Update changelog.

[youtubedl] / youtube_dl / utils.py

diff --git a/youtube_dl/utils.py b/youtube_dl/utils.py
index 2fe9cf585db817e1d86831c71ef28502b4a16ee5..d2d3c1a9fde82510f47b8fcd43c726b51bedf9d6 100644 (file)
--- a/youtube_dl/utils.py
+++ b/youtube_dl/utils.py
@@ -39,6 +39,7 @@ from .compat import (
     compat_HTMLParser,
     compat_basestring,
     compat_chr,
     compat_HTMLParser,
     compat_basestring,
     compat_chr,

+    compat_cookiejar,

     compat_ctypes_WINFUNCTYPE,
     compat_etree_fromstring,
     compat_expanduser,
     compat_ctypes_WINFUNCTYPE,
     compat_etree_fromstring,
     compat_expanduser,


@@ -49,7 +50,6 @@ from .compat import (
     compat_os_name,
     compat_parse_qs,
     compat_shlex_quote,
     compat_os_name,
     compat_parse_qs,
     compat_shlex_quote,

-    compat_socket_create_connection,

     compat_str,
     compat_struct_pack,
     compat_struct_unpack,
     compat_str,
     compat_struct_pack,
     compat_struct_unpack,


@@ -82,7 +82,7 @@ def register_socks_protocols():
 compiled_regex_type = type(re.compile(''))
 
 std_headers = {
 compiled_regex_type = type(re.compile(''))
 
 std_headers = {

-    'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:10.0) Gecko/20150101 Firefox/47.0 (Chrome)',
+    'User-Agent': 'Mozilla/5.0 (X11; Linux x86_64; rv:59.0) Gecko/20100101 Firefox/59.0',

     'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
     'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
     'Accept-Encoding': 'gzip, deflate',
     'Accept-Charset': 'ISO-8859-1,utf-8;q=0.7,*;q=0.7',
     'Accept': 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8',
     'Accept-Encoding': 'gzip, deflate',


@@ -184,6 +184,7 @@ DATE_FORMATS_MONTH_FIRST.extend([
 ])
 
 PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"
 ])
 
 PACKED_CODES_RE = r"}\('(.+)',(\d+),(\d+),'([^']+)'\.split\('\|'\)"

+JSON_LD_RE = r'(?is)<script[^>]+type=(["\'])application/ld\+json\1[^>]*>(?P<json_ld>.+?)</script>'

 
 
 def preferredencoding():
 
 
 def preferredencoding():


@@ -538,10 +539,22 @@ def sanitize_path(s):
     return os.path.join(*sanitized_path)
 
 
     return os.path.join(*sanitized_path)
 
 

-# Prepend protocol-less URLs with `http:` scheme in order to mitigate the number of
-# unwanted failures due to missing protocol

 def sanitize_url(url):
 def sanitize_url(url):

-    return 'http:%s' % url if url.startswith('//') else url
+    # Prepend protocol-less URLs with `http:` scheme in order to mitigate
+    # the number of unwanted failures due to missing protocol
+    if url.startswith('//'):
+        return 'http:%s' % url
+    # Fix some common typos seen so far
+    COMMON_TYPOS = (
+        # https://github.com/rg3/youtube-dl/issues/15649
+        (r'^httpss://', r'https://'),
+        # https://bx1.be/lives/direct-tv/
+        (r'^rmtp([es]?)://', r'rtmp\1://'),
+    )
+    for mistake, fixup in COMMON_TYPOS:
+        if re.match(mistake, url):
+            return re.sub(mistake, fixup, url)
+    return url

 
 
 def sanitized_Request(url, *args, **kwargs):
 
 
 def sanitized_Request(url, *args, **kwargs):


@@ -866,16 +879,54 @@ def _create_http_connection(ydl_handler, http_class, is_https, *args, **kwargs):
     # expected HTTP responses to meet HTTP/1.0 or later (see also
     # https://github.com/rg3/youtube-dl/issues/6727)
     if sys.version_info < (3, 0):
     # expected HTTP responses to meet HTTP/1.0 or later (see also
     # https://github.com/rg3/youtube-dl/issues/6727)
     if sys.version_info < (3, 0):

