- From: Peter McCracken <peterjm@ca.ibm.com>
- Date: Mon, 16 Jun 2003 13:12:48 -0400
- To: www-xml-xinclude-comments@w3.org
In the spec, section 4.5.5, the fourth property of the xml:base attribute
is:
4. A [normalized value] equal to the either [base URI] of the element,
or an equivalent URI reference relative to the [base URI] of the
include parent. The circumstances in which a relative URI is
desireable, and how to compute such a relative URI, are
implementation-dependent.
So it's up to you to decide if you want to output a relative or absolute
URI. I would assume that the xml:base attributes in the test cases are
relative because it doesn't matter what comes before the "Nist/" part of
the path -- it would be entirely dependent on the directory structure of
the machine that created the sample outputs.
Peter McCracken
IBM Toronto Lab
|---------+---------------------------------------->
| | Elliotte Rusty Harold |
| | <elharo@metalab.unc.edu> |
| | Sent by: |
| | www-xml-xinclude-comments-req|
| | uest@w3.org |
| | |
| | |
| | 06/16/2003 09:26 AM |
| | |
|---------+---------------------------------------->
>----------------------------------------------------------------------------------------------------|
| |
| To: daniel@veillard.com |
| cc: www-xml-xinclude-comments@w3.org |
| Subject: Re: DOCTYPEs in result documents |
| |
>----------------------------------------------------------------------------------------------------|
At 11:48 AM +0200 6/16/03, Daniel Veillard wrote:
> Another problem is that the xml:base which needs to be added is
dependant
>on the location of the test-suite at runtime unless one proceed to do
>some convoluted URI relativization of the xml:base values.
Yes, I've also encountered some problems with xml:base attributes in
the result document. However, in that case I'm not yet sure what the
real result should be and whether my output or the test case results
or both are wrong. It would be extremely helpful for the XInclude
spec to provide some more details on this such as whether the value
of the xml:base attribute should or should not be absolute, when the
href is not absolute, and whether or not the value of the xml:base
attribute should contain the fragment identifier. Possibly the
answers to these questions are inferrable from a combination of
different sections of the XInclude and XML Base specs and the URL
RFCs. However, if they are, I haven't yet made the right connections
to infer them.
--
Elliotte Rusty Harold
elharo@metalab.unc.edu
Processing XML with Java (Addison-Wesley, 2002)
http://www.cafeconleche.org/books/xmljava
http://www.amazon.com/exec/obidos/ISBN%3D0201771861/cafeaulaitA
Received on Monday, 16 June 2003 13:13:10 UTC