#! /usr/bin/env python

from useful import tokenize

# Declare file to be worked with
textfile = "furniture.txt"

# Open file & read first line
file = open(textfile,'r')
line = file.readline()

# initialize variables:
Bigrams = {}
prev_word = "START"

while line:
    line = line.rstrip()
    # tokenize the text:
    tokens = tokenize(line)

    # loop over words in input:
    for word in tokens:
	# concatenate words to get bigram:
	bigram = prev_word + ' ' + word
	if bigram in Bigrams:
	    Bigrams[bigram] += 1
	else:
	    Bigrams[bigram] = 1
	# change value of prev_word
        prev_word = word

    line = file.readline()
file.close()

# Write bigrams to output file:
output_file = open('bigrams.txt','w')
for bigram in Bigrams:
    count = Bigrams[bigram]
    output_file.write(str(count)+'\t'+bigram+'\n')
output_file.close()