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

Re: Remove duplicates from a list

---

---

This is entirely conjecture, but I predict that optimization will follow
a similar path as programming language compilers.  In the early days,
you could gain much performance with clever optimizations in your code. 
As time moved on, compilers became increasingly better at optimizing for
you.  Present day, compilers are usually good enough at optimizing that
you're better off writing clear, consistent code.  In parellel,
programming idioms have emerged that are clear and concise, but also
"best practices" as far as writing code that is optimizable.  Of course,
there are exceptions, and no optimization is going to fix a bad
algorithm.

I expect that the same will hold true for XSLT processors.  Early
implementations that are focussing on tracking the spec and being as bug
free as possible will most likely benefit from stylesheets that are
carefull designed with optimization in mind.  Eventually, processors
will be able to optimize better, and you can concentrate on clear and
concise XSL, so long as you don't do anything horrendous.

Just a guess.

Tim

Jarno Elovirta wrote:
> 
> > That was too easy. Now give us an answer that has n*log(n) performance.
> 
> speaking of performance, has anyone written aything about optimization of
> XSLT stylesheets? i myself am not a programmer and don't understand how XSLT
> processors are optimized internally, but i would like to know how i should
> write my stylesheets for better performance. quite often multiple different
> solutions are presented for a given problem on this list and it's not always
> clear which one was the "best". so, are there any resources on design
> principles of XSLT stylesheets when good performance is needed?
> 
> Jarno
> 
>  XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list

-- 
Tim Taylor


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

---