Re: More XSL Discussion

Subject: Re: More XSL Discussion
From: Paul Prescod <papresco@xxxxxxxxxxxxxxxx>
Date: Wed, 25 Feb 1998 13:53:52 -0500
Sean Mc Grath wrote:
> I contend it is XSL. There is nothing in the spec. that I can
> see that says "you must have a flow object". Quite right too

That's because the spec. is not finished. Just as it prohibits global
variables it will also prohibit side effects (println()) for many of the
same reasons. A single construction rule could be executed over and over
again in a conformant XSL implementation. Your println() will then print
your report text many times when you only want it printed once.
> I further contend that using, say, msxsl (yes, I know it is only
> a technology preview of a moving target) without flow objects
> to munge XML is pretty darn useful. It has already saved
> me many hours of tedium doing reports/queries/harvesting
> etc. of XML docs.

No one has argued otherwise.

> If noone else finds XSL very useful for doing this sort of
> thing then I have clearly missed out on some excellent tools
> you guys have that I do not!

I use Jade for the same thing. But when I do it, I do not claim that I
am doing "DSSSL". There is nothing in the DSSSL spec. about report
writing. DSSSL and XSL are stylesheet languages, not report writing
languages. Jade and msxsl are tools, which happen to be useful for
report writing due to implementation choices made by their programmers. 

Some other DSSSL/XSL implementation could choose *not* to be useful in
that manner, either by removing the println() function (which is not an
ECMAScript[1] builtin, but rather an environment-provided function) or
by executing the construction rules in an order that you do not expect.


Paul Prescod  -

[Woody Allen on Hollywood in "Annie Hall"]
Annie: "It's so clean down here."
Woody: "That's because they don't throw their garbage away. They make 
        it into television shows."

 XSL-List info and archive:

Current Thread