-        kwargs[b'strict'] = True
-    hc = http_class(*args, **kwargs)
+        kwargs['strict'] = True
+    hc = http_class(*args, **compat_kwargs(kwargs))

     source_address = ydl_handler._params.get('source_address')
     source_address = ydl_handler._params.get('source_address')

+

     if source_address is not None:
     if source_address is not None:

+        # This is to workaround _create_connection() from socket where it will try all
+        # address data from getaddrinfo() including IPv6. This filters the result from
+        # getaddrinfo() based on the source_address value.
+        # This is based on the cpython socket.create_connection() function.
+        # https://github.com/python/cpython/blob/master/Lib/socket.py#L691
+        def _create_connection(address, timeout=socket._GLOBAL_DEFAULT_TIMEOUT, source_address=None):
+            host, port = address
+            err = None
+            addrs = socket.getaddrinfo(host, port, 0, socket.SOCK_STREAM)
+            af = socket.AF_INET if '.' in source_address[0] else socket.AF_INET6
+            ip_addrs = [addr for addr in addrs if addr[0] == af]
+            if addrs and not ip_addrs:
+                ip_version = 'v4' if af == socket.AF_INET else 'v6'
+                raise socket.error(
+                    "No remote IP%s addresses available for connect, can't use '%s' as source address"
+                    % (ip_version, source_address[0]))
+            for res in ip_addrs:
+                af, socktype, proto, canonname, sa = res
+                sock = None
+                try:
+                    sock = socket.socket(af, socktype, proto)
+                    if timeout is not socket._GLOBAL_DEFAULT_TIMEOUT:
+                        sock.settimeout(timeout)
+                    sock.bind(source_address)
+                    sock.connect(sa)
+                    err = None  # Explicitly break reference cycle
+                    return sock
+                except socket.error as _:
+                    err = _
+                    if sock is not None:
+                        sock.close()
+            if err is not None:
+                raise err
+            else:
+                raise socket.error('getaddrinfo returns an empty list')
+        if hasattr(hc, '_create_connection'):
+            hc._create_connection = _create_connection

         sa = (source_address, 0)
         if hasattr(hc, 'source_address'):  # Python 2.7+
             hc.source_address = sa
         else:  # Python 2.6
             def _hc_connect(self, *args, **kwargs):
         sa = (source_address, 0)
         if hasattr(hc, 'source_address'):  # Python 2.7+
             hc.source_address = sa
         else:  # Python 2.6
             def _hc_connect(self, *args, **kwargs):

-                sock = compat_socket_create_connection(
+                sock = _create_connection(

                     (self.host, self.port), self.timeout, sa)
                 if is_https:
                     self.sock = ssl.wrap_socket(
                     (self.host, self.port), self.timeout, sa)
                 if is_https:
                     self.sock = ssl.wrap_socket(


@@ -1089,6 +1140,33 @@ class YoutubeDLHTTPSHandler(compat_urllib_request.HTTPSHandler):
             req, **kwargs)
 
 
             req, **kwargs)
 
 

+class YoutubeDLCookieJar(compat_cookiejar.MozillaCookieJar):
+    def save(self, filename=None, ignore_discard=False, ignore_expires=False):
+        # Store session cookies with `expires` set to 0 instead of an empty
+        # string
+        for cookie in self:
+            if cookie.expires is None:
+                cookie.expires = 0
+        compat_cookiejar.MozillaCookieJar.save(self, filename, ignore_discard, ignore_expires)
+
+    def load(self, filename=None, ignore_discard=False, ignore_expires=False):
+        compat_cookiejar.MozillaCookieJar.load(self, filename, ignore_discard, ignore_expires)
+        # Session cookies are denoted by either `expires` field set to
+        # an empty string or 0. MozillaCookieJar only recognizes the former
+        # (see [1]). So we need force the latter to be recognized as session
+        # cookies on our own.
+        # Session cookies may be important for cookies-based authentication,
+        # e.g. usually, when user does not check 'Remember me' check box while
+        # logging in on a site, some important cookies are stored as session
+        # cookies so that not recognizing them will result in failed login.
+        # 1. https://bugs.python.org/issue17164
+        for cookie in self:
+            # Treat `expires=0` cookies as session cookies
+            if cookie.expires == 0:
+                cookie.expires = None
+                cookie.discard = True
+
+

 class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
     def __init__(self, cookiejar=None):
         compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar)
 class YoutubeDLCookieProcessor(compat_urllib_request.HTTPCookieProcessor):
     def __init__(self, cookiejar=None):
         compat_urllib_request.HTTPCookieProcessor.__init__(self, cookiejar)


