Re: [xsl] N : M transformation

Subject: Re: [xsl] N : M transformation
From: Tobias Reif <tobiasreif@xxxxxxxxxxxxx>
Date: Sat, 01 Feb 2003 13:28:42 +0100
David Carlisle wrote:

  I asked for
  aspart of a spec draft,
  as opposed to the various extensions.

As far as I know saxon 7's the only experimental xslt2 system that's
been "announced" Although mention of a xalan xslt2 system has been
mentioned on this this list as well (check the archives)

I see.

One should be very careful about using any such systems in production of
course, XSLT2 is only a draft,

Sure! That's why I wrote
"It's your decision if you prefer reliance on drafts or on extensions".

W3C drafts as opposed to W3C recommendations.

and in particular the multiple output
instructions have changed quite a bit over the XSLT2 drafts so
there's no assurance that it will end up looking like it does currently.

OK. Instead of "the XSLT 2.0 syntax will become "standard" (part of a W3C XSLT rec) AFAICS.", I should have written what i meant: "the XSLT 2.0 syntax is more likely become "standard" (part of a W3C XSLT rec) than processor specific extension syntax (which is is its own namespace anyways)."

Using exslt's document extension or the processor specific mechanism provided by an XSLT1 system is to be prefered at present

That's a little general (I think that decision is up to XSLT programmers, to be made anew for each situation/scenario), but I think in general I agree with your point.

for experimenting with the current draft specs.

I use the latest stable Saxon for example. I switched from saxon:document (IIRC) to XSLT 1.1 document. I could switch to EXSLT, but what I currently use works well, so there's no need to change. Again: it's up to each programmer to decide if he wants to rely on extensions or on drafts. I agree that well-specified extension features are a more stable base than W3C drafts, but I don't agree that the latter is the wrong choice in all situations.



XSL-List info and archive:

Current Thread