Guenther Goerz has also created a Mac version of the Image Markup Tool using Wine Bottler (it's a large download). Download it from the IMT Downloads page.
An update to the Image Markup Tool has been released. This includes a couple of small enhancements and fixes, detailed on the Updates page of the site.
A user reported the usual Firefox bug when creating an empty
<ref> element; the empty anchor that results from this kills FF's rendering of the page. Easy enough to trap in XSLT, which I'll do.
Another report concerns the attempt by the application to fix non-well-formed XML. It will successfully escape ampersands which appear in text nodes, but fails when they're in attribute nodes (such as lengthy URLs with multiple parameters). This might or might not be fixable; I'll have to take a look.
Fixed the two problems reported in the last two posts, and provided a new version to the user who reported them. If everything works well, I'll do an official release. Putting this on as a task in case I forget.
<ref> element's @target attribute should be converted to a proper XHTML link on when you export to create a web view.
Interesting issue reported by a user: if you have empty p tags in your annotations, then a self-closing div is output to the HTML. If you then open the file in Firefox, it will screw up, but if you give the file an .xhtml extension, it works fine.
I should put in a trap for empty paragraphs, and insert a non-breaking space into them. That could easily be done in the XSLT. Making this a task.
The Image Markup Tool version 220.127.116.11 has been released. This is a minor bugfix release. See the Update page for more details.
Working at home on a personal project, I've found a bug in the way IMT handles the content of annotation titles (which are
<head> elements). When reading
<head> elements which contain mixed content, a space character preceding an element node is eliminated. I pinned it down and fixed it, and I've created a new release, but I'll test it at home before I release it finally.
They can read and write themselves from a QDomDocument, and seem to be working well. Next, I need to polish up the test application so that it works as a demo of that module of the real app, which means installing some more filtering features (sorting works already, but filtering on any of the values in the fields would be great).
Added transcriptional attribute, and tested sorting with QSortFilterProxyModel.