@@ -1199,6 +1277,11 @@ def unified_timestamp(date_str, day_first=True):
     if m:
         date_str = date_str[:-len(m.group('tz'))]
 
     if m:
         date_str = date_str[:-len(m.group('tz'))]
 

+    # Python only supports microseconds, so remove nanoseconds
+    m = re.search(r'^([0-9]{4,}-[0-9]{1,2}-[0-9]{1,2}T[0-9]{1,2}:[0-9]{1,2}:[0-9]{1,2}\.[0-9]{6})[0-9]+$', date_str)
+    if m:
+        date_str = m.group(1)
+

     for expression in date_formats(day_first):
         try:
             dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta)
     for expression in date_formats(day_first):
         try:
             dt = datetime.datetime.strptime(date_str, expression) - timezone + datetime.timedelta(hours=pm_delta)


@@ -1211,7 +1294,7 @@ def unified_timestamp(date_str, day_first=True):
 
 
 def determine_ext(url, default_ext='unknown_video'):
 
 
 def determine_ext(url, default_ext='unknown_video'):

-    if url is None:
+    if url is None or '.' not in url:

         return default_ext
     guess = url.partition('?')[0].rpartition('.')[2]
     if re.match(r'^[A-Za-z0-9]+$', guess):
         return default_ext
     guess = url.partition('?')[0].rpartition('.')[2]
     if re.match(r'^[A-Za-z0-9]+$', guess):


@@ -1677,6 +1760,28 @@ def parse_count(s):
     return lookup_unit_table(_UNIT_TABLE, s)
 
 
     return lookup_unit_table(_UNIT_TABLE, s)
 
 

+def parse_resolution(s):
+    if s is None:
+        return {}
+
+    mobj = re.search(r'\b(?P<w>\d+)\s*[xX×]\s*(?P<h>\d+)\b', s)
+    if mobj:
+        return {
+            'width': int(mobj.group('w')),
+            'height': int(mobj.group('h')),
+        }
+
+    mobj = re.search(r'\b(\d+)[pPiI]\b', s)
+    if mobj:
+        return {'height': int(mobj.group(1))}
+
+    mobj = re.search(r'\b([48])[kK]\b', s)
+    if mobj:
+        return {'height': int(mobj.group(1)) * 540}
+
+    return {}
+
+

 def month_by_name(name, lang='en'):
     """ Return the number of a month by (locale-independently) English name """
 
 def month_by_name(name, lang='en'):
     """ Return the number of a month by (locale-independently) English name """
 


@@ -1826,6 +1931,13 @@ def strip_or_none(v):
     return None if v is None else v.strip()
 
 
     return None if v is None else v.strip()
 
 

+def url_or_none(url):
+    if not url or not isinstance(url, compat_str):
+        return None
+    url = url.strip()
+    return url if re.match(r'^(?:[a-zA-Z][\da-zA-Z.+-]*:)?//', url) else None
+
+

 def parse_duration(s):
     if not isinstance(s, compat_basestring):
         return None
 def parse_duration(s):
     if not isinstance(s, compat_basestring):
         return None


@@ -2186,6 +2298,20 @@ def try_get(src, getter, expected_type=None):
                 return v
 
 
                 return v
 
 

