def test_get_word_pair(self):
data = "This is a sentence. Hello. World. Good bye; See you later."
expected_pairs = [("this", "is"),
("is", "a"),
("a", "sentence"),
("good", "bye"),
("see", "you"),
("you", "later")]
text_stream = TextStream(data=data)
parser = WordParser(text_stream)
actual_pairs = []
i = 0
limit = 10
while True:
pair = parser.get_word_pair()
prev,curr = pair
if not curr:
break
actual_pairs.append(pair)
# Check that the sizes match
actual_size = len(actual_pairs)
expected_size = len(expected_pairs)
self.assertTrue(actual_size == expected_size,
msg="Expected Size: %s but got %s" %
(expected_size, actual_size))
# Verify contents match
for expected,actual in zip(expected_pairs, actual_pairs):
self.assertTrue(expected == actual,
msg="Expected %s but got %s" %
(expected,actual))
def test_get_word(self):
# Get all the words in text input file.
file_path = self.sample_input_file
expected_words = self.get_words_from_file(file_path)
# Now try to use WordParser to extract words
text_stream = TextStream(file_path)
parser = WordParser(text_stream)
actual_words = []
while True:
word = parser.get_word()
if word:
actual_words.append(word)
else:
break
self.assertTrue(actual_words == expected_words,
msg="Expected: %s \nActual: %s" %
(expected_words, actual_words))
def get_all_sentences(text):
sentences = []
words = WordParser.get_all_words(text)
temp_words = []
for word in words:
if word.type == WordType.END:
temp_words.append(word)
sentences.append(Sentence(temp_words))
temp_words = []
else:
temp_words.append(word)
return sentences
def suggest_word(suggest_dict, curr_word):
""" A suggestion dictionary.
Args:
suggest_dict: A dict that has the the following form.
current_word -> {next_word -> WordCount object }e
curr_word: The current word.
Returns:
A word count with the the most probable next word or None
if no suggestions are available.
"""
# normalize given word
curr_word = WordParser.normalize_word(curr_word)
if curr_word in suggest_dict:
candidates_dict = suggest_dict[curr_word]
top_word_count = None
for item,word_count in candidates_dict.iteritems():
if not top_word_count:
top_word_count = word_count
continue
if word_count.count > top_word_count.count:
top_word_count = word_count
elif word_count.count == top_word_count.count:
# Randomly pick either word: 50/50 chance
if (random.random() > 0.5):
top_word_count = word_count
else:
# The word is less likely than our current top word
# So do nothing, and just skip ahead
pass
return top_word_count
else:
return None
#!/usr/bin/env python
from amcc_ui import Ui
from blessed import Terminal
from configparser import ConfigParser
from word_parser import WordParser
if __name__ == '__main__':
config = ConfigParser()
config.read('amcc.conf')
# Select the 'main' section from the config file.
parser_config = config['main']
ui_config = config['ui']
term = Terminal()
with term.fullscreen():
ui = Ui(ui_config, term)
parser = WordParser(parser_config, ui)
parser.start()
def eqExpr(self):
word = WordParser('equals')
word.setDiscard(True)
sec = SequenceParser()
sec.add(word, self.operand())
return sec
