propername.py

import re

with open('faulkner.txt', 'r') as infile:
	complete_lines = infile.read().split('\n')

stripped_lines = map(lambda x: x.strip(), complete_lines)
lines = filter(lambda x: x and not x.isdigit(), stripped_lines)

print len(lines)

def is_proper_name(prior_word, word):
	proper_name = False
	if word[0].isupper() and word[1:].islower():
		proper_name = True
	if prior_word[-1] in set(['.', '?', '!']):
		proper_name = False
	return proper_name

def get_proper_names(text):
	tokens = text.split()
	output = list()
	for i in range(1, len(tokens)):
		if is_proper_name(tokens[i-1], tokens[i]):
			output.append(tokens[i])
	# starts_with_cap = filter(
	# 	lambda w: w[0].isupper() and w[1:].islower(),
	# 	tokens
	# )
	return output

all_names = list()
for l in lines:
	names = get_proper_names(l)
	all_names.extend(names)

unique_names = list(set(all_names))

new_text = '\n'.join(lines)

def make_reg_exp(name):
	regExp = r'\b' + re.escape(name) + r'\b'
	return regExp

for name in unique_names:
	new_text = re.sub(make_reg_exp(name), 'John', new_text)

print new_text