Untitled

from models.py import Email
import getpass, imaplib, email, sys

from django.http import HttpResponse


class NGramSummer(object): #for tallying total ngram count from someone
		def __init__(self, from_person): #constructor constructor... whats your...
			self.from_person = from_person #self is tagging with the "who" it's from
			self.ngrams = dict()

		def add_ngrams(self, new_ngrams): #this is where the counting is going on
			for word in new_ngrams:
				if word in self.ngrams:
					self.ngrams[word] += new_ngrams[word]
				else:
					self.ngrams[word] = new_ngrams[word]

		def get_ngrams(self):
			return self.ngrams

	# NGramCounter builds a dictionary relating ngrams (as tuples) to the number
	# of times that ngram occurs in a text (as integers)
	class NGramCounter(object):

	  # parameter n is the 'order' (length) of the desired n-gram
	  def __init__(self, text):
	    self.text = text
	    self.ngrams = dict()

		# feed method calls tokenize to break the given string up into units
	  def tokenize(self):
	    return self.text.split(" ")

	  # feed method takes text, tokenizes it, and visits every group of n tokens
	  # in turn, adding the group to self.ngrams or incrementing count in same
	  def parse(self):

	    tokens = self.tokenize()
	    #Moves through every individual word in the text, increments counter if already found
	    #else sets count to 1
	    for word in tokens:
			if word in self.ngrams:
				self.ngrams[word] += 1
			else:
				self.ngrams[word] = 1

	  def get_ngrams(self):
	    return self.ngrams

def get_first_text_part(msg): #setup to cleanup all text
	    maintype = msg.get_content_maintype()
	    if maintype == 'multipart':
	        for part in msg.get_payload():
	            if part.get_content_maintype() == 'text':
	                return part.get_payload()
	    elif maintype == 'text':
	        return msg.get_payload()


def index(request):

	certain_file = open("lists/list_certain.txt", "r")
	negation_file = open("lists/list_negation.txt", "r")
	warm_file = open("lists/list_warm.txt", "r")
	cold_file = open("lists/list_cold.txt", "r")
	firstPOV_file = open("lists/list_1POV.txt", "r")
	secondPOV_file = open("lists/list_2POV.txt", "r")
	thirdPOV_file = open("lists/list_3POV.txt", "r")

	certain_list = [word.strip() for word in certain_file.readlines()]
	negation_list = [word.strip() for word in negation_file.readlines()]
	warm_list = [word.strip() for word in warm_file.readlines()]
	cold_list = [word.strip() for word in cold_file.readlines()]
	firstPOV_list = [word.strip() for word in firstPOV_file.readlines()]
	secondPOV_list = [word.strip() for word in secondPOV_file.readlines()]
	thirdPOV_list = [word.strip() for word in thirdPOV_file.readlines()]


	#loading profile for login
	M = imaplib.IMAP4_SSL('imap.gmail.com')
	M.login("wilo@gmail.com", "pass")
	# M.select("[Gmail]/Sent Mail")
	M.select("[Gmail]/All Mail")

	xml_template = "<email><from>{sender}</from><to>{to}</to><date>{date}</date><subject>{subject}</subject><body>{body}</body><certain>{certain}</certain><negation>{negation}</negation><warm>{warm}</warm><cold>{cold}</cold><firstPOV>{firstPOV}</firstPOV><secondPOV>{secondPOV}</secondPOV><thirdPOV>{thirdPOV}</thirdPOV><ngrams>{ngrams}</ngrams></email>"

	person_summary = "<Summary><ngramSUM>{summery}</ngramSUM></Summary>"

