[
Lists Home |
Date Index |
Thread Index
]
>However it is unfortunate that after purchasing a 10 pound tome like Mike
>Kay's excellent XSLT Programmer's Reference and learning the idiosyncracies of
>both XPath and XSLT one still needs to utilize procedural programming
>technologies be it ECMAScript, Java or C# if one wants to perform non-trivial
>tasks using stylesheets.
>
>This reminds me of a post from a few months ago that stated[0]
>
>"One of the problems I think this highlights is that there is a growing
>disconnect from how the W3C wants XML to be used and how people use XML and
>want to use it. Creating XML content that can be transformed to other formats
>as needed is one of the advantages of using it in a business environment that
>I constantly hear touted, yet it doesn't seem like enough work is being put
>into developing technologies and tools that are aimed at making it easier to
>manipulate XML when dealing with non-trivial problems. (Again, I'd like to
>praise everyone involved in adding XPath support to DOM in any way, shape or
>form)."
>
Well, for me the irreduceable minimun has been getting the input, and putting the output
where it needs to go. This has variously involved file systems, HTTP servers and editor
buffers, but in all cases is beyond XSLT's scope. I've been in the MSXML world, and there I am
loading stuff into DOMDocuments before I invoke the TransformNode method. And likely it's a string at some
point so I can use regexes too.
Horses for courses as they say across the pond.
Frank
|
