This is an archived snapshot of W3C's public bugzilla bug tracker, decommissioned in April 2019. Please see the home page for more details.

Bug 5426 - Revisit whether or allow partial XPath implementation
Summary: Revisit whether or allow partial XPath implementation
Status: RESOLVED FIXED
Alias: None
Product: XML Schema
Classification: Unclassified
Component: Structures: XSD Part 1 (show other bugs)
Version: 1.1 only
Hardware: PC Windows XP
: P1 normal
Target Milestone: ---
Assignee: C. M. Sperberg-McQueen
QA Contact: XML Schema comments list
URL:
Whiteboard: XPath cluster
Keywords: resolved
Depends on:
Blocks:
 
Reported: 2008-01-25 22:18 UTC by David Ezell
Modified: 2008-05-23 23:30 UTC (History)
0 users

See Also:


Attachments

Description David Ezell 2008-01-25 22:18:06 UTC
The status quo says the following things about XPath, formerly dealt with in bug 3673:

1) Schema authors can write any XPath 2.0 expressions they desire.
2) Schema processors are not allowed to report errors on any given XPath 2.0 expression.
3) Processors are required to support the minimum XPath 2.0 subset defined in the spec.
4) The spec allows only certain references using XPath; capacity exceeded errors may arise during assement.

The WG has received repeated requests to revise #3 as #3':
3') Processors are required to support XPath 2.0.
Comment 1 C. M. Sperberg-McQueen 2008-05-07 02:28:45 UTC
A discussion proposal intended to serve the discussion of this topic
and provide a possible final resolution is at 

  http://www.w3.org/XML/Group/2004/06/xmlschema-1/structures.b5426.html
  (member-only lin)

Comment 2 C. M. Sperberg-McQueen 2008-05-23 23:30:38 UTC
The wording proposal mentioned in comment #1 was discussed and adopted by
the WG at its call of 16 May.  The editors were instructed to make three
changes:  (1) add constructors functions and casts to the minimal subset
for CTA, (2) add a note warning that even with full XPath support, ancestors
and descendants and siblings will not be accessible, as a result of the
trimming of the tree during XDM instance construction, and (3) reformulate
the rules which say, in effect, that the only function calls which fall
into the minimal subset are those for fn:not (and the constructors for the
built-ins).  

These changes have now been made and integrated into the status-quo document;
I'm marking this issue as resolved.  David, as the originator of record,
I'll leave it to you to close it, if you are satisfied with its resolution.