On 12/8/05, Perrin Harkins <[EMAIL PROTECTED]> wrote:
> On Thu, 2005-12-08 at 18:24 +0000, Mark Stosberg wrote:
> > Also, at least according to the HT::Compiled benchmark, it's about 10
> > times faster than TT.
>
> It's 3 times faster.  By default, the benchmark tests creating a new
> Template object every time, which no one should ever do.  Switching from
> the "all" test to the "output" tests fixes this.

The TT template in the benchmark also uses INCLUDE instead of PROCESS
to include a template.  INCLUDE will localize all parameters which
takes a bit of a performance hit.  Most users will use PROCESS instead
and only use INCLUDE in specific situations where it is important.  It
doesn't affect the results of the benchmark a great deal, but it does
make the work that TT does closer to what HTC is doing.

Also, for those that are interested, the TT plugin when used in a
persistent environment (mod_perl, SpeedyCGI, etc...) can be setup to
only create the Template object once and reuse it for all future
requests.  See the section on Singleton support in the docs.

Cheers,

Cees

---------------------------------------------------------------------
Web Archive:  http://www.mail-archive.com/[email protected]/
              http://marc.theaimsgroup.com/?l=cgiapp&r=1&w=2
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to