Gutenberg Ipsum

gutenberg_ipsum.pl

#!/usr/bin/perl -w
# Modified from Dr. Drang's original script at http://www.leancrew.com/all-this/2011/02/dissociated-darwin/

use Games::Dissociate;

# Choose the corpus file
if ($#ARGV == -1) {
  $corpus = "totc.txt";
} else {
  $corpus = $ARGV[0];
}

# Slurp in the given corpus as a single string.
open(my $fh, "$ENV{HOME}/bin/gutenberg_ipsum/words/" . $corpus) or die "Can't open";
{local $/; $corpus = <$fh>;}

# Dissociate the corpus, using word pairs, and return 15-50 pairs.
$length = int(15 + rand(35));
$dis = dissociate($corpus, -2, $length);

# Remove quotes and other paired characters, since there might be some that are unmatched
# But this is an incredibly clunky fix. If I had more time/better Perl chops, I'd probably build some algorithm to find unmatched quotes or parentheses and insert them randomly in the text. But that's hard :)
$dis =~ s/[\"\[\]\_\(\)]//gm;

# Capitalize the first word and end it with a period.
$dis =~ s/^(.)/\u$1/;
$dis =~ s/[.);:?'", -]+$/./;

print $dis;