Re: [xsl] Javascript in Xsl using-Xalan-Problem

Subject: Re: [xsl] Javascript in Xsl using-Xalan-Problem
From: "M. David Peterson" <m.david@xxxxxxxxxx>
Date: Sat, 5 Jun 2004 15:41:29 -0600
Hi Barry,

> My main concern with XSLT 2.0 is how well it can be maintained by
> someone else on the project team.  With XSLT 1.0 we could just get the
> newcomers to buy Mr. Kay's book.

Dr. Kay also has an XSLT 2.0 book that can be preordered from Amazon 
(http://www.amazon.com/exec/obidos/tg/detail/-/0764569090/qid=1086468757/sr=1-2/ref=sr_1_2/102-3934922-7293706?v=glance&s=books). 
Its obviously not available now so if your consideration is for the short 
term then this obviously doesn't help.  If its for the long term though 
you're good to go.

> The HTML documentation that comes with
> Saxon is a good reference but is not a tutorial.  This is an argument
> for using JavaScript - its easier to understand and modify for a VB
> programmer than XSL.

Being someone who has gone through the mental transition from procedural to 
declarative/functional styled programming I can assure you that, while it 
does take some effort, its not as big of a deal to make the transition as 
people think it is.  The upside is enormous and the effort fairly minimal. 
Its DEFINITELY worth the effort.  With that said...

The other consideration for a long term solution would be XQuery which gives 
the syntax procedural developers are used to while acting as an excellent 
way to easily query the XML data and process strings etc...  What you 
obviously don't get is the template based matching et al. that is part of 
XSLT.  But it sounds as if you're more concerned with giving developers who 
are coming from a VB/JScript/Javascript background the ability to quickly 
jump into the code base and with minimal effort be able to support it. 
While I disagree with the foundation for the reasoning (again, I just don't 
see the transition to a declarative/functional way of thinking as big of a 
deal as some make it out to be) I do understand that the issue is considered 
by many to be a big deal and as such would definitely recommend XQuery as a 
decent alternative to XSLT for those that do.

I should also reemphasize EXSLT (EXSLT.org) and the fact that Saxon natively 
supports a lot of these functions.  I am currently (along with the project 
lead Pieter Siegers, pietsieg.com) in the early stages of porting Saxon 
7.9.1 over to .NET and can assure you that the processor is very capable of 
performing in a production environment.  And with Chris Bayes (bayes.co.uk) 
excellent regex implementation in Javascript your well on your way to having 
a complete solution that, with minimal effort, should work quite well.  You 
can download his regex libraries from the EXSLT.org site.

Hope I've given you some information you can find helpful!

Best regards,

<M:D/>


> Barry
>
> M. David Peterson wrote:
>
> >Hi Barry,
> >
> >Thanks for the clarification.  Your problem is pretty straight forward 
> >and
> >in fact can be done fairly easily in XSLT 1.0.  But you are correct in 
> >your
> >assumption that 2.0 would be a much more complete and straight forward
> >solution.  If for no other reason (and there are plenty) the regex
> >implementation in 2.0 gives you the much needed ability to do complex 
> >string
> >matching that, while possible in 1.0, is more complex to implement than 
> >is
> >worth the effort.
> >
> >...
> >
> ><M:D/>
> >
> >
>
> --+------------------------------------------------------------------
> XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list
> To unsubscribe, go to: http://lists.mulberrytech.com/xsl-list/
> or e-mail: <mailto:xsl-list-unsubscribe@xxxxxxxxxxxxxxxxxxxxxx>
> --+--
> 

Current Thread