#Felix Zhang, Period 5
#Statistical part of speech tagging - This program tags words based on their most frequently occurring tag in the corpus.
def main():
	en = []
	de = []
	corpusdict = {}
	de = readcorpus("tiger_release_aug07.export")
#	hashtable = findpos(de)
	word = raw_input("Enter word: ")
	findsingleword(word,de)
#	if word in hashtable.keys():
#		print hashtable[word]

def findmaxprob(list): #finds the most frequently occurring tag
	max = 0
	maxtag = ""
	for x in list:
		if x[1] > max:
			max = x[1]
			maxtag = x[0]
	return maxtag

def findsingleword(word,de): #functions the same way as a hashtable, but only stores data for one word - more efficient for single-word inputs
	pos = [[]]
	for x in de:
		list = x.split('\t')
		if list[0][0] is not "#" and list[0][0] is not "%" and list[0][0] not in "1234567890":
			noblanks = removeblanks(list)
			encountered = False
			
			if word == noblanks[1] or word == noblanks[0]:
				if pos == [[]]:
					pos = [[noblanks[2],1]]
				for y in pos:
					if y != []:
						if y[0] == noblanks[2]:
							y[1] = y[1] + 1	
							encountered = True
				if encountered == False:
					pos.append([noblanks[2], 1])
	print word, pos
	print "Most likely tag: " + findmaxprob(pos)
def readcorpus(filename):
	s = open(filename).read().split('\n')[:-1]
	return s
def findpos(readin): 
	list = []
	lookup = {}
	for x in readin:
		list = x.split('\t') #splits line into list
		if list[0][0] != "#" and list[0][0] != "%" and list[0][0] not in "1234567890": #gets rid of "noise" entries - garbled lines
			noblanks = removeblanks(list)
			register(lookup, noblanks[1], noblanks[2])
	return lookup

def removeblanks(list): #gets rid of blank entries when each line is converted into a list
	newlist = []
	for x in list:
		if x != "":
			newlist.append(x)	
	return newlist

def register(table, key, value): #registers the German word in the hashtable as a key, with a list of pairs as values - first element of pair is part of speech tag, second is a numerical value indicating how frequently it appeared
	if key in table:
		invalues = False
		sets = table[key]
		for x in sets:
			if x[0] == value:
				x[1] = x[1] + 1
				invalues = True
			
		if invalues == False:
			table[key].append([value,1])
	else:
		table[key] = [[value,1]]
main()