[XSL-LIST Mailing List Archive Home]
[By Thread]
[By Date]
Andrew and Greg,
Sara is right, the usual workaround is to pass the filename or a meaningful fragment of it (e.g. the base name without the suffix) to the stylesheet.
Andrew, the piping of processes you envision could be achieved this way if you used an extension function designed to write output (for example Saxon's implementation of the shelved XSLT 1.1 instruction xsl:document, which writes a result tree fragment to a document) -- the processor then knows the name of the output file, because it wrote the file, and can write the name somewhere or otherwise make it known to the system.
But the intention of XSLT's designers is that such pipelining be possible without ever writing a file, but simply passing node sets (DOMs or DOM-alikes). Cocoon (Apache's XML infrastructure), I believe, supports this kind of thing out of the box.
Some of us have occasionally thought it'd be useful to have a base-uri() function, which would tell you, in effect, the file name proper to a node. At the moment it slips my mind what the use cases are. The wrinkle is that the source for an XSLT transform may not be in one file (it could be in several entities), or even in a file at all.
At 05:17 AM 1/22/02, Andrew wrote:
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
RE: [xsl] self discovery
Subject: RE: [xsl] self discovery From: Wendell Piez <wapiez@xxxxxxxxxxxxxxxx> Date: Tue, 22 Jan 2002 12:18:01 -0500 |
Andrew and Greg,
Sara is right, the usual workaround is to pass the filename or a meaningful fragment of it (e.g. the base name without the suffix) to the stylesheet.
Andrew, the piping of processes you envision could be achieved this way if you used an extension function designed to write output (for example Saxon's implementation of the shelved XSLT 1.1 instruction xsl:document, which writes a result tree fragment to a document) -- the processor then knows the name of the output file, because it wrote the file, and can write the name somewhere or otherwise make it known to the system.
But the intention of XSLT's designers is that such pipelining be possible without ever writing a file, but simply passing node sets (DOMs or DOM-alikes). Cocoon (Apache's XML infrastructure), I believe, supports this kind of thing out of the box.
Some of us have occasionally thought it'd be useful to have a base-uri() function, which would tell you, in effect, the file name proper to a node. At the moment it slips my mind what the use cases are. The wrinkle is that the source for an XSLT transform may not be in one file (it could be in several entities), or even in a file at all.
Cheers, Wendell
At 05:17 AM 1/22/02, Andrew wrote:
ok.... (to make it clearer or possibly muddy it further)
Imagine: You have an XML file, and apply xsl-1 to it. Xsl-1 takes a node-set of the root (or any node-set) and passes that to some script. The script makes an object out it, and applies xsl-2 (a different stylesheet) to that, and displays the result.
The whole point of this is to be able to apply a second stylesheet without ever knowing the filename of the xml - is this fantasy or a possibility??
andrew
====================================================================== Wendell Piez mailto:wapiez@xxxxxxxxxxxxxxxx Mulberry Technologies, Inc. http://www.mulberrytech.com 17 West Jefferson Street Direct Phone: 301/315-9635 Suite 207 Phone: 301/315-9631 Rockville, MD 20850 Fax: 301/315-8285 ---------------------------------------------------------------------- Mulberry Technologies: A Consultancy Specializing in SGML and XML ======================================================================
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
Current Thread |
---|
|
<- Previous | Index | Next -> |
---|---|---|
[xsl] generate xsl version 1 using , Bryan Rasmussen | Thread | [xsl] Extreme 2002 - Call for Parti, B. Tommie Usdin |
Re: [xsl] differences in methods of, G. Ken Holman | Date | [xsl] Fwd: How to stop XSLT from tr, B. Tommie Usdin |
Month |