def test_get_tree_without_filename(self, output): configurator = Configurator(self.args) tree = configurator.get_tree() output_msg = output.getvalue() self.assertEqual( output_msg, "The bk-tree is building...\n" "The bk-tree built successfully\n") self.assertEqual( tree.tree, BKTree(distance, configurator.get_frequency_words()).tree)
def test_get_tree_correct_load(self, output): configurator = Configurator(self.args_tree) tree1 = bk.BuildBKTree.build_tree(["teststring"]) with open(self.tree_file, 'wb') as file: pickle.dump(tree1, file) tree2 = configurator.get_tree() output = output.getvalue() self.assertEqual( output, f"The bk-tree is loading from {self.tree_file}...\n" "The bk-tree loaded successfully\n") self.assertEqual(tree1.tree, tree2.tree) self.assertNotEqual(BKTree(distance, ['abcdrasdsf']).tree, tree1.tree)
def test_get_tree_without_existing_file_correct_tree(self, output): configurator = Configurator(self.args_tree) tree = configurator.get_tree() output_msg = output.getvalue() self.assertEqual( output_msg, "The bk-tree is building...\n" "The bk-tree built successfully\n" f"The bk-tree is saving to {self.tree_file}...\n" "The bk-tree saved successfully\n") self.assertTrue(os.path.isfile(self.tree_file)) self.assertEqual( BKTree(distance, configurator.get_frequency_words()).tree, tree.tree) self.assertNotEqual(BKTree(distance, ['abcdrasdsf']).tree, tree.tree)
def test_get_tree_with_existing_empty_file(self, output, err): configurator = Configurator(self.args_tree) file = open(self.tree_file, 'w').close() tree = configurator.get_tree() output = output.getvalue() err = err.getvalue() self.assertEqual( output, f"The bk-tree is loading from {self.tree_file}...\n" "The bk-tree is building...\n" "The bk-tree built successfully\n") self.assertEqual( err, "Error: The bk-tree file you specified is empty and can't be loaded\n" ) self.assertEqual( BKTree(distance, configurator.get_frequency_words()).tree, tree.tree) self.assertNotEqual(BKTree(distance, ['abcdrasdsf']).tree, tree.tree)
def test_get_tree_with_existing_wrong_pickling_file(self, output, err): configurator = Configurator(self.args_tree) with open(self.tree_file, 'wb') as file: pickle.dump("teststring", file) tree = configurator.get_tree() output = output.getvalue() err = err.getvalue() self.assertEqual( output, f"The bk-tree is loading from {self.tree_file}...\n" "The bk-tree is building...\n" "The bk-tree built successfully\n") self.assertEqual( err, "Error: You're trying to load from the file not a BK-tree object\n" ) self.assertEqual( BKTree(distance, configurator.get_frequency_words()).tree, tree.tree) self.assertNotEqual(BKTree(distance, ['abcdrasdsf']).tree, tree.tree)
def test_get_tree_with_existing_not_unpickling_file(self, output, err): configurator = Configurator(self.args_tree) with open(self.tree_file, 'w') as file: file.write("teststring") tree = configurator.get_tree() output = output.getvalue() err = err.getvalue() self.assertEqual( output, f"The bk-tree is loading from {self.tree_file}...\n" "The bk-tree is building...\n" "The bk-tree built successfully\n") self.assertEqual( err, "Error: This file of the bk-tree structure doesn't contain any bk-tree structure actually\n" ) self.assertEqual( BKTree(distance, configurator.get_frequency_words()).tree, tree.tree) self.assertNotEqual(BKTree(distance, ['abcdrasdsf']).tree, tree.tree)
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