- Created the release build process and closed that ticket.
- Updated documentation and cleaned up a few things, making the Instructions read-only in Oxygen Author.
- Tested on Windows and confirmed that it's completely broken; no errors are found in the test data under Oxygen.
Category: "Activity log"
- Checking for @xml:base
- Checking for undeclared prefixes
Fixed the bug whereby errors in the old site setup, which were resulting in URLs which had extra path components in them, were not being successfully redirected to the correct new page. Changed version from alpha to beta. I think everything is done now.
SHARP pres is done; should expand notes nearer the time.
Did JT's timesheet; negotiated and wrote a new contract etc. for JT; planned his Endings work; chased up TL's timesheet which is not complete; did some work on diagnostics and Graves.
Addressing some common issues with search on the webapp:
- MoEML did not have the search results highlighting constraint that limits hits within a document to ten, which I'd already implemented for Mariage and Scancan. That's now done.
- It was possible to get eXist error pages if you crafted an ill-formed search. I'm now trapping that with a try-catch and just returning zero hits (all projects).
- Wildcard searching with leading wildcards is now implemented.
- Search tips page has been rewritten for MoEML.
With RE's help, got four new exists deployed, with their apps working, and publicly available. Lesson learned:
- For configuration and deployment of the XAR, use Chrome, and connect to the internal :8080 address. If you connect through Apache, you risk a timeout during deployment, which will bork eXist completely.
Other than one case of that happening with the MoEML app, everything worked as planned; the deployment script goes seamlessly, the XARs all worked, and everything looks good. Couple of bugs to fix in MoEML and Mariage.
The project information essay is going to take a while to write, and I can't live with the mechanical rendering of the teiHeader metadata from the project_metadata page, so I've given it some textual content in the form of a very brief history of the project. That will act as a decent placeholder until EGW has finished her article.