XPath as a stand-alone programming language (Was: Re:

A few months ago I had the pleasure to do "all XPath programming".

I developed the Binary tree data type and the Set data type
essentially completely in XPath 3.0.

These can be found in my blog, respectively at:

    http://dnovatchev.wordpress.com/2012/01/09/the-binary-search-tree-data-st
ructurehaving-fun-with-xpath-3-0/

and

    http://dnovatchev.wordpress.com/2012/01/12/the-set-datatype-implemented-i
n-xpath-3-0/

At the time this happened accidentally. I don't like to work with
XQuery, but at that time Saxon's XSLT 3.0 processor didn't fully
implement higher order functions. Saxon's  XQuery 3.0 did, so I
reluctantly used XQuery 3.0 in my work. Doing so, I planned to move my
whole code to XSLT 3.0 --whenever Saxon's XSLT 3.0 processor was ready
for this (and now, in version 9.4 it is). Therefore I essentially
coded everything as much in pure XPath 3.0 as possible.

This experience clearly showed to me that it is possible to use XPath
3.0 as a stand-alone programming language. This has many benefits, and
one of them is that XPath-only code is fully transferrable between
XSLT and XQuery.

At the end of my latest blog post "Recursion with anonymous (inline)
functions in XPath 3.0", I list a number of nice features we still
need in XPath as a programming language, that could be added in a
future, post-3.0 version:

 1. Stand-alone XPath processors.

 2. Import/include directives for XPath-only files (expression-files).

 3. Separate packaging/compilation of XPath-only programs.

 4. New data structures such as tuples.

 5. Generics b parametric data types.


Cheers,

Dimitre




On Wed, Oct 24, 2012 at 4:55 AM, Costello, Roger L. <costello@xxxxxxxxx>
wrote:
> Thanks a lot Michael for the explanation. Fantastic!
>
> Below is a summary of my question plus Michael's answer (slightly
expanded).
>
> The XPath 3.0 specification says this:
>
>       XPath 3.0 is a composable language
>
> What does that mean?
>
> It means that every operator and language construct allows any XPath
expression to appear as its operand (subject only to operator precedence and
data typing constraints).
>
> For example, take this expression:
>
> 3 + ____
>
> The plus (+) operator has a left-operand, 3. What can the right-operand be?
Answer: any XPath expression! Let's use the max() function as the
right-operand:
>
> 3 + max(___)
>
> Now, what can the argument to the max() function be? Answer: any XPath
expression! Let's use a for-loop as its argument:
>
> 3 + max(for $i in 1 to 10 return ___)
>
> Now, what can the return value of the for-loop be? Answer: any XPath
expression! Let's use an if-statement:
>
> 3 + max(for $i in 1 to 10 return (if ($i gt 5) then $i*3 else $i*2)))
>
> The fact that XPath is a composable language is very cool and very
powerful.
>
> Contrast this with XSLT, which is not fully composable: XPath expressions
can be used as operands to XSLT instructions, but not the other way around.
Similarly Java has constructs called statements that
> cannot be nested inside expressions.
>
> /Roger
>



--
Cheers,
Dimitre Novatchev
---------------------------------------
Truly great madness cannot be achieved without significant intelligence.
---------------------------------------
To invent, you need a good imagination and a pile of junk
-------------------------------------
Never fight an inanimate object
-------------------------------------
To avoid situations in which you might make mistakes may be the
biggest mistake of all
------------------------------------
Quality means doing it right when no one is looking.
-------------------------------------
You've achieved success in your field when you don't know whether what
you're doing is work or play
-------------------------------------
Facts do not cease to exist because they are ignored.
-------------------------------------
Typing monkeys will write all Shakespeare's works in 200yrs.Will they
write all patents, too? :)
-------------------------------------
I finally figured out the only reason to be alive is to enjoy it.