W3C home > Mailing lists > Public > public-xml-core-wg@w3.org > September 2004

(unknown cha

From: (unknown cha <richard@inf.ed.ac.uk>
Date: Fri, 17 Sep 2004 13:12:50 +0100 (BST)
To: (unknown charset) Fran├žois Yergeau <francois@yergeau.com>, public-
Message-Id: <20040917121250.0FE991656BD@macintosh.inf.ed.ac.uk>

> PE132 [1] is about the validity check on default attribute values.  The 
> commenter objects to syntactic checking only (as asserted by E9 to the 
> 2nd edition [2][3]) and points out two tests in the test suite to 
> support his case.  IMHO the test suite needs fixing.

I haven't looked back at our discussions of this, but it's possible that
we didn't carefully enough consider the boundary between syntactic and
semantic validity.

We certainly don't want to have to check that a default IDREF value is
an ID that occurs in the document, because you can't check that until
you have read the whole document.  We might not want to check that an
ENTITY default is the name of a declared unparsed entity (this example
is given in the spec), because you can't check it until the end of the
DTD.  But you can check that the default for an enumerated value is in
the list without any problem.

Probably we did what SGML does, and if so we should leave it like that.

-- Richard
Received on Friday, 17 September 2004 12:12:54 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:21:31 GMT