Re: improving the efficiency of a script

[McDarren]

My dictionary (/usr/share/dict/words) only has 96274 words :)

But anyway, here is my go at this - I'm not sure how efficient (or otherwise) it is - but it seems to work :)

Update 1: blah - I just realised that it doesn't work at all - the shuffle isn't doing anything. I'll update it again later when I get it working.

Update 2: - okay, working now :) (I think)

#!/usr/bin/perl -w
use strict;
use List::Util qw(shuffle);

my $dict = '/usr/share/dict/words';

open DICT, "<", $dict or die "Cannot open $dict:$!\n";

my %words;

WORD:
while (my $word = <DICT>) {
    chomp $word;
    my ($letter) = $word =~ /^(\w)/;
    next WORD if !$letter;
    push @{$words{$letter}}, $word;
}

close DICT;

foreach my $letter (sort keys %words) {
    @{$words{$letter}} = shuffle(@{$words{$letter}});
    for my $number (1 .. 100) {
        print "$words{$letter}[$number]\n";
    }
}

Cheers,
Darren :)

[lima1]

I'm not sure how efficient (or otherwise) it is

Much better than OPs solution ;) but time and space requirements for the data preparation is still O(n). much more than 0(m * log(n)) and O(m), respectively (once again, for sorted dictionaries only).

So if dictionary is unsorted, the memory requirements are not a problem and if this task is not often repeated, yours is the best solution IMHO.