Re: [xsl] Recursive xsl parsing.

Subject: Re: [xsl] Recursive xsl parsing.
From: Jeni Tennison <mail@xxxxxxxxxxxxxxxx>
Date: Fri, 12 Jan 2001 09:48:46 +0000
Hi Gurvan,

> The first step parses a document A through an xsl AX (using the
> "xml-stylesheet" tag). The result is the valid, well formed xsl or xml
> document AXA. We then parse a document B through an XSL document BX which
> is calling the resulting AXA (using an include or a xmlDom.load  with a
> "msxml:script" tag).
> If we copy the resulting code of AXA directly in BX then, when we process B
> through BX, the result BXB is a valid well formed XML/XSL document.
> If we just link the document AXA in BX and we parse B then we don't get the
> expected result from AXA. At this point it seems that parsing B through BX
> which should be parsing A through AX misses the link. A is not parsed
> through AX and we only get the A document and not the expected AXA.

When you say 'link the document AXA in BX' I guess that you mean
that you're using the document() function to access the document A?
Something like:


When a XSLT processor sees that instruction, it just goes and gets the
document 'A.xml' - it doesn't interpret any xml-stylesheet processing
instructions in it or anything, just gets the XML.

I guess that one solution would be to have some kind of automatic
server-side transformation behind the scenes so that accessing 'A.xml'
(through a URL) actually gets you the transformed 'AXA' document.

Another possibility is to combine AX and BX together using xsl:import
and modes, then have a root-node-matching template that manages all
the transformations.

I'm really not sure that I've interpreted what you're doing correctly,
but if any of the above looks promising, feel free to follow it up with
another question and more details.



Jeni Tennison

 XSL-List info and archive:

Current Thread