hm

#!/usr/bin/env perl
use strict;
use warnings;
use Getopt::Std;

my @marks = ('', '.'x(int(rand(3))+1), '!');
my %options;
my $beg;
my $mid;
my $end = '';

sub parse($) {
	my $word = shift;
	$beg = substr($word, 0, -1);
	$mid = substr($word, -1, 1);
}

getopts('u', \%options);

if ($options{u}) {
	push(@marks, '…');
} if (@ARGV == 1) {
	parse(shift);
} elsif (@ARGV > 1) {
	$beg = shift;
	$mid = shift;
	$end = shift || '';
} else {
	parse((split(/\//, $0))[-1]);
}
print(($beg.$mid x(int(rand(4))+1).$end)x(int(rand(12)/10)+1).$marks[rand@marks]."\n")

__END__

=head1 NAME

hm - say hm

=head1 SYNOPSIS

B<hm> [ B<-u> ] [ I<string> | I<begin> I<middle> [I<end>] ]

=head1 DESCRIPTION

B<hm> says 'hm' in a variety of ways, such as 'hmm...', 'hm!' or 'hmmhmm'.

It uses three strings, which we will call begin, middle and end, and which are
processed roughly like this:

phrase = I<begin> [repeated 0 to 3 times: I<middle>] I<end>;  Print phrase
one or two times;  Print punctuation.

If I<begin>, I<middle> and I<end> are specified on the commandline, they will
be directly put into the output. If I<end> is left out, it will be blank.

Otherwise, B<hm> takes its first argument (I<string>) or its command name
(B<hm>), uses its last character for middle, the rest for begin, and leaves end
empty.

=head1 OPTIONS

=over

=item B<-u>

Use the unicode variant of '...' in the punctuation

=back

=head1 AUTHOR

Copyright (c) 2009 Birte Kristina Friesel <derf@derf.homelinux.org>

L<https://derf.homelinux.org/~derf/code/hm>

=head1 LICENSE

  0. You just DO WHAT THE FUCK YOU WANT TO.