Jenda Krynicky wrote:
>
> From: "John W. Krahn" <[EMAIL PROTECTED]>
> >
> > $ perl -MPOSIX -MTime::Local -e'
> > $local = time;
> > $gm = timelocal( gmtime $local );
> > $sign = qw( + + - )[ $local <=> $gm ];
> > $calc = sprintf "%s%02d%02d", $sign, (gmtime abs( $local - $gm
> > ))[2,1];
> >
> > print "Calculated: $calc\n", strftime( "strftime: %z\n", localtime );
> > ' Calculated: -0800 strftime: -0800
>
> I like this code. It's a lot shorter than mine. It's a bit slower
> though :
>
> Benchmark: timing 100000 iterations of John, Mine...
> John: 7 wallclock secs ( 6.75 usr + 0.00 sys = 6.75 CPU)
> @ 14814.81/s (n=100000)
> Mine: 3 wallclock secs ( 2.90 usr + 0.00 sys = 2.90 CPU)
> @ 34435.26/s (n=100000)
>
> Not that it matters though ... noone will compute this thousand times
> :-)
>
> I guess I'll keep my code in Mail::Sender though.
In my test it didn't do as poorly. However if the %z format works then
POSIX::strftime is much faster.
Benchmark: timing 100000 iterations of Jenda, John, POSIX...
Jenda: 62 wallclock secs (57.66 usr + 1.38 sys = 59.04 CPU) @ 1693.77/s
(n=100000)
John: 85 wallclock secs (78.25 usr + 1.26 sys = 79.51 CPU) @ 1257.70/s
(n=100000)
POSIX: 16 wallclock secs (15.62 usr + 0.91 sys = 16.53 CPU) @ 6049.61/s
(n=100000)
John
--
use Perl;
program
fulfillment
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