	theperson = ["alinjen@bellsouth.net", "ajarnp@gmail.com", "alinjen@me.com", "dlbergman@gmail.com", "donjen@bellsouth.net", "hljen@bellsouth.net", "trixietree@hotmail.com", "wendycantdrive@hotmail.com", "avery@averymax.com", "chris.laniosz@gmail.com", "frannie.hall@gmail.com", "ian.oliver@flawlessfuture.com", "jasonaston@gmail.com", "seatubers@gmail.com", "seatubers@gmail.com", "idralcar@hotmail.com", "matthew.d.rader@gmail.com", "rader@matthewrader.com", "mslyssa@gmail.com", "guerrajmichael@gmail.com", "pamela@pamelareed.com", "justkeepgoing@excite.com", "studio@reedandrader.com", "ryan@letsneverdie.net", "4stepan@gmail.com", "cmae.oliver@gmail.com", "conniemae.olive@gmail.com", "larissa_bemis@yahoo.com", "larissarbemis@gmail.com", "lbemis@apple.com", "sushionthego@gmail.com", "acm466@nyu.edu", "lia.martinez@nyu.edu", "lia@potiondesign.com", "kiwi@smirkyplop.com", "roisin.stack@gmail.com", "ohannamarie@gmail.com", "bryan.baxter@gmail.com", "davidestici@hotmail.com", "genny.hoffman@gmail.com", "sheenamcneal@gmail.com", "daniel.shiffman@gmail.com", "daniel.shiffman@nyu.edu", "dan.osullivan@nyu.edu",  "edward.gordon@nyu.edu", "marianne.petit@nyu.edu", "midori.yasuda@nyu.edu", "nh19@nyu.edu", "rob.ryan@nyu.edu", "EFarnon@wsgc.com", "lori@scoreatthetop.com"]

	for person in theperson:
		print "Searching for:", repr(person)
		type, data = M.search(None, 'FROM', person) #Gets ALL messages
		summer = NGramSummer(person) #a NGramSummer object,this will sum all the ngrams together
		new = open(person+"_from.xml", 'w')
		for num in data[0].split(): #Loops through all messages
		    yp, data = M.fetch(num, '(RFC822)') #Pulls Message
	    	    msg = email.message_from_string(data[0][1]) #Puts message into easy to use python objects
	    	    _from = msg['from'] #pull from
	    	    _to = msg['to'] #pull to
	    	    _subject = msg['subject'] #pull subject
	    	    _date = msg['date']
	    	    _body = get_first_text_part(msg) #pull body
	            certain_sum = 0
	            negation_sum = 0
	            warm_sum = 0
	            cold_sum = 0
	            firstPOV_sum = 0
	            secondPOV_sum = 0
	            thirdPOV_sum = 0
	            for word in _body.split(' '):
					word = word.lower()
					if word in certain_list:
						certain_sum += 1
					if word in negation_list:
						negation_sum += 1
					if word in warm_list:
						warm_sum += 1
					if word in cold_list:
						cold_sum += 1
					if word in firstPOV_list:
						firstPOV_sum += 1
					if word in secondPOV_list:
						secondPOV_sum += 1
					if word in thirdPOV_list:
						thirdPOV_sum += 1
	    	    if _body:
		        ngrams = NGramCounter(" ".join(_body.strip(">").split()))
			ngrams.parse()
			_feed = ngrams.get_ngrams()
			summer.add_ngrams(_feed)
			email_obj = Email()
			email_obj.from = _from
			email_obj.to = _to
			email_obj.subject = _subject
			email_obj.date = _date
			email_obj.body = _body
			email_obj.certain = certain_sum
			email_obj.negation = negation_list
			email_obj.warm = warm_sum
			email_obj.cold = cold_sum
			email_obj.firstPOV = firstPOV_sum
			email_obj.secondPOV = secondPOV_sum
			email_obj.thirdPOV = thirdPOV_sum
			email_obj.save()
			#print _feed

	    	        my_message = xml_template.format(sender=_from, to=_to, date=_date, subject=_subject, ngrams=_feed, body=_body, certain=certain_sum, negation=negation_sum, warm=warm_sum, cold=cold_sum, firstPOV=firstPOV_sum, secondPOV=secondPOV_sum, thirdPOV=thirdPOV_sum)
	    	        new.write(my_message)
	    	        print repr(my_message)

	    # print 'Content-Type:',msg.get_content_type()
	# last_message = xml_template.format(sender="_from", to=_to, date=_date, subject=_subject, body=_body, ngrams=_summer.get_ngrams())

	# new.write('------summmary----------')
	# new.write(str(summer.get_ngrams()))
	# print '------summmary----------'
			the_summary = summer.get_ngrams()
			sum_persons = person_summary.format(summery=the_summary)
			new.write(sum_persons)
	        print repr(summer.get_ngrams())

	new.close()
	M.close()
	M.logout()

	return HttpResponse("Hello, world. You're  at the mailerz index")