def test_get_mode(self):
args = [
'-w', 'word', '-f', 'frequency', '-clr', '-cost', 'cost.txt',
'-basic'
]
configurator = Configurator(args)
mode = configurator.get_mode()
self.assertEqual(mode.get(cfg.MODE_CLEAR), 'STDOUT')
self.assertEqual(mode.get(cfg.MODE_COST), 'cost.txt')
self.assertEqual(mode.get(cfg.MODE_BASIC), 'STDOUT')
self.assertIsNone(mode.get(cfg.MODE_CORRECT))
def build(configurator: cfg.Configurator):
analyzer = WordAnalyzer()
analyzer.words = configurator.get_words(verbose=True)
analyzer.frequency_words = configurator.get_frequency_words(
verbose=True)
analyzer.splitter = TextSplitter(analyzer.frequency_words)
analyzer.tree = configurator.get_tree()
analyzer.mode = configurator.get_mode()
analyzer.verbose = configurator.get_verbose()
# This flag specifies that's need add extra information with a word when analyzing words
analyzer.verbose_file = False
# If the word isn't in the dictionary, then its cost is default_cost
analyzer.default_cost = 1000
# How many similar words will be returned by get_similar_words method
analyzer.number_similar_words = configurator.get_configuration_values(
)['similar_words']
# What distance will be used to search for similar words
analyzer.distance = configurator.get_configuration_values()['distance']
# How many parts will be spliced in the get_correct_words method
analyzer.threshold = configurator.get_configuration_values(
)['threshold']
# How many words will be returned by get_correct_words method
analyzer.number_of_corrected_words = configurator.get_configuration_values(
)['number_corrected']
analyzer.max_len = max(len(x) for x in analyzer.words)
# Define dictionary with values like (number: list_of_divisors).
# There are defined all numbers from 1 to max length of all words.
# It's necessary to improve efficiency, because divisors won't calculated again for same number
analyzer.divisors = dict((number, factorize(number))
for number in range(1, analyzer.max_len + 1))
return analyzer