Following the success of the MoEML document type taxonomy as a way of organizing, listing and searching resources, I've decided to do the same thing with the Coldesp documents, so I've started outlining a taxonomy. The project schema has also been updated to allow the appropriate header elements.
I now have a bash script,
utilities/generate_oai_records.sh, which does the following:
<earliestDatestamp>for the records.
With this, we have a fully-automated process to regenerate the complete OAI record set whenever we need to. It takes about fifteen minutes to complete, and has been pretty thoroughly tested.
The old XQuery-based way of generating OAI files was very time-consuming and required an offline Cocoon/eXist to run. I've now completely rewritten it as a single XSLT transformation, which takes about a minute to generate all the files. It has a couple of parameters we can tweak if we decide to change ids to remove the .scx, and it builds the files to a separate folder. I'm going to combine it with a shell script that will:
Today I've started work on a completely new web application for the Despatches project. The first stage has been to get a good working build script; I'm using the one Greg created in the svn/hcmc/utils repo, but I've tweaked it a bit (and recommitted the changes) so that it will include the Scheduler extension, which I think we may use for building OAI files.
I've decided to build this app (initially at least) in the form of an eXist .xar file, so I've been learning how xar files work and how they're built. I've configured the ant build.xml script so that it can update the svn copy of the data and copy that into the xar file from the original svn location, excluding certain files. This is a good start, and it means we should be able to create an easily-updated, easily-deployed version of the webapp for testing with new builds of eXist as we move forward with the rewrite. There's a lot of stuff from the standard xar setup that I'm not going to follow, and I definitely won't be using any of the templating stuff, but I see no reason not to use the basic xar deployment functionality.
I'm also going to reconfigure the SVN file structure a little too, to make it more friendly for XQuery targetting the despatches. I'll move the "year" directories into a single "despatches" directory.
Reviewed KSW's rewrite of the grant application.
KSW and I met with CP, LW and CW at the library to review the situation regarding grant applications. We completed the interim report for this year, which is now with CW and will be submitted next week. We also plan to write the final report for this year within the next two months, so that it's in before the grant application for next year goes in, and the latter has to be there by December. CW will send us a copy of last year's final report, so that we can start drafting this year's. Meanwhile, TB will be asked to read through the 1863 transcripts and relevant external sources to list the significant events from that year, to help with writing the justification for the new application.
1863 is about the same size as 1862 in terms of the transcripts we have; we would want to generate images from CO 60 / 15, 16 and 17, which we have (B-088, B-089, B-090); ditto CO 305 / 20 and 21 (B-244 and B-245); CO 398 / 2 is done, as is CO 410 / 1. CO 6 / 37 and 38 we'll have to get from LAC, and while we're at it I think we should get 39 through 43 (the last CO 6), so they're there when we need them. RG7 G8C vol 10 is already done and covers 1863; it runs to the middle of reel C15634, and vol 11 starts in the middle of that reel.
A previous attempt to format vessel names correctly was incomplete. What we need to do is to always format vessel names in italics when they're not in correspondence documents; but in correspondence, we stick with the formatting of the source document (of course). I've now tweaked the CSS and that seems to be the result we're seeing.
Hadn't grabbed a copy of the stats for a few months. Now done.
It looks as though I'd never finished implementing and testing the use of @type on placeName and @subtype on label/name for vessels for incomplete or unavailable entries, so I updated the schema to allow these and include documentation, and fixed the XSLT which wasn't quite doing the right thing.
The Colonial Despatches is an XML database project which is creating a digital archive containing the original correspondence between the British Colonial Office and the colonies of Vancouver Island and British Columbia. The project lives at http://bcgenesis.uvic.ca, and the web application runs on the Pear dev Tomcat. The XML data is managed in SVN at http://revision.tapor.uvic.ca/svn/coldesp/.
|<< <||> >>|