[
Lists Home |
Date Index |
Thread Index
]
> -----Original Message-----
> From: Michael Kay [mailto:michael.h.kay@ntlworld.com]
> Sent: Friday, May 10, 2002 7:37 AM
> To: 'Jeni Tennison'; 'Uche Ogbuji'
> Cc: 'Jonathan Robie'; xml-dev@lists.xml.org
> Subject: RE: [xml-dev] XPath 2.0 - how much of XQuery should
> it include?
>
>
> > "Shortcutting" means that you only evaluate the first part of
> > the expression if you can determine the result of the
> > expression, correct? In XPath 1.0, 'and' and 'or' are shortcutting:
> >
> >
> > I hadn't noticed that this changes in XPath 2.0 (it's not
> > listed as one of the incompatibilities):
> >
>
> Yes - this should have been listed as an incompatibility.
>
> This change was made because database systems need to use indexes to
> evaluate boolean expressions. In a database system, if someone writes
>
> colour=my-favourite-colour() and home-town="Wigan"
>
> then if you have an index on home-town, you want to use it.
>
> I've been thinking we might try to add something to XSLT that imposes
> rules for XPath-in-XSLT that are stricter than the rules in
> XPath itself
> to deal with such situations.
>
<sarcasm>
It's good to see that the W3C does extensive research on the user
scenarios for the "new science" they develop before foisting them on the
WWW community.
</sarcasm>
Changing the semantics of XPath's logic operators because of DBMSs is
ridiculous when one considers that DBMSs are not and most likely will
not in future be the primary users of XPath.
--
PITHY WORDS OF WISDOM
Two wrongs are only the beginning.
This posting is provided "AS IS" with no warranties, and confers no
rights.
You assume all risk for your use. (c) 2002 Microsoft Corporation. All
rights reserved.
|
