DOM object to input source for XT and SAXON : reaults report

[Sebastien Sahuc <ssahuc@xxxxxxxxxxxxxx>]

I'm experimenting strange behaviours while injecting DOM object to XSL
Processor :
With both processor (XT and Saxon ), two methods are used to perform
the XSL processing :
        - Method 1 - DOM object is (kind of) 'serialized' into byte[]
to  build an InputSource.
        - Method 2 - the DOMParser (from John Cowan) is used and
regitered as is in the XSLProcessor.

For the details on implementation, see previous discussions around the
same subject in this list.

Results are as follows :
--------------------------------------
! Method #           ! XT    ! SAXON !
--------------------------------------
! Serializing method ! 160ms ! 350ms !
! DOMParser          ! 250ms ! 290ms !
--------------------------------------

The absolute time isn't important, what's significant is the relative
performance : XT and SAXON don't even follow the same logic... I
really don't understand.

I'm ready to share the source if someone want it.

Sebastien

> -----Original Message-----
> From: Kay Michael [mailto:Michael.Kay@xxxxxxx]
> Sent: vendredi 27 août 1999 10:59
> To: 'xsl-list@xxxxxxxxxxxxxxxx'
> Subject: RE: Converting Document object (DOM) into inputsource for
XSL
> pro cessing in XT
> 
> 
> > 
> >   You could write a "DOM to SAX tree walker", i.e. a class
> > that implements org.xml.sax.Parser and that traverses a DOM
> > tree in pre order, and for each Element node visited : 
> > 
> Or you could one that has already been written! See John Cowan's
"DOM
> Parser" on http://www.ccil.org/~cowan/XML/
> 
> (I also wrote one that was part of an earlier version of 
> SAXON, but I can't
> lay my hands on it right now, and John's is just as good).
> 
> Mike Kay
> 
> 
>  XSL-List info and archive:
http://www.mulberrytech.com/xsl/xsl-list
> 


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