teitoschematron fails miserably

[sydb]

The teitoschematron program, whether specified with --localsource or not, fails to generate actual Schematron. The output has an <sch:title>, several comments, and a single <sch:ns> declaration, but no rules; the fact that <sch:ns> declarations for xs: and rng: are missing makes me think it is never reading in p5subset.xml at all.

I do not know for sure if this is caused by ADHO server failure or not, but I note that Stylesheets/profiles/default/schematron/to.xsl simply imports the extract-isosch.xsl file, and does nothing else. This seems wrong. That XSLT program expects a “complied” ODD as input.

[dmj]

I am not sure what the bug is here. I tested it with tei_lite.xml downloaded from roma2 and running bin/teitoschematron. This did create a Schematron w/ only a <sch:title> and a <sch:ns>.

However: This is to be expected because the tei_lite.xml did not contain any Schematron rules.

[trishaoconnor]

Assigning to @sydb to test to see if this error is still a problem.

[ebeshero]

@trishaoconnor @peterstadler @raffazizzi @sydb @HelenaSabel : I'm just adding a few people to this ticket b/c it seems like it might be connected with TEI Garage support. I think we might need to investigate what scripts are being run to see what they are failing to read to produce meaningful schematron.

I have very basic questions about how people write TEI that is expected to generate Schematron (as opposed to writing a constraintSpec in an ODD, or just writing Schematron directly). What is teitoschematron expected to do, when it's doing the right thing?