[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: Deterministic content models???

• To: drudder@d... (Doug Rudder)
• Subject: Re: Deterministic content models???
• From: John Cowan <jcowan@r...>
• Date: Wed, 7 Aug 2002 12:01:06 -0400 (EDT)
• Cc: jcowan@r... ('John Cowan'), drudder@d... (Doug Rudder), andy@e..., robin.lafontaine@d..., xml-dev@l...
• In-reply-to: <7FEEE73A06F3D411A99D00508BD634100104E537@FACEXCH1> from "Doug Rudder" at Aug 07, 2002 10:54:28 AM

Doug Rudder scripsit:

> Exactly.  But I can see where someone with an SGML DTD background (like me,
> for instance) might initially try use this construct out of habit in an XML
> DTD -- until of course it fails to validate in XML and requires debugging.

Or, of course, you can go to RELAX NG where there are no such silly rules,
and write something like this:

paracontent = (\text | list | (table, tabnote*) | figure)+

\text = text | this | that ...

list = foo | bar | baz ...

table = element table { ... }

tabnote = element tabnote { ... }

figure = element figure { ... }

(The \ before "text" prevents it from being confused with the builtin
word "text", which is the equivalent of #PCDATA.)

-- 
John Cowan                                <jcowan@r...>     
http://www.reutershealth.com              http://www.ccil.org/~cowan
Yakka foob mog.  Grug pubbawup zink wattoom gazork.  Chumble spuzz.
    -- Calvin, giving Newton's First Law "in his own words"

• References:
  • RE: Deterministic content models???
    • From: Doug Rudder <drudder@d...>

• Prev by Date: RE: Deterministic content models???
• Next by Date: Re: Deterministic content models???
• Previous by thread: RE: Deterministic content models???
• Next by thread: Tree View + / - Lists with XSL
• Index(es):
  • Date
  • Thread