])
PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
-JSON_LD_RE = r'(?is)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>'
+JSON_LD_RE = r'(?is)<script[^>]+type=(["\']?)application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>'
def preferredencoding():
path = path.decode('utf-8')
if not isinstance(path, compat_str) or not path:
return None
- if re.match(r'^(?:https?:)?//', path):
+ if re.match(r'^(?:[a-zA-Z][a-zA-Z0-9+-.]*:)?//', path):
return path
if isinstance(base, bytes):
base = base.decode('utf-8')
'gv': 'glv',
'ha': 'hau',
'he': 'heb',
+ 'iw': 'heb', # Replaced by he in 1989 revision
'hi': 'hin',
'ho': 'hmo',
'hr': 'hrv',
'hz': 'her',
'ia': 'ina',
'id': 'ind',
+ 'in': 'ind', # Replaced by id in 1989 revision
'ie': 'ile',
'ig': 'ibo',
'ii': 'iii',
'wo': 'wol',
'xh': 'xho',
'yi': 'yid',
+ 'ji': 'yid', # Replaced by yi in 1989 revision
'yo': 'yor',
'za': 'zha',
'zh': 'zho',