[
Lists Home |
Date Index |
Thread Index
]
Jeff Lowery <Jeff.Lowery@creo.com> writes:
> Okay, just a quick test using the attached schema and document:
>
> XSV 2.0-3: okay
> SQC 2.1.1: error (although it's surely wrong):
> <field xpath="act:name|act:desc"/>
> evaluates to the following two objects which may appear at the same time in
> an instance document :
> <element maxOccurs="1" minOccurs="1" name="name" type="normalizedString"
> xmlns=" http://www.w3.org/2001/XMLSchema"/
> <http://www.w3.org/2001/XMLSchema> >
>
> <element maxOccurs="1" minOccurs="1" name="desc" type="normalizedString"
> xmlns=" http://www.w3.org/2001/XMLSchema"/
> <http://www.w3.org/2001/XMLSchema> >
>
> I'm not sure either one is doing a complete job of validating the schema.
> While XSV is the most correct, it doesn't complain if you remove either
> act:name or act:desc from the key's field XPath (neither does SQC). It
> would seem that it should, since both fields are in a choice model group,
> and if the one in the key field is missing XSV will throw a document
> validation error.
Nothing in the REC requires processors to do static analysis of a
schema with keys in it to determine if the keys are in principle
satisfiable. This is a hard problem, and it didn't seem necessary to
require processors to solve it.
ht
--
Henry S. Thompson, HCRC Language Technology Group, University of Edinburgh
Half-time member of W3C Team
2 Buccleuch Place, Edinburgh EH8 9LW, SCOTLAND -- (44) 131 650-4440
Fax: (44) 131 650-4587, e-mail: ht@cogsci.ed.ac.uk
URL: http://www.ltg.ed.ac.uk/~ht/
[mail really from me _always_ has this .sig -- mail without it is forged spam]
|
