XSLT is fun
Can't Understand It? Don't Worry. For some time I used this argument against getting deeply into XSLT. I had used XSLT many times as part of an MVC architecture: an XML document is the model, an XSLT file is the view and some ASP JavaScript is the controller. In this architecture it's important to put all processing in the controller, so I would do whatever manipulation of the XML was needed to keep the XSLT as simple as possible. For example, rather than do any complex processing in XSLT to calculate a value, I would do the calculation in script and make an XML element or attribute containing the calculated value before doing the transform. Good separation of data, format, code. So all of the hard stuff in XSLT was not needed and this fits in well with Joel's point, I think. COM monikers = <xsl:key> and <xsl:variable> and all the other hard stuff.
Thankfully, I came to my senses a few months ago when I realized this simple fact: XSLT is fun. It's a challenge. I put it in the same category as regular expressions and advanced SQL. A regular expression replaces many lines of simple code. A complex SQL select statement does the same. I love both those technologies and relish the complexity. (I'm so glad we finally have a compiled language with regexes built in.) So now I jump at any chance to use XSLT in a new way (and there have been many opportunities).
I'd like to thank Michael Kay and his participation on XML-DEV and his most excellent book for helping me to see the light.
8:27:58 AM 
|
