oh no. i was afraid you would say that. your suggestion would have significant reprecussions.
i agree that it's probably less difficult if one thinks outside the XMLSchema box, but as of now, this needs to be 100% XMLSchema. On Fri, 15 Feb 2002, Eddie Robertsson wrote: > > so, A,AB,ABA,ABAB,... etc are legal > > I think this would be hard to implement in W3C XML Schema alone. However by > embedding Schematron rules within the XML Schema this would be trivial to do. > I've put up a draft paper describing this technique at [1]. > > Cheers, > /Eddie > > [1] http://www.topologi.com/public/Schtrn_XSD/Paper.html > > > > > > > but B,BA, AA, BB, ABB, etc are not. > > > > this was the first implementation, but it violates UPA as far as xerces-j > > 2.0.0 is concerned: > > > > <sequence> > > <element name="A" type="string"/> > > <sequence minOccurs="0" maxOccurs="unbounded"> > > <element name="B" type="string"/> > > <element name="A" type="string"/> > > </sequence> > > <element name="B" type="string" minOccurs="0"/> > > </sequence> > > > > among other things. i was thinking that this was a recursion problem and > > this was what i came up with after google-mining for recursion and > > XMLSchema: > > > > http://castor.exolab.org/list-archive/msg08774.html > > > > this was my distilled contribution, but it's not correct because it would > > permit AA: > > > > <element name="AB-list-ref"> > > <complexType> > > <sequence> > > <element ref="AB-list"/> > > <sequence> > > <complexType> > > <element> > > > > <element name="AB-list"> > > <complexType> > > <sequence> > > <element name="A" type="string"/> > > <choice> > > <element name="B" type="string"/> > > <element ref="AB-list-ref" minOccurs="0"/> > > <choice> > > <sequence> > > <complexType> > > <element> > > > > so, mine is wrong. and i have been beating the heck out of this and not > > getting anywhere. > > > > if somebody knows of a canonical example of such a thing, i'd really like > > to see it. if anybody has implemented something that sorta works, i'd love > > to see that too! > > > > one thing that has occurred to me as a sortof nifty tool: a way to express > > regexps as XMLSchema: > > > > ie, enter a regexp and get back a correct XMLSchema for same. > > > > anybody ever seen such a beasty? > > > > tnx for any help that anyone can provide! > > > > johnu > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [EMAIL PROTECTED] > > For additional commands, e-mail: [EMAIL PROTECTED] > > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [EMAIL PROTECTED] > For additional commands, e-mail: [EMAIL PROTECTED] > > --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
