[xsl] Re: Re: questions about XSLT philosophy: how much is too much?

["Dimitre Novatchev" <dnovatchev@xxxxxxxxx>]

"Robert P. J. Day" <rpjday@xxxxxxxxxxxxxx> wrote in message
news:Pine.LNX.4.44.0303200629520.3554-100000@xxxxxxx
> On Thu, 20 Mar 2003, Dimitre Novatchev wrote:
>
> > How can a newcomer judge what is the "basic design" and what is
"natural"
> > for a language?
> >
> > The fact that you may be experiencing difficulties in understanding
these
> > does not mean that the language is used unnaturally.
> >
> > Or shall we start a witch hunt because we do not understand how someone
is
> > using the language and proclaim this use as "unnatural"?
>
> i didn't mean for this thread to take on a life of its own like
> this, and i think you're overreacting to what i wrote.

Didn't you have to tell something similar to other people today?

>
> i never claimed that certain usage of XSLT was "unnatural".  what i
> *said* was that, for someone not used to it and coming in with
> the understanding that XSLT was being promoted as a *functional*
> language, seeing it being used in what appears to be a non-functional
> (i.e., procedural) way would *seem* not natural.  that's all,
> nothing more.

Can you give examples of using XSLT in non-functional way on xsl-list?

>
> once upon a time, i programmed in prolog, and even taught the
> language in a college course in AI.  a lot of students had difficulty
> getting their head around their first functional programming language
> until they finally clicked on how functional programming worked --
> then things went smoothly.
>
> my only point from previous posts was that there is potential
> confusion for newcomers like me when we're told up front that
> XSLT is functional, then almost immediately see it being used
> to solve problems in what appears to be decidedly non-functional
> ways.
>
> one example of potential confusion is in tidwell's book, p. 3,
> where he *first* writes, "XSLT is heavily influenced by the design
> of functional programming languages ... functional programming
> languages define programs as a series of functions ..."
>
> and yet, in the very next paragraph, he writes, "Instead of
> looping, XSLT uses iteration and recursion."  excuse me but,
> where i grew up, iteration is just another word for looping.
> and looping sure seems to have a procedural programming history.
> and the potential for confusion for beginners increases when
> they learn that XSLT does indeed have "if" statements,
> case statements and loops.  suddenly, this emphasis on
> functional programming seems not so strong anymore.

Even functions have to be built up from some promitive constructs.

If you agree that Haskell is a functional language, there one will find:

if ... then ... else...

case

an iterate() function

some

xxxWhile() and xxxUntil() functions

Ther's even "do"  :o)

The functions mentioned above are from the Standard Haskell Prelude.

>
> finally, not to beat this to death, since i read that it's
> coming in XSLT 2.0, but i was amused to see how one finds,
> say, the largest value among sibling nodes, which as i recall,
> involved comparing to all nodes along the preceding-sibling::
> and following-sibling:: axis, yes?
>
> when i saw that, my first reaction was, why is there no max()
> function?  i mean, given that XSLT was promoted as being
> functional, it struck me as ironic that the one thing that
> would have made this solution easy -- a max() function --
> wasn't there.

Why should a max() function be provided? In a functional languge only a few
functions need be provided as primitives (not defined by other functions).
Again in Haskell the max() function is not provided as a primitive function
of the language and one can see that it is defined as:

maximum :: Ord a => [a] -> a

maximum = foldl1 max



Where max() is also a non-primitive function:

max x y | x <= y = y

| otherwise = x



Exactly in the same way, it is up to the XSLT programmer to define their
implementation of max or maximum -- this is not difficult.


>
> anyway, i didn't mean for this observation to become such
> an issue.

It is an issue, because it is not correct.


=====
Cheers,

Dimitre Novatchev.
http://fxsl.sourceforge.net/ -- the home of FXSL




 XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list