Merge b83889a20eb01094a3361cf08f27895fea164049 into da7223d4aa42ff9fc680b0951d043dd03cec2d30

* improve extraction of global "useful data" Array from player JS
* also handle tv-player and add tests: thx seproDev (yt-dlp/yt-dlp#12684)

Co-Authored-By: sepro <sepro@sepr0.com>

test/test_jsinterp.py

@@ -11,6 +11,7 @@ sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
 
 import math
 import re
+import time
 
 from youtube_dl.compat import compat_str as str
 from youtube_dl.jsinterp import JS_Undefined, JSInterpreter
@@ -208,6 +209,34 @@ class TestJSInterpreter(unittest.TestCase):
         self._test(jsi, 86000, args=['12/31/1969 18:01:26 MDT'])
         # epoch 0
         self._test(jsi, 0, args=['1 January 1970 00:00:00 UTC'])
+        # undefined
+        self._test(jsi, NaN, args=[JS_Undefined])
+        # y,m,d, ... - may fail with older dates lacking DST data
+        jsi = JSInterpreter(
+            'function f() { return new Date(%s); }'
+            % ('2024, 5, 29, 2, 52, 12, 42',))
+        self._test(jsi, (
+            1719625932042                           # UK value
+            + (
+                + 3600                              # back to GMT
+                + (time.altzone if time.daylight    # host's DST
+                   else time.timezone)
+            ) * 1000))
+        # no arg
+        self.assertAlmostEqual(JSInterpreter(
+            'function f() { return new Date() - 0; }').call_function('f'),
+            time.time() * 1000, delta=100)
+        # Date.now()
+        self.assertAlmostEqual(JSInterpreter(
+            'function f() { return Date.now(); }').call_function('f'),
+            time.time() * 1000, delta=100)
+        # Date.parse()
+        jsi = JSInterpreter('function f(dt) { return Date.parse(dt); }')
+        self._test(jsi, 0, args=['1 January 1970 00:00:00 UTC'])
+        # Date.UTC()
+        jsi = JSInterpreter('function f() { return Date.UTC(%s); }'
+                            % ('1970, 0, 1, 0, 0, 0, 0',))
+        self._test(jsi, 0)
 
     def test_call(self):
         jsi = JSInterpreter('''
@@ -463,6 +492,14 @@ class TestJSInterpreter(unittest.TestCase):
         self._test('function f(){return NaN << 42}', 0)
         self._test('function f(){return "21.9" << 1}', 42)
         self._test('function f(){return 21 << 4294967297}', 42)
+        self._test('function f(){return true << "5";}', 32)
+        self._test('function f(){return true << true;}', 2)
+        self._test('function f(){return "19" & "21.9";}', 17)
+        self._test('function f(){return "19" & false;}', 0)
+        self._test('function f(){return "11.0" >> "2.1";}', 2)
+        self._test('function f(){return 5 ^ 9;}', 12)
+        self._test('function f(){return 0.0 << NaN}', 0)
+        self._test('function f(){return null << undefined}', 0)
 
     def test_negative(self):
         self._test('function f(){return 2    *    -2.0    ;}', -4)

test/test_youtube_signature.py

@@ -223,6 +223,42 @@ _NSIG_TESTS = [
         'https://www.youtube.com/s/player/9c6dfc4a/player_ias.vflset/en_US/base.js',
         'jbu7ylIosQHyJyJV', 'uwI0ESiynAmhNg',
     ),
+    (
+        'https://www.youtube.com/s/player/f6e09c70/player_ias.vflset/en_US/base.js',
+        'W9HJZKktxuYoDTqW', 'jHbbkcaxm54',
+    ),
+    (
+        'https://www.youtube.com/s/player/f6e09c70/player_ias_tce.vflset/en_US/base.js',
+        'W9HJZKktxuYoDTqW', 'jHbbkcaxm54',
+    ),
+    (
+        'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
+        'W9HJZKktxuYoDTqW', 'larxUlagTRAcSw',
+    ),
+    (
+        'https://www.youtube.com/s/player/e7567ecf/player_ias_tce.vflset/en_US/base.js',
+        'Sy4aDGc0VpYRR9ew_', '5UPOT1VhoZxNLQ',
+    ),
+    (
+        'https://www.youtube.com/s/player/d50f54ef/player_ias_tce.vflset/en_US/base.js',
+        'Ha7507LzRmH3Utygtj', 'XFTb2HoeOE5MHg',
+    ),
+    (
+        'https://www.youtube.com/s/player/074a8365/player_ias_tce.vflset/en_US/base.js',
+        'Ha7507LzRmH3Utygtj', 'ufTsrE0IVYrkl8v',
+    ),
+    (
+        'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
+        'N5uAlLqm0eg1GyHO', 'dCBQOejdq5s-ww',
+    ),
+    (
+        'https://www.youtube.com/s/player/69f581a5/tv-player-ias.vflset/tv-player-ias.js',
+        '-qIP447rVlTTwaZjY', 'KNcGOksBAvwqQg',
+    ),
+    (
+        'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js',
+        'ir9-V6cdbCiyKxhr', '2PL7ZDYAALMfmA',
+    ),
 ]
 
