W3C home > Mailing lists > Public > www-xml-schema-comments@w3.org > January to March 2008

[Bug 5193] Validation Rule: Element Locally Valid (Complex Type)

From: <bugzilla@wiggum.w3.org>
Date: Fri, 04 Jan 2008 02:23:43 +0000
To: www-xml-schema-comments@w3.org
Message-Id: <E1JAcDn-00005Q-Mf@wiggum.w3.org>


cmsmcq@w3.org changed:

           What    |Removed                     |Added
             Status|NEW                         |ASSIGNED
  Status Whiteboard|                            |thimble, easy

------- Comment #1 from cmsmcq@w3.org  2008-01-04 02:23 -------
On the first point, see also bug 4368.

On the second point, you are right.  And yet -- is it a fool's errand to wish
make explicit that where the treatment of whitespace (or white space) differs
between XML 1.0 and 1.1 (I mean in particular the treatment of U+0085, the NEL
of ISO 6429) the schema processor may do either, and should document which?
The wording quoted talks about white space, not about the 's' production in 
the XML spec; I have a vague recollection of wording it carefully to ensure
it made sense even though the 's' production has not change.

Of course XSDL defines its input (for better or worse) as an infoset, not 
an XML document, and so one could argue that the difference is really utterly
out of scope for the spec.  In that case, I think I would probably argue for
a non-normative Note pointing out the issue in terms that make clear that it's
(a) a question of how the input infoset is created, and thus out of scope for
the XSDL spec, and (b) a question of how an XML document taken as input is
read, and thus of some interest for users who hope to use our technology 
instead of only admiring the purity of our spec's concentration on 
untestable abstractions, and who care about getting the same results from
one processor to another.  
Received on Friday, 4 January 2008 02:23:53 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:50:07 UTC