I have to finish implementing the tests, but I've done 17 so far and they're all passing. 120 minutes.
Left early for appointment.
This is the emergency info for the Digital Victorian Periodical Project.
- Site repository: hcmc.uvic.ca/svn/dvpp. This contains all the XML and programming code.
- Online database: this AdaptiveDB database contains metadata only, and should be phased out by the end of 2019. The live db is at hcmc.uvic.ca/people/martin/vpn/live/editor/, or for read-only access, hcmc.uvic.ca/people/martin/vpn/live/reader/. There is a dev version at hcmc.uvic.ca/people/martin/vpn/dev/editor/. The AdaptiveDB configuration files for these dbs are in the svn repo under 'mysql'.
- Jenkins site: the Jenkins job is at jenkins.hcmc.uvic.ca/job/DVPP, and it runs when commits are made to the svn repo, as well as once every night, just after midnight. A copy of the Jenkins configuration file is also stored in the svn repo inside the jenkins folder.
- Development site: the build process creates a site on Jenkins at jenkins.hcmc.uvic.ca/job/DVPP/lastSuccessfulBuild/artifact/products/index.html.
- Live site (future): Published versions of the site will be on nfs at hcmc/www/dvpp/. The subdomain dvpp.uvic.ca should point at this location.
- Documentation: The documentation is part of the dev site: jenkins.hcmc.uvic.ca/job/DVPP/lastSuccessfulBuild/artifact/products/dvpp.html, and it is also stored in the svn repo at hcmc.uvic.ca/svn/dvpp/documentation/dvpp.html. This document has a section for programmers which explains how to do stuff.
One question on alphabetization outstanding. 60 minutes.
Late duty.
Implemented the new category "preface" for documents that should be processed as though they were articles, but shouldn't be preceded by an Articles page. Seems to be working OK. Then did SH corrections, all except for the references (which will be a lot of work too). 180 minutes.
XSLT in place but not finished; target in build file but not called by default. 100 minutes.
Met with JJ and MH to talk about the LEMDO infrastructure, including build jobs, shared resources, and the potential for ODD chaining. Will do some more research into ODD chaining and think about cost/benefit.
Late duty.
Made substantial progress on the Porter2 stemmer in XSLT; I've done the first few steps (which I think are the trickiest), and all tests are passing so far. Looking at the algorithm, though, I think it's going to be way slower than the original one. Don't know how much that really matters. 120 minutes.