Re: [xsl] Using or ignoring Types in XSLT 2.0 / XPath 2.0

Subject: Re: [xsl] Using or ignoring Types in XSLT 2.0 / XPath 2.0
From: Jeni Tennison <jeni@xxxxxxxxxxxxxxxx>
Date: Wed, 14 May 2003 09:40:57 +0100
Hi Mike,

>> The error isn't recoverable, but it is preventable. Before
>> attempting the conversion of a string to a dateTime, you can test:
>> <xsl:if test="$x castable as xs:dateTime">
> Thanks Michael, that's what I was missing. This just invokes the
> lexical parsing of $x and compares with the lexical requirements for
> xs:dateTime, correct? Or is its behavior more determined than that?

It checks whether all the requirements of xs:dateTime are met (not
just that it's in the correct format, but that it actually is a legal
dateTime) -- in other words it tells you precisely whether or not the
cast will fail.

> Just a minor nit, wouldn't syntactic consistency suggest
> *castable-as* for the operator name? Is it really two tokens?

Yes, it's really two tokens. There are a number of operators in XPath
2.0 that are two tokens, such as "cast as", "instance of", "treat as".
In contrast, "isnot" is one word...



Jeni Tennison

 XSL-List info and archive:

Current Thread