OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

relax-ng message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Subject: Re: [relax-ng] relaxng.dtd


Thanks for doing this.  I noticed one major problem.  In a case like this:

<!ELEMENT element     ((%nameClass;)?, (%pattern;)+)>

the content model is non-deterministic because a choice element can match
both nameClass and pattern. Similar problem with <div>.  Here are the
warnings I get from rxp -V:

Warning: Content model for choice is not deterministic.   At start of
content th
ere are multiple choices when the next element is choice.
 (detected at end of prolog of document file:///E:/temp/test.rng)
Warning: Content model for element is not deterministic.   At start of
content t
here are multiple choices when the next element is choice.
 (detected at end of prolog of document file:///E:/temp/test.rng)
Warning: Content model for attribute is not deterministic.   At start of
content
 there are multiple choices when the next element is choice.
 (detected at end of prolog of document file:///E:/temp/test.rng)
Warning: Content model for except is not deterministic.   At start of
content th
ere are multiple choices when the next element is choice.
 (detected at end of prolog of document file:///E:/temp/test.rng)
Warning: Content model for div is not deterministic.   At start of content
there
 are multiple choices when the next element is start.
 (detected at end of prolog of document file:///E:/temp/test.rng)

XML makes non-deterministic content models an error (i.e. processsing result
is completely undefined), so I think this has to be fixed.

My other suggestion would to add some parameter entity hooks for
extensibility.  For example, RELAX NG allows all RELAX NG elements to have
arbitrary namespace qualified atts.  So I would introduce a parameter entity

<!ENTITY % extensionAtts "">

and reference that at the end of commonAtts.  Similarly for element
extensions.

----- Original Message -----
From: "DuCharme, Bob (LNG)" <bob.ducharme@lexisnexis.com>
To: <relax-ng@lists.oasis-open.org>
Sent: Friday, December 07, 2001 3:18 AM
Subject: [relax-ng] relaxng.dtd


> (First: I originally subscribed to relax-ng-comment-request, but there
> doesn't seem to be much action there and LexisNexis is part of Reed
> Elsevier, an OASIS member, so I've joined this list.)
>
> I put together a RELAX NG DTD to make it easier to create RNG schemas with
> DTD-driven XML editors and put it up at www.snee.com/xml/relaxng. Please
let
> me know of any bugs, suggestions, 1.0 schemas that won't validate against
> it, etc.
>
> thanks,
>
> Bob DuCharme
> Consulting Software Engineer, LexisNexis
> Data Architecture, Editorial Systems and Content Engineering
>
>
> ----------------------------------------------------------------
> To subscribe or unsubscribe from this elist use the subscription
> manager: <http://lists.oasis-open.org/ob/adm.pl>
>
>
>



[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]


Powered by eList eXpress LLC