Re: [xsl] XSL vs. Parsing/Filling and Reading from an Object

[Ivan Rubin <iayma at mcsla dot com dot ar>]

Katie,

I'm working on almost the same thing.

I had a system to improve that worked almost exactly as the one you 
described, with a poorer design unfortunately, and a worst separation of 
concerns, but that way. Servlets, JSP, 'third party' content provider.

I started working with cocoon, sax parsing (it was DOM of course) and 
xsl transformations. SAX parsing is fantastic. Performance boost is 
really impressive.

As this system is a circle, not only generating -> transforming -> 
serializing but serializing -> requesting -> applying -> generating -> 
transforming -> and so on, I had a bit of a fight with cocoon flow 
architecture, concentrated on the sitemap, which is the place I think 
they have to put work on. (My personal opinion is that the flow between 
components should be a programmers thing, with a good and flexible xml 
schema, and not a site-master thing. If you want a "web-master" to 
connect things, just build and give him a case utility to generate the 
xml with pipelines definitions.)

On the other side, the framework over which cocoon is based, and tomcat, 
  is rock-solid, and the idea of 'components' is what should make cocoon 
take off.

Hope to encourage you with my primitive english,
(and push any further discussion to the right list)

Ivan Rubin

Katie McNally wrote:

> We are trying to decide which approach to use to create new pages for 
> the site we are building.  The current approach for the existing pages 
> involves a servlet calling a session bean, which sends an XML request to 
> a 3rd party vendor, which returns XML.  The session bean parses the xml 
> into a hashtable and then passes the hashtable back to the servlet.  The 
> servlet puts the data from the hashtable into an object, and the object 
> is placed in the session.  The jsp then retrieves the object from the 
> session to build the page presented to the user.
> 
> We are considering an XSL approach (using the Xalan processor) to 
> transform the XML returned by the 3rd party vendor.  The things we are 
> considering for our decision are performance, time-to-build and 
> complexity of page maintenance/creation of future pages (the current 
> pages are read-only, but future pages may be transactional/editable).  
> The pages currently being designed will involve the retrieval of XML 
> about 35 KB in size (although future pages may require the retrieval of 
> XML greater in size).  The logic necessary to create the read-only pages 
> is medium in complexity, but future pages may require greater complexity.
> 
> If you have any suggestions/input on either approach, I would appreciate 
> it.  Let me know if I need to provide additional information.
> 
> Thanks, Katie
> 
> _________________________________________________________________
> MSN Photos is the easiest way to share and print your photos: 
> http://photos.msn.com/support/worldwide.aspx
> 
> 
> XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list
> 
> 

</ayma>


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