RE: Shouldn't the XMLDSig XPath here function be defined in a separate namespace?

Hi Sean,

I think you are generalizing something that is not in the XPath spec, i.e. you are
generalizing from "unqualified if built-in" to "unqualified if and only if built-in".

Generally, there is no requirement to namespace-qualify functions that are added to 
an XPath evaluation context.  For example, XSLT consumes XPath and adds a number of 
functions, which it does not place in the XSL namespace.  The same is true of XForms.  
Generally, the XPaths appearing in these languages appear as element content or 
attribute values of elements or attributes *in the consuming language's namespace* and 
are therefore evaluated in a context identified by the consuming language.
(There are three or four more things to say here, but they have been omitted for now
in the interest of brevity).

Namespace qualification is particularly not required for DSig because the full context 
of evaluating the XPath v1 expression is defined to contain the XPath v1 functions plus 
here(), not the XPath v1 functions, here() and any other random assortment of functions 
that happen to be available on the local computer.  If other functions were available and 
were used during signing, then there would be interoperability problems during validation, 
so no other functions are allowed in the evaluation context.

Hope this helps,
John Boyer

-----Original Message-----
From: Sean Mullan [mailto:Sean.Mullan@Sun.COM]
Sent: Tuesday, March 23, 2004 12:49 PM
To: w3c-ietf-xmldsig@w3.org
Subject: Shouldn't the XMLDSig XPath here function be defined in a
separate namespace?



Why isn't the here function [1] defined in the dsig or some
other namespace? In XPath v1, functions without a namespace
are builtin. This is not a builtin function.

[1]: http://www.w3.org/TR/xmldsig-core/#function-here

Thanks,
Sean

Received on Tuesday, 23 March 2004 17:40:56 UTC