RE: XForms CR - 7.10.1 Another XPath 1.0 Conformance Question

Hi Andrew,
 
The words 'does something like this' meant 'access out of band data'.  XML DSig's here() function generates an error because that is appropriate in the context (please read XML DSig further for more info).  Nonetheless, XPath extension functions have to be able to use such out of band data to generate an error rather than crash.
 
John Boyer, Ph.D.
Senior Product Architect
PureEdge Solutions Inc.
  

-----Original Message-----
From: AndrewWatt2001@aol.com [mailto:AndrewWatt2001@aol.com]
Sent: Monday, November 18, 2002 12:11 PM
To: John Boyer
Cc: www-forms-editor@w3.org; xforms@yahoogroups.com; www-forms@w3.org
Subject: Re: XForms CR - 7.10.1 Another XPath 1.0 Conformance Question


In a message dated 18/11/2002 19:28:07 GMT Standard Time, JBoyer@PureEdge.com writes:




An excellent exposition, Ryan.

In essence, XPath functions supplied by the consuming application are free to access out of band data.  For another example of a function that does something like this, see the here() function in the XML Signature recommendation.

John Boyer, Ph.D.
Senior Product Architect
PureEdge Solutions Inc.



John,

The here() function in XML Signature is an interesting comparison. I assume you are intimately familiar with it.

One aspect of the description of the here() function is as follows,
"This expression results in an error if the containing XPath expression does not appear in the same XML document against which the XPath expression is being evaluated."

At least at first glance what the XForms WG is suggesting is legimate is what is stated normatively to be an error for the XML Signature here() function.

I would be interested in your exposition of why, when the instance() function does something similar to the here() function, it too should not generate an error.

Andrew Watt