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 1275 - [F&O] Need QName -> URI mapping for functions namespace
Summary: [F&O] Need QName -> URI mapping for functions namespace
Status: CLOSED WONTFIX
Alias: None
Product: XPath / XQuery / XSLT
Classification: Unclassified
Component: Functions and Operators 1.0 (show other bugs)
Version: Last Call drafts
Hardware: PC Linux
: P2 normal
Target Milestone: ---
Assignee: Norman Walsh
QA Contact: Mailing list for public feedback on specs from XSL and XML Query WGs
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-04-26 14:27 UTC by C. M. Sperberg-McQueen
Modified: 2005-09-29 10:41 UTC (History)
1 user (show)

See Also:


Attachments

Description C. M. Sperberg-McQueen 2005-04-26 14:27:17 UTC
I can't find a qname->uri mapping in
http://www.w3.org/TR/2005/WD-xpath-functions-20050404/#namespace-prefixes

ala
http://www.w3.org/TR/webarch/#qname-mapping

[This entry added by CMSMcQ from Dan Connolly's email at 
http://lists.w3.org/Archives/Public/public-qt-comments/2005Apr/0112.html]
Comment 1 Norman Walsh 2005-05-02 21:52:24 UTC
Our specifications appear to make use of the following namespaces:

Standard  Standard
Prefix    URI

  dm:       none
  err:      http://www.w3.org/2004/07/xqt-errors
  fn:       http://www.w3.org/2005/04/xpath-functions
  fs:       none
  local:    none
  op:       none
  xdt:      http://www.w3.org/2005/04/xpath-datatypes
  xqx       http://www.w3.org/2005/04/XQueryX
  xsl:      http://www.w3.org/1999/XSL/Transform

We have defined several of these prefixes without a corresponding URI
because they exist only as an editorial device or because they
describe things that are completely abstract and we have no reason to
refer to them in any concrete sense.

But Dan has demonstrated that it may be useful for the larger
community as a whole to be able to identify even the things that we
consider abstract using URIs. In particular, I believe that Dan wanted
to define one of our operations in a semantic web context and found
that there was no URI for it.

In that light, I think it probably makes sense to provide real URIs
for the dm: fs: and op: prefixes, in addition to the prefixes that we
have already defined. Concretely, I propose:

Standard  Standard
Prefix    URI

  dm:       http://www.w3.org/YYYY/MM/xpath-data-model
  fs:       http://www.w3.org/YYYY/MM/xpath-formal-semantics
  op:       http://www.w3.org/YYYY/MM/xpath-operators

I really don't think there's any reason to define a URI for the local:
prefix, but I'm not really sure I understand how "local:" works in
XQuery so I could be wrong.

Each of our namespaces needs a namespace document. I'm willing to
produce the namespace documents for the dm:, fn:, and op: prefixes. I
can probably be persuaded to take on some others, if no one else
volunteers.

Per the good practice[1] in WebArch[2], we must provide a mapping
from QNames to URIs. That is, there must be a way to translate
the QName pfx:localname into a URI. I propose that we adopt a single
definition for all of our specs: the URI of the resource identified
by pfx:localname is

 concat(namespace-uri(pfx:localname), '#', local-name(pfx:localname))

I suggest that we write a working group note that outlines our suite
of namespaces and our QName to URI mapping.

[1] http://www.w3.org/TR/webarch/#qname-mapping
[2] http://www.w3.org/TR/webarch/
Comment 2 Michael Rys 2005-05-03 03:36:33 UTC
Sorry, but adding namespace URIs to the abstract prefixes are highly 
problematic since we would have to define semantics for XSLT and XQuery 
specifications to disallow users to use these URIs which is problematic in its 
own right. If the RDF folks want to refer to these functions, they better 
define their own URIs in their domain and refer to the definition.
Comment 3 Ashok Malhotra 2005-05-20 22:45:54 UTC
The QT WGs discussed this on 5/19/2005 and agreed with the form suggested for
the URIs:
> the URI of the resource identified by pfx:localname is

> concat(namespace-uri(pfx:localname), '#', local-name(pfx:localname))

They did not, however, approve the rest of the suggestions for QName to URI mapping.