Hi Andiih, I had one more thought. I don't usually setTemplateURL abd
ibstead use setTemplateElement and have the template locally in a hidden
text area. I was wondering if it might be the difference because of the
network communication that occurs in your case. Might be worth a try to see
if thats the extra time drain.
Also JavaScriptTemplates from trimpath is a really great template engine as
well and really not too different syntatically than jTemplate so you should
be able to modify your templates relatively easily to use them if that what
you decide to do.
On Fri, Jun 27, 2008 at 5:53 AM, Andiih <[EMAIL PROTECTED]> wrote:
>
> Machine A
> FF3 xml to json 1380, Template Processing 619
> IE6 xml to json 1782, Template Processing 22593
>
> Machine B
> IE7 xml to json 765, Template Processing 8579
> FF2 xml to json 2094, Template Processing 1492
> FF3 Beta xml to json 1151, Template Processing 1054
>
> Machine C
> IE 7 Xml to json 1187, Template Processing 9344
> FF 2 Xml to json 3281, Template Processing 1328
>
> OK - its the tempaltes. Will try the other template engine above...or
> perhaps I should consider xml/xsl and skip the json step ? I think I
> saw a jQuery xls processor somewhere...
>
> On Jun 27, 12:03 am, "chris thatcher" <[EMAIL PROTECTED]>
> wrote:
> > I think you are measuring the combined time. Try this:
> >
> > var xmltojson_start = new Date().getTime();
> > ret = $.xmlToJSON(responseXML);
> > var xmltojson_stop = new Date().getTime();
> >
> > var template_start = new Date().getTime();
> > $('#output2').processTemplate(ret);
> > var template_stop = new Date().getTime();
> > alert("****************************************"+
> > "\n\tXml to Json: " + (xmltojson_stop - xmltojson_start) +
> > "\n\tTemplate Processing: " + (template_stop - template_start)
> +
> > "\n********************************************");
> >
> >
> >
> > On Thu, Jun 26, 2008 at 4:36 PM, Andiih <[EMAIL PROTECTED]> wrote:
> >
> > > Jack, I'll take a look
> > > Chris, I *think* the timing code I've added in my sample above is only
> > > timing the template step ?
> >
> > > I've had the opportunity to try this out this afternoon on a number of
> > > machines in the organization, and noted a couple of things...
> > > Its quick in FF3 (which I didn't expect given what I have been
> > > reading)
> > > Its *usually* a lot slower in IE than FF, but some machines are the
> > > opposite.
> >
> > > So, I need to revise the question : what browser configuration
> > > changes, plugins, settings etc might make a *huge* difference to inter-
> > > browser performance ?
> >
> > > I'll see if JavaScriptTemplates looks like I can integrate it, and re-
> > > test.
> >
> > > TIA
> >
> > > Andrew
> >
> > > On Jun 26, 6:02 pm, "chris thatcher" <[EMAIL PROTECTED]>
> > > wrote:
> > > > Andiih, just curious becuase I use jtemplates and havent seen that
> issue,
> > > > though can you verify that the slow code is not 'xmlToJSON'. I have
> seen
> > > > the marshalling process take up a lot of time in IE when the xml is
> > > > substantial in size .
> >
> > > > Thatcher
> >
> > > > On Thu, Jun 26, 2008 at 12:30 PM, Jack Killpatrick <[EMAIL PROTECTED]>
> > > wrote:
> >
> > > > > I don't know how it compares as far as speed, but you might want to
> try
> > > > > this:
> >
> > > > >http://code.google.com/p/trimpath/wiki/JavaScriptTemplates
> >
> > > > > We've been using it for more than a year on many of our projects,
> > > including
> > > > > some with large table row outputs, and it's worked for us.
> >
> > > > > - Jack
> >
> > > > > Andiih wrote:
> >
> > > > >> I am using jTemplates (jquery-jtemplates.js) to render a large xml
> > > > >> response by running xmlToJSON then processTemplate. Although the
> code
> > > > >> works fine, and performance in FF2.0 is acceptable (2672ms) on my
> test
> > > > >> system, I am getting a result of 9827ms when running in IE7. Is
> there
> > > > >> a known performance issue with jtemplates ? Are other templte
> modules
> > > > >> better ?
> >
> > > > >> (p.s. the real world code uses jQuery Form plugin and web
> services,
> > > > >> but the sample below reproduces the issue)
> >
> > > > >> Code, and template follow.
> >
> > > > > ... snip...
> >
> > > > --
> > > > Christopher Thatcher
> >
> > --
> > Christopher Thatcher
>
--
Christopher Thatcher
