def get_titles(self):
possible_titles = self.parser.parse()
logger.debug("Found possible titles %s for title %s" % (possible_titles, self.title))
possible_exact_titles = flatten(
[self.db.query_title(title, year) for title, year in possible_titles])
possible_fuzzy_titles = flatten(
[self.db.query_fuzzy(title, year) for title, year in possible_titles])
titles = set(possible_exact_titles).intersection(set(possible_fuzzy_titles))
logger.debug("Found normalized titles %s" % titles)
return titles
def populate_actors_file(input_files=INPUT_FILES,
write_file=OUTPUT_FILE, cleanup=True):
data = set(
map(tuple,
flatten(
map(clean_scraped_data, input_files)
)
)
)
data_dicts = map(lambda x: {
'f_name': x[0],
'l_name': x[1],
}, data)
write_scraped_data(write_file, data_dicts)
if cleanup:
map(os.remove, INPUT_FILES)
def extract_credits(path, wildcard=EXT):
"""
path: Input path where processed movie is contained
Performs OCR on each image found at path/*wildcard.
The text is then cleaned and normalized to the closest
match to known actors and actresses.
A set of names is returned.
"""
logger.debug("Extracting credits at %s" % path)
cleaner = StringCleaner()
img_paths = glob("%s/*%s" % (path, wildcard))
img_text = [_text_from_img(img_path, path) for img_path in img_paths]
logger.debug("img_text %s" % img_text)
clean_text = flatten(map(cleaner.clean, img_text))
logger.debug("clean_text %s" % clean_text)
return clean_text
def _split_text(clean_text):
"""
split text into terms for fuzzy matching
"""
return flatten(map(
lambda tokens: map(parse_name, tokens), clean_text))
