Re: [docbook-tc] Schematron problem in DB5 grammar

[Norman Walsh <ndw@nwalsh.com>]

/ Jirka Kosek <jirka@kosek.cz> was heard to say:
| Norman Walsh wrote:
|
|> It looks like we're encountering some interoperability issues with the
|> schematron rules embedded directly in the schema. I'm tempted to yank
|> them out entirely in 5.0b1 and simply ship a separate Schematron
|> schema.
|
| I agree that we should ship separate Schematron schema. I think that
| we can left Schematron rules inside RELAX NG because in some tools it
| works and some are just ignoring it. I didn't came across tool that
| will complain about current schema (no sch:pattern and sch:ns). But if
| there is such tool, we should definitively remove Schematron rules
| until this problem gets resolved.

I added the patterns, because I can't see how to generate them (they
need to have a name) so now MSV falls over.

|> If we can get the tools vendors to some consensus on how to encode
|> Schematron rules inside a RELAX NG grammar, we can start shipping them
|> that way again.
|
| Yep. Probably we should try to put together list of RELAX
| NG+Schematron implementations against we could test schemas. There are
| validators that I know:
|
| MSV (+relames)
| Topologi Schematron Validator
| oXygen
| Relaxed (product of my student, see http://badame.vse.cz/validator/)

Cool. There are more than I thought.

                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com>      | All knowledge is of itself of some
http://www.oasis-open.org/docbook/ | value. There is nothing so minute
Chair, DocBook Technical Committee | or inconsiderable, that I would
                                   | not rather know it than not.--Dr.
                                   | Johnson

PGP signature