RE: [xsl] Processing a huge XML document

> 	I am quite puzzled as to which method would be faster
> to transform
> a 50 MB XML file: an XSLT, or using SAX call-backs.

If your transformation is serial, then a SAX application will certainly be
much faster, as you don't have to build the tree in memory (the tree for a
50Mb file will be vast).

You might like to look at Saxon's "preview mode" if you want an XSLT
solution, this essentially does an XSLT transform on one subtree at a time,
avoiding the need to build the complete tree in memory.

Mike Kay
Software AG


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

Current Thread
Re: [xsl] generating a repeatable unique id, (continued) Michael Beddow - Sun, 29 Jul 2001 22:44:05 +0100 Mahesh V Kondwilkar - Sun, 29 Jul 2001 18:28:23 -0500 (CDT) Kurt Cagle - Sun, 29 Jul 2001 20:34:08 -0700 Mahesh V Kondwilkar - Mon, 30 Jul 2001 01:54:45 -0500 (CDT) Michael Kay - Mon, 30 Jul 2001 09:24:45 +0100 <= Sebastian Rahtz - Sun, 29 Jul 2001 21:33:32 +0100 Robert Koberg - Sun, 29 Jul 2001 13:58:48 -0700 Michael Beddow - Sun, 29 Jul 2001 22:39:21 +0100 Sebastian Rahtz - Mon, 30 Jul 2001 11:28:10 +0100

Current Thread

Re: [xsl] generating a repeatable unique id, (continued)

<- Previous	Index	Next ->
Re: [xsl] Processing a huge XML doc, Mahesh V Kondwilkar	Thread	Re: [xsl] generating a repeatable u, Sebastian Rahtz
RE: [xsl] SAXON : EOF, Michael Kay	Date	RE: [xsl] choose document() string , Michael Kay
	Month

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