 
@@ -284,7 +320,7 @@ def t_factory(name, sig_func, url_pattern):
 
 
 def signature(jscode, sig_input):
-    func = YoutubeIE(FakeYDL())._parse_sig_js(jscode)
+    func = YoutubeIE(FakeYDL({'cachedir': False}))._parse_sig_js(jscode)
     src_sig = (
         compat_str(string.printable[:sig_input])
         if isinstance(sig_input, int) else sig_input)
@@ -292,9 +328,10 @@ def signature(jscode, sig_input):
 
 
 def n_sig(jscode, sig_input):
-    funcname = YoutubeIE(FakeYDL())._extract_n_function_name(jscode)
+    ie = YoutubeIE(FakeYDL({'cachedir': False}))
-    return JSInterpreter(jscode).call_function(
+    jsi = JSInterpreter(jscode)
-        funcname, sig_input, _ytdl_do_not_return=sig_input)
+    jsi, _, func_code = ie._extract_n_function_code_jsi(sig_input, jsi)
+    return ie._extract_n_function_from_code(jsi, func_code)(sig_input)
 
 
 make_sig_test = t_factory(

youtube_dl/__init__.py

@@ -18,7 +18,7 @@ from .compat import (
     compat_getpass,
     compat_register_utf8,
     compat_shlex_split,
-    workaround_optparse_bug9161,
+    _workaround_optparse_bug9161,
 )
 from .utils import (
     _UnsafeExtensionError,
@@ -50,7 +50,7 @@ def _real_main(argv=None):
     # Compatibility fix for Windows
     compat_register_utf8()
 
-    workaround_optparse_bug9161()
+    _workaround_optparse_bug9161()
 
     setproctitle('youtube-dl')
 

youtube_dl/compat.py

@@ -16,7 +16,6 @@ import os
 import platform
 import re
 import shlex
-import shutil
 import socket
 import struct
 import subprocess
@@ -24,11 +23,15 @@ import sys
 import types
 import xml.etree.ElementTree
 
+_IDENTITY = lambda x: x
+
 # naming convention
 # 'compat_' + Python3_name.replace('.', '_')
 # other aliases exist for convenience and/or legacy
+# wrap disposable test values in type() to reclaim storage
 
-# deal with critical unicode/str things first
+# deal with critical unicode/str things first:
+# compat_str, compat_basestring, compat_chr
 try:
     # Python 2
     compat_str, compat_basestring, compat_chr = (
@@ -39,18 +42,23 @@ except NameError:
         str, (str, bytes), chr
     )
 
-# casefold
+
+# compat_casefold
 try:
     compat_str.casefold
     compat_casefold = lambda s: s.casefold()
 except AttributeError:
     from .casefold import _casefold as compat_casefold
 
+
+# compat_collections_abc
 try:
     import collections.abc as compat_collections_abc
 except ImportError:
     import collections as compat_collections_abc
 
+
+# compat_urllib_request
 try:
     import urllib.request as compat_urllib_request
 except ImportError:  # Python 2
@@ -79,11 +87,15 @@ except TypeError:
     _add_init_method_arg(compat_urllib_request.Request)
     del _add_init_method_arg
 
+
+# compat_urllib_error
 try:
     import urllib.error as compat_urllib_error
 except ImportError:  # Python 2
     import urllib2 as compat_urllib_error
 
+
+# compat_urllib_parse
 try:
     import urllib.parse as compat_urllib_parse
 except ImportError:  # Python 2
@@ -98,17 +110,23 @@ except ImportError:  # Python 2
 compat_urlparse = compat_urllib_parse
 compat_urllib_parse_urlparse = compat_urllib_parse.urlparse
 
+
+# compat_urllib_response
 try:
     import urllib.response as compat_urllib_response
 except ImportError:  # Python 2
     import urllib as compat_urllib_response
 
+
+# compat_urllib_response.addinfourl
 try:
     compat_urllib_response.addinfourl.status
 except AttributeError:
     # .getcode() is deprecated in Py 3.
     compat_urllib_response.addinfourl.status = property(lambda self: self.getcode())
 
