def test_read(self):
# test read_terms function
self.assertEqual(
autocomplete_me.read_terms("wiktionary.txt").root.weight, -1)
self.assertEqual(
autocomplete_me.read_terms("wiktionary.txt").root.word, None)
self.assertEqual(
autocomplete_me.read_terms("wiktionary.txt").root.maxWeight,
5627187200)
self.assertRaises(ValueError, autocomplete_me.read_terms, " ")
def test_autocomplete_for_wiki(self):
wikiTrie = ac.read_terms("wiktionary.txt")
result = ac.autocomplete("the", wikiTrie, 5)
expected = [(5627187200, 'the'), (334039800, 'they'),
(282026500, 'their'), (250991700, 'them'),
(196120000, 'there')]
self.assertEqual(expected, result)
def test_autocomplete_for_movies(self):
movTrie = ac.read_terms('movies.txt')
result = ac.autocomplete("The", movTrie, 5)
expected = [(623357910, 'The Avengers (2012)'),
(534858444, 'The Dark Knight (2008)'),
(448139099, 'The Dark Knight Rises (2012)'),
(422783777, 'The Lion King (1994)'),
(408010692, 'The Hunger Games (2012)')]
self.assertEqual(expected, result)
class graph:
def random_file(newfile, filename="movies.txt", encoding='UTF-8'):
''' creates a text file that is a random subset of a given file
filename is the original file's name
newfile is the name of the random subset
'''
with open(filename, "r", encoding=encoding) as file:
allLines = file.readlines()
fileSize = random.randrange(10000, 100000)
lines = random.sample(allLines[1:], fileSize)
with open(newfile, "w", encoding=encoding) as file:
for line in lines:
strLine = str(line)
file.write(strLine)
return [newfile, fileSize]
time1, time2, size = [], [], []
for i in range(30):
[randFile, fileSize] = random_file("rand_i", "movies.txt")
#times building trie and records time to list
start_i = time.clock()
trie = ac.read_terms(randFile)
size_i = fileSize
#adds filesize to size list
size.append(size_i)
running1 = time.clock() - start_i
#times matcher and records time to list
time1.append(running1)
newStart_i = time.clock()
ac.autocomplete('T', trie, 5)
running2 = time.clock() - newStart_i
time2.append(running2)
plt.title('Performance by Input Size')
plt.xlabel('Input Size')
plt.ylabel('Execution Time (s)')
plt.grid(True)
plt.xlim(10000, 100000)
plt.ylim(0, 10)
#plot filesize against two time lists with legends
plt.scatter(size, time1, color="red", label="Loading data")
plt.scatter(size, time2, color="blue", label="Matcher")
plt.legend()
plt.show()
words, wordList = [], []
with open(file, 'r') as txt:
next(txt)
for line in txt:
if line != '\n':
item = line.strip().split('\t')
words.append((int(item[0]), item[1]))
words = sorted(words, reverse=True)
for item in words:
if prefix == item[1][0:len(prefix)] and len(wordList) < k:
wordList.append(item)
return wordList
# four tries
wikTrie = autocomplete_me.read_terms('wiktionary.txt')
pokTrie = autocomplete_me.read_terms('pokemon.txt')
babTrie = autocomplete_me.read_terms('baby-names.txt')
movTrie = autocomplete_me.read_terms('movies.txt')
# random tries
randFile = create_random_terms('random.txt', 2000)
randTrie = autocomplete_me.read_terms(randFile)
if __name__ == '__main__':
unittest.main()
import autocomplete_me as ac
import time
import cProfile
#times read_terms() process
begin1 = time.clock()
words = ac.read_terms("movies.txt")
stop1 = time.clock()
diff1= stop1 - begin1
print("Building the Trie takes: {}".format(diff1))
print("\n")
#times autocomplete() process
begin2 = time.clock()
ac.autocomplete("The", words, 5)
stop2 = time.clock()
diff2 = stop2 - begin2
print("Finishing autocomplete for 'The' takes: {}".format(diff2))
print("\n")
#times all functions separately with "The" as search string and "movies.txt"
# as text input
cProfile.run('ac.autocomplete("The", ac.read_terms("movies.txt"), 5)')
def test_autocomplete_for_empty_input(self):
pokTrie = ac.read_terms("pokemon.txt")
result = ac.autocomplete("", pokTrie, 5)
expected = []
self.assertEqual(expected, result)
def test_autocomplete_for_input_not_in_trie(self):
babyTrie = ac.read_terms('baby-names.txt')
result = ac.autocomplete("XXX", babyTrie, 5)
expected = []
self.assertEqual(expected, result)
def test_autocomplete_for_babynames(self):
babyTrie = ac.read_terms('baby-names.txt')
result = ac.autocomplete("L", babyTrie, 5)
expected = [(16709, 'Liam'), (13066, 'Logan'), (10623, 'Lucas'),
(9319, 'Landon'), (8930, 'Luke')]
self.assertEqual(expected, result)
def test_autocomplete_for_pokemon(self):
pokTrie = ac.read_terms("pokemon.txt")
result = ac.autocomplete("Sh", pokTrie, 5)
expected = [(81075, 'Sharpedo'), (55024, 'Shedinja'),
(43597, 'Shaymin'), (42367, 'Shuckle'), (31091, 'Shiftry')]
self.assertEqual(expected, result)