def _extract_id_from_html(cls, lines, ident):
h = HTMLParser()
suffix = extract_name_suffix(ident)
if suffix:
suffix = suffix.replace('(', '\(', 1).replace(')', '\)', 1)
suffix = '(:?%s)?\s*' % suffix
person_id_string = cls._person_id_string % (rname(clean_name(ident)),
suffix)
re_person_id = re.compile(person_id_string, re.I)
for line in imap(h.unescape, lines):
id_match = re_person_id.search(line)
if id_match:
return int(id_match.group(1))
return None
def produce_persons(cls, role_type, idents_only=False, sans_roles=False):
re_person_start = re.compile('^----\t\t\t------$')
re_person_name = re.compile('^(.*?)\t+(.*)$')
type_path = config.imdb['%s_path' % role_type]
log.info('Loading roles for "%s" from %s' % (role_type, type_path))
f = open(type_path, 'r')
while not re_person_start.match(f.readline()):
pass
person_new = lambda: { 'name' : None, 'id' : None, 'roles' : [] }
person_ident, person = None, person_new()
for line in imap(lambda l: l.strip().decode('latin_1'), f):
if line[:9] == '---------':
log.info('End of File, done importing %s' % role_type)
break
if not person_ident:
name_match = re_person_name.match(line)
if name_match:
person_ident = name_match.group(1)
person['name'] = rname(clean_name(person_ident))
role_ident = name_match.group(2)
elif not line:
if person_ident and person['roles']:
if idents_only:
yield person_ident
else:
person['id'] = cls.person_ident_to_id.get(person_ident)
person['href'] = cls._person_href(person['id'],
ident=person_ident)
if sans_roles:
del person['roles']
yield person
person_ident, person = None, person_new()
continue
else:
role_ident = line
role = cls._parse_role_ident(role_ident, role_type)
if role['title_ident'] in cls.title_ident_to_id:
role['title_id'] = cls.title_ident_to_id[role['title_ident']]
del role['title_ident']
person['roles'].append(role)
f.close()