+def merge_dicts(*dicts):
+    merged = {}
+    for a_dict in dicts:
+        for k, v in a_dict.items():
+            if v is None:
+                continue
+            if (k not in merged or
+                    (isinstance(v, compat_str) and v and
+                        isinstance(merged[k], compat_str) and
+                        not merged[k])):
+                merged[k] = v
+    return merged
+
+

 def encode_compat_str(string, encoding=preferredencoding(), errors='strict'):
     return string if isinstance(string, compat_str) else compat_str(string, encoding, errors)
 
 def encode_compat_str(string, encoding=preferredencoding(), errors='strict'):
     return string if isinstance(string, compat_str) else compat_str(string, encoding, errors)
 


@@ -2219,13 +2345,16 @@ def parse_age_limit(s):
         return int(m.group('age'))
     if s in US_RATINGS:
         return US_RATINGS[s]
         return int(m.group('age'))
     if s in US_RATINGS:
         return US_RATINGS[s]

-    return TV_PARENTAL_GUIDELINES.get(s)
+    m = re.match(r'^TV[_-]?(%s)$' % '|'.join(k[3:] for k in TV_PARENTAL_GUIDELINES), s)
+    if m:
+        return TV_PARENTAL_GUIDELINES['TV-' + m.group(1)]
+    return None

 
 
 def strip_jsonp(code):
     return re.sub(
         r'''(?sx)^
 
 
 def strip_jsonp(code):
     return re.sub(
         r'''(?sx)^

-            (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]+)
+            (?:window\.)?(?P<func_name>[a-zA-Z0-9_.$]*)

             (?:\s*&&\s*(?P=func_name))?
             \s*\(\s*(?P<callback_data>.*)\);?
             \s*?(?://[^\n]*)*$''',
             (?:\s*&&\s*(?P=func_name))?
             \s*\(\s*(?P<callback_data>.*)\);?
             \s*?(?://[^\n]*)*$''',


@@ -2376,7 +2505,7 @@ def parse_codecs(codecs_str):
     vcodec, acodec = None, None
     for full_codec in splited_codecs:
         codec = full_codec.split('.')[0]
     vcodec, acodec = None, None
     for full_codec in splited_codecs:
         codec = full_codec.split('.')[0]

-        if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1'):
+        if codec in ('avc1', 'avc2', 'avc3', 'avc4', 'vp9', 'vp8', 'hev1', 'hev2', 'h263', 'h264', 'mp4v', 'hvc1', 'av01'):

             if not vcodec:
                 vcodec = full_codec
         elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):
             if not vcodec:
                 vcodec = full_codec
         elif codec in ('mp4a', 'opus', 'vorbis', 'mp3', 'aac', 'ac-3', 'ec-3', 'eac3', 'dtsc', 'dtse', 'dtsh', 'dtsl'):


@@ -2535,8 +2664,8 @@ def _match_one(filter_part, dct):
         return op(actual_value, comparison_value)
 
     UNARY_OPERATORS = {
         return op(actual_value, comparison_value)
 
     UNARY_OPERATORS = {

-        '': lambda v: v is not None,
-        '!': lambda v: v is None,
+        '': lambda v: (v is True) if isinstance(v, bool) else (v is not None),
+        '!': lambda v: (v is False) if isinstance(v, bool) else (v is None),

     }
     operator_rex = re.compile(r'''(?x)\s*
         (?P<op>%s)\s*(?P<key>[a-z_]+)
     }
     operator_rex = re.compile(r'''(?x)\s*
         (?P<op>%s)\s*(?P<key>[a-z_]+)


@@ -2611,6 +2740,7 @@ def dfxp2srt(dfxp_data):
     ]
 
     _x = functools.partial(xpath_with_ns, ns_map={
     ]
 
     _x = functools.partial(xpath_with_ns, ns_map={

+        'xml': 'http://www.w3.org/XML/1998/namespace',

         'ttml': 'http://www.w3.org/ns/ttml',
         'tts': 'http://www.w3.org/ns/ttml#styling',
     })
         'ttml': 'http://www.w3.org/ns/ttml',
         'tts': 'http://www.w3.org/ns/ttml#styling',
     })


@@ -2702,7 +2832,9 @@ def dfxp2srt(dfxp_data):
     repeat = False
     while True:
         for style in dfxp.findall(_x('.//ttml:style')):
     repeat = False
     while True:
         for style in dfxp.findall(_x('.//ttml:style')):

