progressive translation

As I understand it, the general XSL translation
process cannot emit anything until the entire
incoming XML document has been parsed because the 
translation rules can potentially require access
to the full parse tree.

However, there will be some applications where 
"chunks" of incoming XML can be processed in 
relative isolation.

The potential benefit is to improve response time
when a user is wanting to view what is being 
translated - they can see the start of the document
before all of it  has been translated.  In cases 
where the whole XML input stream can take some time
to generate, or is large (eg, is being progressively
generated by scanning a large data warehouse), this
improvement could be considerable.  As well, it gives
the user the chance to abort the process if what they
are looking at isn't what they wanted, potentially
saving resources and their time.

One way we can achieve this now is to generate the 
start of the outer container "manually" 
(eg, for HTML:  <HTML><TITLE>...</TITLE><BODY>),
then process and emit the chunks progressively,
end then generate the end of the outer container
manually.

Are there better ways in existence or planned?

Kent Fitch                           Ph: +61 2 6276 6711
ITS  CSIRO  Canberra  Australia      kent.fitch@xxxxxxxxxxxx



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

<- Previous	Index	Next ->
RE: Case usage in element names, at, Rob McDougall	Thread	Re: progressive translation, James Clark
Re: Case usage in element names, at, Betty Harvey	Date	Re: progressive translation, James Clark
	Month

<-prev [Thread] next->	<-prev [Date] next->
Month Index \| List Home