
- Attended Tupelo meeting in Champaign.
- Continue working on KISTI eAIRS-web. The lucene index should be written to disk and should be updated when a user submits a workflow. The search view needs to be updated to limit search by user or by all users. I also plan to create a tupelo-tool version of the eAIRS parameters tool to make it easier to add metadata from within the tool.
- Attend 10am staff meeting on November 7th
- Drive to Champaign to attend Tupelo meeting on November 11th.
This week went as planned.