Re: Unusual sorting requirements; comparing three implementations.

[BrowserUk]

Ignore. Dumb mistake. Corrected:

 ...

sub x {
    my @sorted = map $_->[1],
    sort{ $a->[0] cmp $b->[0] } map[
        $_->title eq 'Manager' ? 'A'.$_->name : 'B'.$_->name, $_
    ], @employees;
}

 cmpthese( -1, {
     obvious    => \&obvious,
     subtle     => \&subtle,
     functional => \&functional,
     x          => \&x,
 });

__END__
C:\test>junk87
             Rate    obvious     subtle functional          x
obvious     185/s         --       -15%       -60%       -83%
subtle      218/s        18%         --       -54%       -80%
functional  469/s       153%       115%         --       -58%
x          1111/s       500%       411%       137%         --

[afoken]

sub x {
    my @sorted = map $_->[1],
    sort{ $a->[0] cmp $b->[0] } map[
        $_->title eq 'Manager' ? 'A'.$_->name : 'B'.$_->name, $_
    ], @employees;
}

Faster, but not correct for the original problem. tobyink's Test::More script tested for ->title =~ /Manager/, not for ->title eq 'Manager', and so all those non-generic Manager employees, like the Sales Manager, the Finance Manager, and the Marketing Manager will be sorted as NON-Manager employees. tobyink's benchmark script wrongly reduces the possible titles to "Staff" and "Manager".

Alexander

[BrowserUk]

Correcting that makes barely any difference to the performance:

sub x {
    my @sorted = map $_->[1],
    sort{ $a->[0] cmp $b->[0] } map[
        $_->title =~ /Manager/ ? 'A'.$_->name : 'B'.$_->name, $_
    ], @employees;
}

 cmpthese( -1, {
     obvious    => \&obvious,
     subtle     => \&subtle,
     functional => \&functional,
     x          => \&x,
 });
C:\test>junk87
             Rate    obvious     subtle functional          x
obvious     183/s         --       -18%       -60%       -83%
subtle      224/s        22%         --       -51%       -79%
functional  458/s       150%       105%         --       -57%
x          1070/s       483%       378%       134%         --

[tobyink]

Not sure if your computer's architecture is radically different to mine or something, but for me, this technique is still slightly slower than the functional approach, and much slower than keysort.

Read more... (2 kB)

Results:

              Rate     obvious       subtle pure_schwarz  functional     sortkey
obvious      133/s          --         -28%         -60%        -65%        -80%
subtle       186/s         39%           --         -45%        -52%        -72%
pure_schwarz 337/s        153%          81%           --        -12%        -50%
functional   383/s        187%         107%          14%          --        -43%
sortkey      671/s        403%         262%          99%         75%          --