+
+# compat_http_cookiejar
 try:
     import http.cookiejar as compat_cookiejar
 except ImportError:  # Python 2
@@ -127,12 +145,16 @@ else:
     compat_cookiejar_Cookie = compat_cookiejar.Cookie
 compat_http_cookiejar_Cookie = compat_cookiejar_Cookie
 
+
+# compat_http_cookies
 try:
     import http.cookies as compat_cookies
 except ImportError:  # Python 2
     import Cookie as compat_cookies
 compat_http_cookies = compat_cookies
 
+
+# compat_http_cookies_SimpleCookie
 if sys.version_info[0] == 2 or sys.version_info < (3, 3):
     class compat_cookies_SimpleCookie(compat_cookies.SimpleCookie):
         def load(self, rawdata):
@@ -155,11 +177,15 @@ else:
     compat_cookies_SimpleCookie = compat_cookies.SimpleCookie
 compat_http_cookies_SimpleCookie = compat_cookies_SimpleCookie
 
+
+# compat_html_entities, probably useless now
 try:
     import html.entities as compat_html_entities
 except ImportError:  # Python 2
     import htmlentitydefs as compat_html_entities
 
+
+# compat_html_entities_html5
 try:  # Python >= 3.3
     compat_html_entities_html5 = compat_html_entities.html5
 except AttributeError:
@@ -2408,18 +2434,24 @@ except AttributeError:
     # Py < 3.1
     compat_http_client.HTTPResponse.getcode = lambda self: self.status
 
+
+# compat_urllib_HTTPError
 try:
     from urllib.error import HTTPError as compat_HTTPError
 except ImportError:  # Python 2
     from urllib2 import HTTPError as compat_HTTPError
 compat_urllib_HTTPError = compat_HTTPError
 
+
+# compat_urllib_request_urlretrieve
 try:
     from urllib.request import urlretrieve as compat_urlretrieve
 except ImportError:  # Python 2
     from urllib import urlretrieve as compat_urlretrieve
 compat_urllib_request_urlretrieve = compat_urlretrieve
 
+
+# compat_html_parser_HTMLParser, compat_html_parser_HTMLParseError
 try:
     from HTMLParser import (
         HTMLParser as compat_HTMLParser,
@@ -2432,22 +2464,33 @@ except ImportError:  # Python 3
         # HTMLParseError was deprecated in Python 3.3 and removed in
         # Python 3.5. Introducing dummy exception for Python >3.5 for compatible
         # and uniform cross-version exception handling
+
         class compat_HTMLParseError(Exception):
             pass
+
 compat_html_parser_HTMLParser = compat_HTMLParser
 compat_html_parser_HTMLParseError = compat_HTMLParseError
 
+
+# compat_subprocess_get_DEVNULL
 try:
     _DEVNULL = subprocess.DEVNULL
     compat_subprocess_get_DEVNULL = lambda: _DEVNULL
 except AttributeError:
     compat_subprocess_get_DEVNULL = lambda: open(os.path.devnull, 'w')
 
+
+# compat_http_server
 try:
     import http.server as compat_http_server
 except ImportError:
     import BaseHTTPServer as compat_http_server
 
+
+# compat_urllib_parse_unquote_to_bytes,
+# compat_urllib_parse_unquote, compat_urllib_parse_unquote_plus,
+# compat_urllib_parse_urlencode,
+# compat_urllib_parse_parse_qs
 try:
     from urllib.parse import unquote_to_bytes as compat_urllib_parse_unquote_to_bytes
     from urllib.parse import unquote as compat_urllib_parse_unquote
@@ -2598,6 +2641,8 @@ except ImportError:  # Python 2
 
 compat_urllib_parse_parse_qs = compat_parse_qs
 
+
+# compat_urllib_request_DataHandler
 try:
     from urllib.request import DataHandler as compat_urllib_request_DataHandler
 except ImportError:  # Python < 3.4
@@ -2632,16 +2677,20 @@ except ImportError:  # Python < 3.4
 
             return compat_urllib_response.addinfourl(io.BytesIO(data), headers, url)
 
+
+# compat_xml_etree_ElementTree_ParseError
 try:
     from xml.etree.ElementTree import ParseError as compat_xml_parse_error
 except ImportError:  # Python 2.6
     from xml.parsers.expat import ExpatError as compat_xml_parse_error
 compat_xml_etree_ElementTree_ParseError = compat_xml_parse_error
 
-etree = xml.etree.ElementTree
+
+# compat_xml_etree_ElementTree_Element
+_etree = xml.etree.ElementTree
 
 
-class _TreeBuilder(etree.TreeBuilder):
+class _TreeBuilder(_etree.TreeBuilder):
     def doctype(self, name, pubid, system):
         pass
 
