Lists Home |
Date Index |
From: Mike Champion [mailto:firstname.lastname@example.org]
Sent: Sat 7/6/2002 12:47 PM
To: email@example.com; Dare Obasanjo
Subject: RE: [xml-dev] XQuery and DTD/Schema?
> I've never heard anyone pine for the ability to reject instances from a query because they didn't validate against
> a schema (most use fairly open content models in their storage schemas to make
> sure that trivially invalid instances aren't rejected at storage time),
validate() is poorly named. It is a cast operation for converting arbitrarily constructed XML to a specific named type not an operation meant for validating instances in the middle of a query. It would be extremely wise for the XML Query WG to change its name because I am yet to meet anyone who on hearing the term "validate" doesn't think it is a ridiculous idea until it is explained that it is simply the complex type version of CAST AS.
This can be gleaned from how validate() is used such as in the example I posted to the list a few days ago.
> and I've never heard anyone but Jonathan want the ability to ensure that the results of
> a query are valid instances of a schema without running them through a validator.
That's probably because most people realize that this problem is beyond the ability of computer science to solve and even for decidable cases may be expensive to implement or utilize.