You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@xalan.apache.org by "Voytenko, Dimitry" <DV...@SECTORBASE.COM> on 2000/09/08 03:09:35 UTC

XSLT and Input XML

I use XalanJ
As I understand XSLT processor works on the prepared DOM tree. Is any way to
extend source tree during XSLT-processor work. I found extension
RowSetLocator. It's supposed to retrieve data from DB and extend source tree
with it. But it's not clear how it's called (how XSLT-processor decides that
it should use its XLocator, or something like that). Correct me if I wrong.

Thank you


Performance: attribute sets, named templates, entities, etc

Posted by Alistair Hopkins <al...@berthengron.co.uk>.
I am using Xalan to build a website and the stylesheets are getting fairly 
big, as each stylesheet copes with quite a wide range of data 
elements.  They're still processing nice and fast once they are compiled to 
StylesheetRoot, but I'm not sure what the relative costs of various nice 
features are.

1/ Attribute sets.  Are these inserted into the element during compilation 
or processing?  Is there any reason not to use them everywhere (because 
they make html so easy)?

2/ Named templates.  I want to use named templates for little snippets of 
'stuff': for example, a 'compulsory field' marker, or a navigational 
image.  This lets me change my mind later.  Is the template inserted at 
compile or runtime?  What is the overhead of calling templates?

3/ Variables.  I use a lot of 'sub stylesheets' for different data 
entities, then call them from a master which has a variable declaring which 
servlet it is invoked by.  This lets me do 'loopback URLs' without worrying 
about what the current servlet is.  Is this costing me?