Subject: RE: Beware the count method with Muenchian grouping (was: Testing by counting or positional predicate)
From: "Michael Kay" <mhkay@xxxxxxxxxxxx>
Date: Wed, 17 Jan 2001 15:18:28 -0000
|
Daniel Bowen wrote:
>
> > I did some time runs with MSXML 3.0 and Saxon 6.0.2 (with Sun JDK
> > 1.3) on Windows NT 4.0 Workstation with Sp5.
Thanks very much for collecting this data. It reveals that generate-id() in
Saxon, when using the "tinytree" tree model and a large source document, is
extremely (and quite unnecessarily) inefficient. This will be fixed!
A couple of points about the measurements:
- if you were using Instant Saxon, you were using Microsoft Java, not JDK
1.3 as stated. JDK 1.3 is generally 2-3 times faster.
- you can get timings from Saxon by using -t on the command line.
Saxon 6.1 shouldn't make any difference in this area.
Mike Kay
XSL-List info and archive: http://www.mulberrytech.com/xsl/xsl-list
| Current Thread |
- Re: Beware the count method with Muenchian grouping (was: Testing by counting or positional predicate), (continued)
- Jeni Tennison - Sun, 14 Jan 2001 11:26:55 -0500 (EST)
- Daniel Bowen - Tue, 16 Jan 2001 15:46:30 -0500 (EST)
- Jeni Tennison - Wed, 17 Jan 2001 04:46:34 -0500 (EST)
- Francis Norton - Wed, 17 Jan 2001 09:32:23 -0500 (EST)
- Michael Kay - Wed, 17 Jan 2001 10:16:50 -0500 (EST) <=
- Daniel Bowen - Wed, 17 Jan 2001 09:54:46 -0500 (EST)
- Francis Norton - Sun, 14 Jan 2001 17:38:47 -0500 (EST)
- Francis Norton - Sun, 14 Jan 2001 17:53:39 -0500 (EST)
- Francis Norton - Mon, 15 Jan 2001 06:20:47 -0500 (EST)
|
|