def spot_keywords_in_pages(self, pages, keywords):
file_filters = []
for page in pages:
file_filters.append(str(page) + '-*.png')
word_images = sorted(
fnmatch.filter(os.listdir(self.paths["resized_word_images"]),
file_filters))
validate_word_features = self.get_word_features(word_images)
return self.spot_keywords(validate_word_features, keywords)
def train(self, train_pages, save_file_name=''):
file_filters = []
for page in train_pages:
file_filters.append(str(page) + '-*.png')
word_images = sorted(
fnmatch.filter(os.listdir(self.paths["resized_word_images"]),
file_filters))
self.words_features = self.get_word_features(word_images)
if len(save_file_name) != 0:
self.save_word_features(self.words_features, save_file_name)
for wf in self.words_features:
self.learned_words.add(wf[0])
return self.words_features
def test_limit_filter(self):
"""Test expansion limit of `filter`."""
with self.assertRaises(_wcparse.PatternLimitException):
fnmatch.filter(['name'], '{1..11}', flags=fnmatch.BRACE, limit=10)
def test_filter(self):
"""Test exclusion with filter."""
self.assertEqual(fnmatch.filter(['name', 'test'], '*', exclude='test'), ['name'])