OASIS Darwin Information Typing Architecture (DITA) TC

XSD Generation Status: 26 June 2014

  • 1.  XSD Generation Status: 26 June 2014

    Posted 06-26-2014 17:19
    The current status is: - ditabase.xsd is not yet generated correctly for 1.2 and 1.3. This is special-case processing that just needs to be worked out. Shouldn't be hard, I just need to focus on it. - All other 1.2 XSDs appear to be valid. There are some catalog generation details that I need to work out with Eric Sirois but the catalogs currently generated for the URN-based XSDs appear to be correct. - For 1.3, there are issues with the L&T learningAssessment, learningGroupMap, and learningObjectMap XSDs. All other XSDs appear to be correct. Having thought more about the issues with constrainable content models and getting more information from Eric to help me understand the technical XSD details, I think that automating the generation of constrainable content models is beyond my abilities and available time so I'm going to abandon that effort in favor of simply generating XSDs that work for the TC-defined 1.3 vocabulary and shells. We can make the issue known to the community and handle constraint requirements on a case-by-case basis or continue to work on a more general solution as a post-1.3 activity. Or just wait for general implementation of xs:override in common XSD processors (it may already be in the latest Xerces and Saxon implementations, I'm not sure) and look at redefining the XSDs to use xs:override rather than xs:redefine. I will continue to work on the XSD generation today and tomorrow and hopefully resolve all the issues. Cheers, Eliot ————— Eliot Kimber, Owner Contrext, LLC http://contrext.com