Re: [XPath Filter 2.0] Problem with Transform model

r/[email protected]/2002.05.29/11:27:31
>
>On Sunday 26 May 2002 08:31 am, Gregor Karlinger wrote:
>> A possible solution for this problem could be to explicitely
>> forbid an emtpy input for the XPath Filter 2.0 transform. Any
>> thought?
>
>How would one forbid this? We can't stop a user from specifying an XPath 
>expression that misses the mark, right? If we said throw an error, that 
>would stop the processing of the signature -- maybe there are other useful 
>XPath transforms. We could add a processing step that says if the input 
>node-set is empty the transform is done.

Just as it is an error to use here() in an XPath transform that is
processing a different document from the signature document, it is
an error to supply an empty node set to the XPath filter transform.
I agree with Gregor; the transform should fail and processing of
the reference should stop.

Merlin