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 1519 - WXS's "data type namespace" is not reserved
Summary: WXS's "data type namespace" is not reserved
Status: CLOSED INVALID
Alias: None
Product: XPath / XQuery / XSLT
Classification: Unclassified
Component: XSLT 2.0 (show other bugs)
Version: Last Call drafts
Hardware: Other Linux
: P2 normal
Target Milestone: ---
Assignee: Michael Kay
QA Contact: Mailing list for public feedback on specs from XSL and XML Query WGs
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2005-07-01 19:58 UTC by Frans Englich
Modified: 2005-07-21 20:05 UTC (History)
0 users

See Also:


Attachments

Description Frans Englich 2005-07-01 19:58:09 UTC
Among the namespaces categorized as reserved, as per 3.2 Reserved Namespaces, I   
don't find the W3C XML Schema namespace, which only allows addressing built-in   
types(http://www.w3.org/2001/XMLSchema-datatypes).   
   
I cannot comment on the issue of the two namespaces(if there is an issue at 
all), but  
my current thought is that the data types namespace should at least be  
reserved. It would otherwise be possible to for example create a stylesheet  
function(xsl:function) with a local name identical to a primitive type and the  
namespace being WXS-datatypes, hence looking like a constructor function, and  
creating potential problems.  
  
  
Cheers,   
Frans
Comment 1 Michael Rys 2005-07-01 20:05:00 UTC
Since the -datatypes namespace is not normative and would lead to a problem 
with having two non-related types in our type system we discourage the use of 
these types and decided a long time ago to not include them in the type system 
or provide a predefined namespace prefix binding.

Best regards
Michael (speaking for himself)
Comment 2 Frans Englich 2005-07-01 21:19:06 UTC
   
Ok. In effect, the -datatypes namespace is an arbitrary namespace with no   
special meaning in any context.   
   
This namespace-issue is a bit confusing matter, in my opinion. It might clear 
future confusion if the -datatypes namespace was explicitly mentioned in a 
specification(xslt20?), and outline the rationale. It might be informative, but 
also promote conformance among implementations, since if special treatment of 
the -datatypes namespace is implemented, it can lead to clashes with user 
code(this change to the specification is a suggestion, not a request). 
  
(I am fine with having the bug report closed as INVALID, or any other 
resolution.) 
   
 
Cheers,   
Frans   
   
Comment 3 Michael Kay 2005-07-01 21:24:55 UTC
A personal response: XSLT attaches no special meaning to names in the W3C Schema
datatypes namespace, for example, data types in this namespace are not
recognized as data types, and functions in this namespace are not recognized as
constructors. Therefore, the namespace falls into the same category as many
other namespaces that some other group has defined: it's good etiquette to avoid
using them with a meaning other than the meaning defined by the namespace's
owner, but the XSLT processor won't complain if you do otherwise.

The decision not to recognize http://www.w3.org/2001/XMLSchema-datatypes (let's
call it xsdt:) was a deliberate one: there's a technical problem with the
definition of this namespace in that xsdt:integer is a subtype of xs:integer,
and xsdt:decimal is a subtype of xs:decimal, and therefore xsdt:integer is not a
subtype of xsdt:decimal.

Michael Kay