HCMC Blogs

More bugs reported by a user that are now fixed. Mostly these were due to the failure to specify the full range of variant spellings when "double u" is involve (e.g. adiuuo, which can be adiuvo, as well as adivuo). Rebuilt two units, after a lot of editing in HotPot.

The custom fields in the search page now have ids constructed similarly to the main fields, using the [tableName].filter. at the beginning, so they're being correctly parsed and submitted by the JavaScript. I've also tidied up some of the appearance issues by making the long text field render as a single-line input when it's being used as a filter. Next is adding the back-end handling that adds appropriate search SQL to the MdhRecordList object based on the submitted custom fields.

KSW was having trouble running oXygen on Karmic (Ubuntu 9.10). Keystrokes would not generate characters in the editor for a second or two after being hit. The oXygen help file provided the solution:

Display problems like screen freeze or momentary menu pop-ups during mouse movements over screen on Linux or Solaris can be solved by specifying the parameter

-Dsun.java2d.pmoffscreen=false

for the Java virtual machine. This parameter disables off-screen pixmap support and must be added to the Java command line which starts the Java virtual machine at the end of the file oxygen10.3 located in the install directory.

We added this to both the oxygen and oxygen.sh files, grouping it with other similar parameters, and remembering to put quotes around it where required.

Thanks to all of you for the continuing solid work.

I have begun to go through the reference file, where I think some of the references can be trimmed to make them more focused to the context at hand. This will take me a while as I'm away from the computer on and off until October 14, but the process has begun.

In general, if information is being taken from a source (Robert Illustré or Universalis, etc), the important thing is to acknowledge that source (or those sources) whether quoting directly with quote tags, or paraphrasing. A combination of sources may be desirable when tailoring the reference to the context. As you develop experience with the corpus, this will be easier to do, but in any case, I will continue to monitor the reference file.

CP wrote asking for some sort capabilities on the search page (sort by date, etc.). The framework for these was mostly already done in the back-end XQuery that does the search, so I tweaked and refined it a bit to fix a couple of bugs, and then implemented an interface for it based on a drop-down list control. My original intention had been to enable sorting by column headers in the table of search results, but it's actually impossible to figure out exactly how that should work when there are multiple pages of results, so the drop-down is a more elegant solution.

In the process, I also noticed some hard-coded caption strings in the search.xsl file, so I replaced them with calls to variables in strings_en.xsl.

I also fixed a couple of bugs in the search_query_data.xq file. This is the XQuery that I run every now and then (by accessing it in the web interface) to refresh the lists of authors, addressees etc. in the drop-downs in the search page. CP pointed out that Moody was in the author list twice; it turned out that there was a return character in the <author> tag in five documents with his name, and this was resulting in two different entries for him (although in the end result on the page, they appeared identical). I've now updated the XQuery to make it more intelligent with regard to whitespace. This file can be as complicated as it needs to be, because it's only run once every now and then. Eventually it will also have to take account of names, places and vessels (I think -- perhaps that would be too much for dropdowns?).

Had to apply for another secondary account for English and couldn't find my previous post on how to find the application form, so here it is again:
To apply for a departmental secondary account:
go to http://helpdesk.uvic.ca/
click on Netlink Tools in nav list on left of window
authenticate
click on Netlink in nav list of window
click on Apply for Secondary in sublist under Netlink
fill in form

Here we go with my list.

1. Proof the last 40 pages of Varin. Tiny alterations probably still needed for how the text is showing up on the website (e.g. linebreaks, etc. to add).
2. Candale: write the references. Get a consensus on how best to write the references (am I now directly quoting from the Petit Robert or am I still to paraphrase while carefully changing vocab so as not to plagiarize?
3. Agreemens: proof both volumes, add a note at the end of the second volume saying that the third and four are missing.
4. Agreemens: references.
5. Isembourg: references.
6. References: overhaul, put MY references in some kind of intelligible order to make working with them easier, make everything cohesive.

KSW finished several hundred scan images for the 1847 documents, so I integrated those into the db, and also refreshed the 1846 and 1847 XML files (the latter not finished yet, but getting close). Found a validation error in one 1846 file, and warned KSW to fix it in his copy. The handling of the images is quite time-consuming partly because of the number of images involved, but also because they're organized (for good reason) by year in KSW's work directories, but have to be re-organized by CO volume on the server and in the XML index.

JJ reported a problem with the site today. The eXist service wasn't running so I started it, but the experimental map wasn't running. Toolbar and so forth appear but no map.

I've tailed the access and error logs and find no errors.

Posting the time spent over the last week or so on back-and-forth about a grant application related to the IMT. These things eat so much time...