wordcount

gistfile1.py

import sys

def calculate_data (line_data):
#	Got a sring and creates dictionary of (word : count)
	dict = {}
	for word in line_data:
		if word in dict:
			dict[word] += 1
		else:
			dict[word] = 1
	return dict

def file_to_dict(file1):
# 	convert file into string by lower case and removing \n
	data = file1.read()
	data = data.replace ('\n', ' ')
	data = data.lower()

#	Split by space
	for word in data:
		line_data = data.split (" ");

#	Sending to function that creates and return dictionary
	return calculate_data (line_data)
	
	
	
def print_top (filename):
	dict1 = {}
	file1 = open (filename,'rU')
	dict1 = file_to_dict (file1) # convert file into dictionary

#	Print the dictionary top20 sorted by values- The solution.
	dict1 =  sorted(dict1.items(), key=lambda (k,v):(v,k), reverse=True)
	for i in range (5):
		print dict1[i]
	
	file1.close ()
	return filename
	
def print_words (filename):
	dict1 = {}
	file1 = open (filename,'rU')
	dict1 = file_to_dict (file1) # convert file into dictionary

#	Print the dictionary sorted- The solution.
	print sorted (dict1.items())
		
	file1.close ()
	return filename
	
def main():
  if len(sys.argv) != 3:
    print 'usage: ./wordcount.py {--count | --topcount} file'
    sys.exit(1)

  option = sys.argv[1]
  filename = sys.argv[2]
  if option == '--count':
    print_words(filename)
  elif option == '--topcount':
    print_top(filename)
  else:
    print 'unknown option: ' + option
    sys.exit(1)

if __name__ == '__main__':
  main()