-            style_id = style.get('id')
+            style_id = style.get('id') or style.get(_x('xml:id'))
+            if not style_id:
+                continue

             parent_style_id = style.get('style')
             if parent_style_id:
                 if parent_style_id not in styles:
             parent_style_id = style.get('style')
             if parent_style_id:
                 if parent_style_id not in styles:


@@ -2836,6 +2968,7 @@ class ISO639Utils(object):
         'gv': 'glv',
         'ha': 'hau',
         'he': 'heb',
         'gv': 'glv',
         'ha': 'hau',
         'he': 'heb',

+        'iw': 'heb',  # Replaced by he in 1989 revision

         'hi': 'hin',
         'ho': 'hmo',
         'hr': 'hrv',
         'hi': 'hin',
         'ho': 'hmo',
         'hr': 'hrv',


@@ -2845,6 +2978,7 @@ class ISO639Utils(object):
         'hz': 'her',
         'ia': 'ina',
         'id': 'ind',
         'hz': 'her',
         'ia': 'ina',
         'id': 'ind',

+        'in': 'ind',  # Replaced by id in 1989 revision

         'ie': 'ile',
         'ig': 'ibo',
         'ii': 'iii',
         'ie': 'ile',
         'ig': 'ibo',
         'ii': 'iii',


@@ -2959,6 +3093,7 @@ class ISO639Utils(object):
         'wo': 'wol',
         'xh': 'xho',
         'yi': 'yid',
         'wo': 'wol',
         'xh': 'xho',
         'yi': 'yid',

+        'ji': 'yid',  # Replaced by yi in 1989 revision

         'yo': 'yor',
         'za': 'zha',
         'zh': 'zho',
         'yo': 'yor',
         'za': 'zha',
         'zh': 'zho',


@@ -3481,10 +3616,13 @@ class GeoUtils(object):
     }
 
     @classmethod
     }
 
     @classmethod

-    def random_ipv4(cls, code):
-        block = cls._country_ip_map.get(code.upper())
-        if not block:
-            return None
+    def random_ipv4(cls, code_or_block):
+        if len(code_or_block) == 2:
+            block = cls._country_ip_map.get(code_or_block.upper())
+            if not block:
+                return None
+        else:
+            block = code_or_block

         addr, preflen = block.split('/')
         addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
         addr_max = addr_min | (0xffffffff >> int(preflen))
         addr, preflen = block.split('/')
         addr_min = compat_struct_unpack('!L', socket.inet_aton(addr))[0]
         addr_max = addr_min | (0xffffffff >> int(preflen))


@@ -3499,7 +3637,7 @@ class PerRequestProxyHandler(compat_urllib_request.ProxyHandler):
             setattr(self, '%s_open' % type,
                     lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
                         meth(r, proxy, type))
             setattr(self, '%s_open' % type,
                     lambda r, proxy='__noproxy__', type=type, meth=self.proxy_open:
                         meth(r, proxy, type))

-        return compat_urllib_request.ProxyHandler.__init__(self, proxies)
+        compat_urllib_request.ProxyHandler.__init__(self, proxies)

 
     def proxy_open(self, req, proxy, type):
         req_proxy = req.headers.get('Ytdl-request-proxy')
 
     def proxy_open(self, req, proxy, type):
         req_proxy = req.headers.get('Ytdl-request-proxy')


@@ -3841,8 +3979,12 @@ def write_xattr(path, key, value):
 
 
 def random_birthday(year_field, month_field, day_field):
 
 
 def random_birthday(year_field, month_field, day_field):

+    start_date = datetime.date(1950, 1, 1)
+    end_date = datetime.date(1995, 12, 31)
+    offset = random.randint(0, (end_date - start_date).days)
+    random_date = start_date + datetime.timedelta(offset)

     return {
     return {

-        year_field: str(random.randint(1950, 1995)),
-        month_field: str(random.randint(1, 12)),
-        day_field: str(random.randint(1, 31)),
+        year_field: str(random_date.year),
+        month_field: str(random_date.month),
+        day_field: str(random_date.day),

     }
     }