W3C home > Mailing lists > Public > public-xml-processing-model-comments@w3.org > November 2009

Revisions to p:document and p:load

From: Norman Walsh <ndw@nwalsh.com>
Date: Tue, 03 Nov 2009 15:05:24 -0800
To: public-xml-processing-model-comments@w3.org
Message-ID: <m2zl7343bf.fsf@nwalsh.com>
Hello world,

At the f2f meeting today, we happened to notice some (apparently long
standing) problems with p:document. Further discussion lead us to make
fairly large changes to the p:document and p:load sections (and the now
penultimate paragraph of 2.2 immediately before 2.2.1).

Before these changes our spec allowed, even encouraged, the
construction of documents that could not be processed with XPath
(because they would contain unexpanded entity reference information
items which are not allowed).

The changes we are now visible in the current editor's draft:

1. Change p:document so that it does what an XPath processor needs: fully
expand entities, etc. 

2. Change p:load to make it clear that it does the same thing as p:document,
only with the option of performing DTD validation and with a possibly computed
href value.

3. Limit the scope of the binding to external documents to be only
those documents that don't have unexpanded entity reference
information items in them.


                                        Be seeing you,
                                          norm

-- 
Norman Walsh <ndw@nwalsh.com> | The reverse side also has a reverse
http://nwalsh.com/            | side.

Received on Tuesday, 3 November 2009 23:06:25 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 3 November 2009 23:06:26 GMT