# No.29 Perl Sample (code key - 41)
# Learning Perl: Count Word Types 1 （単語のタイプ数を数える）
# Tomonori Nagano <tnagano@gc.cuny.edu>
# Last Update: January 18, 2008
#
# This file is encoded in Unicode (UTF-8). If you see gibberish characters, 
# please re-encode the file in utf-8. 
# 

# this sample script is simplified for pedagogical purposes. Those who are
# interested in advanced Perl programming are encouraged to consult with 
# relevant sections of "Perl Cookbook" by Christiansen & Torkington

my $numTypes = 0 ;

while(<>){						# this reads the default file line by line
	chomp ;						# delete the end-of-line character
	$_ = lc($_) ;				# change to lower case
	s/([^\- a-zA-Z0-9])/ \1 /g;	# a negative definition of punctuation
	@words = split (" ",$_) ;	# split into words
	foreach (@words) {			# process each word
		next if /[^a-zA-Z]/ ;	# skip if words contain non-alphabet
		if (!defined $seen{$_}) {	# if you haven't seen the word before
			$numTypes++ ;			# increase the type frequency
			$seen{$_} = 1 ;			# define a new hash key
		}
	}		
}

print "numTypes\t$numTypes\n" ;