RE: in memory tree representation SAXON

If you supply the input to the Saxon transformation as a StreamSource or
SAXSource, then the result tree will be built from scratch each time.
You can arrange to reuse the same tree by building the tree yourself:

XPathEvaluator xp = new XPathEvaluator();
DocumentInfo doc = xp.setSource(new StreamSource(...))

transformer.transform(doc, ...)

(this is Saxon 7.x, it's a bit more cumbersome in 6.x., but you can
still achieve it)

If you do two transformations on the same document tree using the same
compiled stylesheet, then Saxon will reuse the indexes defined using
xsl:key.

Michael Kay


> -----Original Message-----
> From: owner-xsl-list@xxxxxxxxxxxxxxxxxxxxxx 
> [mailto:owner-xsl-list@xxxxxxxxxxxxxxxxxxxxxx] On Behalf Of 
> Philippe Dubreuil
> Sent: 06 June 2003 12:00
> To: xsl-list@xxxxxxxxxxxxxxxxxxxxxx
> Subject:  in memory tree representation SAXON
> 
> 
> Hi,
> I 'd like to know, if the in-memory tree representation in saxon is 
> removed or not after each transformation.
> Is it  possible to deduce the in-memory  result tree 
> representation from 
> the first one. i'd like to transform the first one to get the 
> result tree.
> 
> I'm trying to find a way to conserve the indexation through a 
> transformations composition.
> 
> thanks
> 
> phil. Dubreuil
> 
> 
>  XSL-List info and archive:  http://www.mulberrytech.com/xsl/xsl-list
> 


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