HCMC Blogs

Images for 1855 (CO 305 06) are now on the site, indexed and working, and I'm running the rsync operation from yesterday to back them up to Rutabaga.

A new issue is emerging with images: where a figure wraps to the next page, a space is left. We may have to carefully manage the insertion of images to place them optimally on the page.

Deep in code, unable to swim to surface...

I have a pilot thing working which parses an XML file, displays it in a syntax-highlighted editor, and simultaneously displays its DOM tree in an expanding tree widget and a list of its xml:ids (with the first 30 characters of the element's text) in a list view. This is pretty good, but I know in my heart that what I really need is a model/view architecture whereby the QDomDocument is the model, and these three views are all synchronized so that when you click on (say) an xml:id in the id list, the dom tree opens and highlights the corresponding element, and the editor scrolls to, and highlights, the same element. QT's model/view stuff hurteth my brane mightily, but I managed to understand it for long enough to write the little SVG resource applet, so I'm sure I can do it again.

Finally did something I've been meaning to do for ages: created a full backup of the processed page-images from Lettuce to Rutabaga. This took literally all day (I throttled rsync a bit to make sure Lettuce didn't struggle). I realized I could also have logged on directly to nfs.tapor.uvic.ca, taking Lettuce out of the equation; might do that in future. This is the process: ssh into lettuce, navigate to the coldesp www folder, and run rsync --verbose --progress --stats --compress --recursive --times --bwlimit=10000 jpg_scans/ -e ssh mholmes@rutabaga.hcmc.uvic.ca:/"home/mholmes/backups/Martin/Colonial\ Despatches/www/jpg_scans/" . I did the equivalent for the maps directory as well.

Mainly as part of testing the PDF rendering.

One of the few remaining reasons for working in the XP VM was that I had the old oXygen/XEP system from 2005 set up there for ScanCan rendering. I've finally got around to porting that over to my Jaunty setup. This was the process:

• Copied the whole RenderX folder from Program Files to a shared dir.
• Ran the installer jar file that was in there, which installed for Linux without problems, allowing me to specify the licence.xml file which was in the program folder.
• Copied all the fonts over from the fonts subdir of the original XEP folder to the same folder in the install location.
• Opened the xep.xml configuration file, original and new, and added my customizations, copying them from the original file to the new.
• Set up the appropriate transformations in oXygen.
• Edited the path_config.xsl file to point to the correct location for the images file on my machine (needs to be hard-coded).

Tested and seems to be working (and much faster than on Windows). Haven't yet tested a build of a full volume; I'll do that when we get close to having volume 19 ready.

After more hassle than I expected getting forms to submit with correct values, I now have a flexible and responsive map gallery, with paging features, and the ability to change the sort sequence and the number of items displayed on a page. Next is getting the individual map display code working, which means figuring out how MJ's code works, and trying to fit it into the context of the site.

Note to self: remember, in future, that rather than trying to get a form to submit itself in the old-fashioned manner, it's much simpler to write a bit of script that grabs all the values you want and constructs a GET URL, then sets the location to that. Even simpler might be AJAX, but in this case it seemed overkill so I went the route of a traditional form page. And gave myself loads of trouble as a result (e.g.: if you try to trigger submission from the onchange event of a <select> element, you'll have trouble getting the value of the selected option, and the onsubmit event of the form will never fire).

Tested and working in FF, WebKit (Epiphany), Opera, and IE8.