On 19/05/15 20:51, Martin Holmes wrote:
I've been bitten by it too, and I was the once who made the change. :-) I think any significant change to class structures like this will generate similar problems. I have tried to think of a way around this, but the only approach I can think of involves specially-coded traps in the schema creation process, and that's a bit ungainly. We could add Schematron rules to apply to the ODD, but for that to work, you would need to be validating your ODD against a fresh TEI schema at some point. Would that have worked for you in this case?
It would indeed have worked for me. I was at work so subscribed to the draft oxygen-tei framework created with the most recent TEI. So if you had added a schematron rule that a classSpec with @ident of att.global that has a valList for @rend inside it should show a warning that it should be att.global.rendition, then I'd probably have noticed quicker. :-) Probably. Seems a reasonable schematron warning to add, but I'd only want this to be a warning because I could, for example, delete @rend on att.global.rendition and make a brand new @rend on att.global, for some bizarre reason, and this should technically be legal, no? -James -- Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT Services, University of Oxford