W3C home > Mailing lists > Public > xml-editor@w3.org > January to March 1999

Re: 3.2.1 Emphasize on undefined cps

From: C M Sperberg-McQueen <cmsmcq@acm.org>
Date: Fri, 19 Mar 1999 18:04:08 -0600
Message-Id: <199903200004.SAA284580@tigger.cc.uic.edu>
To: omeyer@i3.informatik.rwth-aachen.de
CC: xml-editor@w3.org, cmsmcq@acm.org
>From: Oliver Meyer <omeyer@i3.informatik.rwth-aachen.de>
>Date: Thu, 4 Mar 1999 13:30:14 +0100 (MET)

>after a short discussion in comp.lang.xml, I'd like to submit a
>request for clarification in section '3.2.1 Element Content' of
>REC-XML-19980210. 
>
>Make clear how an XML processor should behave if the Name used in
>[48]cp references an undeclared element.
 ...
>Some parser already issue warnings for this case, others don't.
>
>Please let me know, if the above does NOT describe the bahaviour
>intended by the XML Spec.

You are correct in saying that reference to undeclared element
is valid according to the XML 1.0 spec.

I agree also (a) that processors should be allowed to issue warnings
in this case, (b) that the warnings must be at user option (otherwise
some widely used DTDs will become unbearable to use), and (c) that the
spec should say that processors may at user option issue warnings
about this case (if only to make clear that it must be a warning, not
an error!).  I don't believe the words 'for interoperability' should
be used, since conforming SGML processors are required to accept
content-model references to undeclared elements.

Thanks for your note.

-C. M. Sperberg-McQueen
 Senior Research Programmer, University of Illinois at Chicago
Received on Friday, 19 March 1999 19:05:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:59:29 GMT