W3C home > Mailing lists > Public > public-qt-comments@w3.org > January 2005

Re: [F&O] INF, -INF and NaN - literals?

From: David Carlisle <davidc@nag.co.uk>
Date: Fri, 14 Jan 2005 11:19:50 GMT
Message-Id: <200501141119.LAA04050@penguin.nag.co.uk>
To: colin@colina.demon.co.uk
CC: public-qt-comments@w3.org

> If you allow that, then you need something to handle the case of an
> element named NaN, etc.

which of course would be incompatible with Xpath 1, and also break the
general design of XPath (and Xquery) that there are no reserved words
and in a context where an element name is accepted _any_ element name
(even inconvenient ones like div or NaN) is accepted, and refers to an
element of that name, not an Xpath language construct.

number("NaN") has worked perfectly OK to represent NaN in Xpath 1, I
don't see why it should be confusing in Xpath 2.  This doesn't conflict
with the Schema usage; NaN would still be a valid literal for a double
in a schema validated attribute of type double.


This e-mail has been scanned for all viruses by Star. The
service is powered by MessageLabs. For more information on a proactive
anti-virus service working around the clock, around the globe, visit:
Received on Friday, 14 January 2005 11:20:36 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 16:57:03 UTC