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 15912 - [XP3] Function items based on collation-dependent functions
Summary: [XP3] Function items based on collation-dependent functions
Alias: None
Product: XPath / XQuery / XSLT
Classification: Unclassified
Component: XPath 3.0 (show other bugs)
Version: Last Call drafts
Hardware: PC All
: P2 normal
Target Milestone: ---
Assignee: Jonathan Robie
QA Contact: Mailing list for public feedback on specs from XSL and XML Query WGs
Depends on:
Reported: 2012-02-06 15:57 UTC by Michael Kay
Modified: 2013-06-19 09:13 UTC (History)
1 user (show)

See Also:


Description Michael Kay 2012-02-06 15:57:07 UTC
We have a number of bug reports against tests (*) that use constructs such as contains#2 or deep-equal#2 - constructs which we have disallowed because the functions in question are context-dependent. I think this points to a usability problem which we should fix. Users can work around the problem by defining their own function as a wrapper around the system function, but it's clumsy, and users are going to trip over this all the time.

I propose that we should remove the ban on forming function items from context-dependent system functions in the case where the only dependency is on the static context. In very many of these cases the dependency is on the default collation. It makes sense to take the default collation from the static context at the point where the function literal appears. There is no ambiguity, as there could be in the case of a dependency on the dynamic context.

* Examples: bug #15911, bug #15642
Comment 1 Jonathan Robie 2012-03-20 16:23:54 UTC
From the Prague face to face:

Bug 15912 (Context dependent function items). DECIDED to resolve bug 
15912 by splitting the category of functions that are context-dependent 
into two (static-context-dependent and dynamic-context-dependent), and 
permitting use of those that are only static-context-dependent in 
function item literals and dynamic function calls.
Comment 2 Jonathan Robie 2012-03-20 16:29:16 UTC
Leaving open pending WG discussion to specify in more detail.
Comment 3 Jonathan Robie 2012-05-07 18:58:07 UTC
The Working Group has accepted the proposal in the following message:

With editorial changes noted here: