This is the mail archive of the
xsl-list@mulberrytech.com
mailing list .
RE: submitting rtf contained within param
- To: xsl-list at mulberrytech dot com
- Subject: RE: submitting rtf contained within param
- From: David_Marston at lotus dot com
- Date: Tue, 15 Aug 2000 10:20:25 -0400
- Reply-To: xsl-list at mulberrytech dot com
Matthew Bentley wants to do
<xsl:apply-templates select="$gentext" />
where $gentext is a result tree fragment (RTF). He notes that
Saxon complains, as would Xalan, but MSXML accepts it.
Then Joshua Allen (joshuaa@microsoft.com) says:
>MSXML treats everything as a node-set, thus no extension function.
>Most other processors have a node-set extension function (which is
>how Saxon does things). My vote is "spec vagueries",...
My vote is that the spec is clear enough, if just barely.
First, note the syntax in Section 5.4
<xsl:apply-templates select= node-set-expression ...
Now, look at Section 11.1, which defines RTFs. It talks about
binding a variable to an RTF "instead of" a node-set, and goes on
to use verbiage that consistently avoids portraying RTFs as a
subclass of node-sets.
As for automatic conversion, 11.1 also says "An operation is
permitted on a result tree fragment only if that operation would
be permitted on a string..." and XPath 3.3 says that no objects
can be converted to node-sets. That xsl:apply-templates
instruction takes a node-set, not a string.
Taken together, these provisions eliminate all the claimed
vagueness in the specs.
.................David Marston
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list