W3C home > Mailing lists > Public > public-grddl-wg@w3.org > April 2007

Re: #xmlbase [1234]

From: Jeremy Carroll <jjc@hpl.hp.com>
Date: Thu, 26 Apr 2007 17:10:35 +0100
Message-ID: <4630CEFB.30807@hpl.hp.com>
To: ogbujic@ccf.org
CC: GRDDL Working Group <public-grddl-wg@w3.org>

If I understand you correctly, your reading of the specs, and the WG 
decision concerning baseURIs, is as follows below.

As far as I can tell, your reading is better than mine.

Please read this, and confirm that I have correctly understood your 
position. Then I will modify the xmlbase tests to comply.

I will report in next message as to other issues arising.


Stepping through again (skip to ISSUE for the divergence between what I 
was thinking, and what I think you think)


The file is retrieved from:
and has contents
<foo xmlns="http://www.example.org"

whichdir.xsl is a relative reference, and, GRDDL assumes that xml:base
has been legally used (i.e. that the namespace http://www.example.org
supports XML Base), thus, the absolute form is:

The xsl is run to give:

    <rdf:Description rdf:about="">
 >Transformed by stylesheet from /2001/sw/grddl-wg/td/</

(my line breaks)


rdf:about="" is a same document reference, that hence resolves to the
baseURI of this intermediate representation.

Now, the phrase "baseURI of this intermediate representation" was
considered by the WG before I joined, and the resolution was:
RESOLVED: Given that a base URI parameter is a parameter whose value is
the base URI of the source document, the WG RESOLVES not to define a
base URI parameter for transforms.

so we devolve to our dependency chain.
If xml:base appears, we assume that it's use is legal
that we should be using
as the baseURI for interpreting the RDF/XML and so get that as the
subject of the triple.

I have been previously arguing, that sorting out xml:base and html:base 
is down to the transform, and if it doesn't, it doesn't. So that, the 
baseURI to be used should be
the retrieval URI.

I think we are all agreed that #xmlbase1 and #xmlbase3 have the same
answer - it's just what is it?

Chime points to section 5.1.3 of RFC 2396
actually the picture is helpful.
I think Chime's argument is compelling.

   |  .----------------------------------------------------.  |
   |  |  .----------------------------------------------.  |  |
   |  |  |  .----------------------------------------.  |  |  |
   |  |  |  |  .----------------------------------.  |  |  |  |
   |  |  |  |  |       <relative-reference>       |  |  |  |  |
   |  |  |  |  `----------------------------------'  |  |  |  |
   |  |  |  | (5.1.1) Base URI embedded in content   |  |  |  |
   |  |  |  `----------------------------------------'  |  |  |
   |  |  | (5.1.2) Base URI of the encapsulating entity |  |  |
   |  |  |         (message, representation, or none)   |  |  |
   |  |  `----------------------------------------------'  |  |
   |  | (5.1.3) URI used to retrieve the entity            |  |
   |  `----------------------------------------------------'  |
   | (5.1.4) Default Base URI (application-dependent)         |

We have a relative reference "" inside an RDF/XML representation.
There is no base URI embedded in content, so 5.1.1 does not apply.
Thus we use 5.1.2 to get the Base URI of the 'encapsulating entity',
which was the XML document above.

So we start again: the 'encapsulating entity' has a base URI embedded in 
content, so we never get to 5.1.3 URI used to retrieve the entity, and 
hence my current code that assumes that we do, is broken.

I think this means we need to review some of the tests we passed last night.


Hewlett-Packard Limited
registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England
Received on Thursday, 26 April 2007 16:10:58 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:39:11 UTC