[XSL-LIST Mailing List Archive Home] [By Thread] [By Date]

Re: [xsl] RSS transform


Subject: Re: [xsl] RSS transform
From: "Jon Gorman" <jonathan.gorman@xxxxxxxxx>
Date: Tue, 21 Feb 2006 08:49:05 -0600

On 2/21/06, THUFIR HAWAT <hawat.thufir@xxxxxxxxx> wrote:
> I'd like to do an "identity transform" with relaxer.  By identity
> transform, I mean that foo.xml becomes bar.xml, and vice versa, when
> they're transformed.  (Perhaps not "exactly," but that's the ideal.)

Isn't this something of a pseudo-tautology?  Typically people use XSLT
to create an output document.  Hence no matter what the stylesheet
does, foo.xml will become bar.xml.  (Example: I'd like to "delete" all
the nodes in the source document, by that I mean foo.xml becomes
bar.xml.)   Are you trying to describe a round-about crosswalk?  That
is transform a document from one set of schema rules to another and
then back to the original form.  This isn't always posssible, and
would be difficult to do in one stylesheet.


> a post of mine from usenet:

Did you get any useful answers when you posted it there?  Next time
might be nice to spend a little time adapting the message to the forum
instead of just copy and pasting it from one to the other.  (Pet peeve
of mine.)


> Relaxer will generate an XSD file but I'm not sure what to do with that.
> What I'd like to do is to do an identity transform as described in the
> tutorial introduction
> <http://www.relaxer.org/doc/tutorial/tutorial_s1.html>:

Most people don't follow links in this forum.  I actually tried it for
a bit, but I only have so much time and I stopped after the third or
so page.  Also, I've never used relaxer before, but I have heard of
it.


>      * XSLT stylesheets (identity transforms) based on instances
>
>
> I'm just mucking about here, and am just looking for a simple result.  I'd
> like to use relaxer to do this but can't figure out how to do an identity
> transform such that the result is for all intents and purposes exactly
> what was transformed.

Huh?  Google xslt identity transform.  I imagine the archives and faq
of this list comes up frequently.  (And probably the section in the
specs called "Identity transform").

http://www.dpawson.co.uk/xsl/sect2/identity.html (the faq)



> If I can get an identity transform for one of these RSS feeds that'd give
> me a jumping off point to muck around, but right now I'm feeling a bit
> stuck with how to "do" XSLT.  (I've got Xalon up and running, but nothing
> to run through it.)

You really need to give more information.  I have a feeling you're
trying to automatically generate the XSLT using the relaxer program.
What problem you're having is not clear though.  Googling "generating
styelsheets relaxer" gives me quite a few examples.

If you want to create a transformation by hand though, this is the
place to do it.  We don't tend to talk much about specific
applications. We will occasionally talk about XSLT processors, but
that's about it.  In other words, imagine a list that talks about c++
that might occasionally get compilier-specific.  Now you're asking
about how to automatically generate c++ from some code-templating
program.

> [thufir@localhost feedsRSS]$
[snipped the unnecessary info ]

I don't see you using the xslt option.  Might be an issue here.  I
looked up the home page of the project and I'm assuming you've read
through the manual about how to generate xslt pages right?
http://www.relaxer.org/doc/refman/1.0/html/refman_p10_en.html#doc1_48

Have you tried asking questions in the mailing list suggested by what
appears to be the author (I got this by google relaxer mailing list)?
See http://www.asahi-net.or.jp/~dp8t-asm/java/tools/Relaxer/.

Jon Gorman


Current Thread
Keywords