GRT or ST are not the fastest ways to sort in perl anymore:
Bold claim, but in my tests a simple sort wins from 10 to 1000+ data items, with a GRT taking over from 10,000 upto 1 million data items:
P:\test>466286 -N=10
Rate S::M GRT sort
S::M 15958/s -- -22% -60%
GRT 20443/s 28% -- -49%
sort 39790/s 149% 95% --
P:\test>466286 -N=100
Rate S::M GRT sort
S::M 1814/s -- -17% -40%
GRT 2173/s 20% -- -28%
sort 3024/s 67% 39% --
P:\test>466286 -N=1000
Rate S::M GRT sort
S::M 157/s -- -5% -21%
GRT 166/s 6% -- -17%
sort 200/s 27% 20% --
P:\test>466286 -N=10000
Rate S::M sort GRT
S::M 13.2/s -- -2% -3%
sort 13.5/s 2% -- -0%
GRT 13.6/s 3% 0% --
P:\test>466286 -N=100000
Rate sort S::M GRT
sort 0.938/s -- -7% -11%
S::M 1.01/s 7% -- -4%
GRT 1.05/s 12% 4% --
P:\test>466286 -N=1000000
(warning: too few iterations for a reliable count)
(warning: too few iterations for a reliable count)
(warning: too few iterations for a reliable count)
s/iter sort S::M GRT
sort 13.8 -- -8% -22%
S::M 12.7 9% -- -16%
GRT 10.7 29% 19% --
Benchmark
#! perl -slw
use strict;
use Benchmark qw[ cmpthese ];
use Sort::Key::Maker sort_num_str =>
sub { /^(\d+)(.*)/ }, qw(num str);
sub log10{ log( $_[ 0 ]) / log( 10 ) }
our $N ||= 1000;
our $CHECK ||= 0;
our $ITERS = $CHECK ? 1 : -int( log10 $N );
our @data = map{ int( rand 1000 ) . ('A'..'Z')[ rand 26 ] } 1 .. $N;
my %tests = (
sort => q[
my @sorted = sort{ $^W=0; $a<=>$b or $a cmp $b } @data;
print "s:@sorted" if $CHECK;
],
'S::M' => q[
my @sorted = sort_num_str @data;
print "S::M:@sorted" if $CHECK;
],
GRT => q[
my @sorted = map {
substr $_, 4
} sort map {
$^W=0;
pack 'NA*', 0+$_, $_
} @data;
print "GRT:@sorted" if $CHECK;
],
);
cmpthese $ITERS, \%tests;
Also, I got make test failures when building?
C:\Perl\packages\Sort-Key-0.10>nmake test
Microsoft (R) Program Maintenance Utility Version 7.00.9466
Copyright (C) Microsoft Corporation. All rights reserved.
C:\perl5.8.6\bin\perl.exe "-MExtUtils::Command::MM" "-e" "test
+_harness(0, 'blib\lib', 'blib\arch')" t\floats.t t\ints.t t\multi.t t
+\strings.t
t\floats.....ok
t\ints.......ok 3/5
t\ints.......NOK 4# Failed test (t\ints.t at line 17)
# Structures begin differing at:
# $got->[0] = '9.1766357421875'
# $expected->[0] = '9.2138671875'
t\ints.......ok 5/5# Looks like you failed 1 test of 5.
t\ints.......dubious
Test returned status 1 (wstat 256, 0x100)
DIED. FAILED test 4
Failed 1/5 tests, 80.00% okay
t\multi......ok
t\strings....ok
Failed Test Stat Wstat Total Fail Failed List of Failed
----------------------------------------------------------------------
+---------
t\ints.t 1 256 5 1 20.00% 4
Failed 1/4 test scripts, 75.00% okay. 1/21 subtests failed, 95.24% oka
+y.
NMAKE : fatal error U1077: 'C:\perl5.8.6\bin\perl.exe' : return code '
+0xff'
Stop.
Examine what is said, not who speaks -- Silence betokens consent -- Love the truth but pardon error.
Lingua non convalesco, consenesco et abolesco. -- Rule 1 has a caveat! -- Who broke the cabal?
"Science is about questioning the status quo. Questioning authority".
The "good enough" maybe good enough for the now, and perfection maybe unobtainable, but that should not preclude us from striving for perfection, when time, circumstance or desire allow.