@@ -2650,7 +2699,7 @@ try:
     # xml.etree.ElementTree.Element is a method in Python <=2.6 and
     # the following will crash with:
     #  TypeError: isinstance() arg 2 must be a class, type, or tuple of classes and types
-    isinstance(None, etree.Element)
+    isinstance(None, _etree.Element)
     from xml.etree.ElementTree import Element as compat_etree_Element
 except TypeError:  # Python <=2.6
     from xml.etree.ElementTree import _ElementInterface as compat_etree_Element
@@ -2658,12 +2707,12 @@ compat_xml_etree_ElementTree_Element = compat_etree_Element
 
 if sys.version_info[0] >= 3:
     def compat_etree_fromstring(text):
-        return etree.XML(text, parser=etree.XMLParser(target=_TreeBuilder()))
+        return _etree.XML(text, parser=_etree.XMLParser(target=_TreeBuilder()))
 else:
     # python 2.x tries to encode unicode strings with ascii (see the
     # XMLParser._fixtext method)
     try:
-        _etree_iter = etree.Element.iter
+        _etree_iter = _etree.Element.iter
     except AttributeError:  # Python <=2.6
         def _etree_iter(root):
             for el in root.findall('*'):
@@ -2675,27 +2724,29 @@ else:
     # 2.7 source
     def _XML(text, parser=None):
         if not parser:
-            parser = etree.XMLParser(target=_TreeBuilder())
+            parser = _etree.XMLParser(target=_TreeBuilder())
         parser.feed(text)
         return parser.close()
 
     def _element_factory(*args, **kwargs):
-        el = etree.Element(*args, **kwargs)
+        el = _etree.Element(*args, **kwargs)
         for k, v in el.items():
             if isinstance(v, bytes):
                 el.set(k, v.decode('utf-8'))
         return el
 
     def compat_etree_fromstring(text):
-        doc = _XML(text, parser=etree.XMLParser(target=_TreeBuilder(element_factory=_element_factory)))
+        doc = _XML(text, parser=_etree.XMLParser(target=_TreeBuilder(element_factory=_element_factory)))
         for el in _etree_iter(doc):
             if el.text is not None and isinstance(el.text, bytes):
                 el.text = el.text.decode('utf-8')
         return doc
 
-if hasattr(etree, 'register_namespace'):
+
-    compat_etree_register_namespace = etree.register_namespace
+# compat_xml_etree_register_namespace
-else:
+try:
+    compat_etree_register_namespace = _etree.register_namespace
+except AttributeError:
     def compat_etree_register_namespace(prefix, uri):
         """Register a namespace prefix.
         The registry is global, and any existing mapping for either the
@@ -2704,14 +2755,16 @@ else:
         attributes in this namespace will be serialized with prefix if possible.
         ValueError is raised if prefix is reserved or is invalid.
         """
-        if re.match(r"ns\d+$", prefix):
+        if re.match(r'ns\d+$', prefix):
-            raise ValueError("Prefix format reserved for internal use")
+            raise ValueError('Prefix format reserved for internal use')
-        for k, v in list(etree._namespace_map.items()):
+        for k, v in list(_etree._namespace_map.items()):
             if k == uri or v == prefix:
-                del etree._namespace_map[k]
+                del _etree._namespace_map[k]
-        etree._namespace_map[uri] = prefix
+        _etree._namespace_map[uri] = prefix
 compat_xml_etree_register_namespace = compat_etree_register_namespace
 
+
+# compat_xpath, compat_etree_iterfind
 if sys.version_info < (2, 7):
     # Here comes the crazy part: In 2.6, if the xpath is a unicode,
     # .//node does not match if a node is a direct child of . !
@@ -2898,7 +2951,6 @@ if sys.version_info < (2, 7):
         def __init__(self, root):
             self.root = root
 
-    ##
     # Generate all matching objects.
 
     def compat_etree_iterfind(elem, path, namespaces=None):
@@ -2933,13 +2985,15 @@ if sys.version_info < (2, 7):
 
 
 else:
-    compat_xpath = lambda xpath: xpath
     compat_etree_iterfind = lambda element, match: element.iterfind(match)
+    compat_xpath = _IDENTITY
 
 
+# compat_os_name
 compat_os_name = os._name if os.name == 'java' else os.name
 
 
+# compat_shlex_quote
 if compat_os_name == 'nt':
     def compat_shlex_quote(s):
         return s if re.match(r'^[-_\w./]+$', s) else '"%s"' % s.replace('"', '\\"')
@@ -2954,6 +3008,7 @@ else:
                 return "'" + s.replace("'", "'\"'\"'") + "'"
 
 
+# compat_shlex.split
 try:
     args = shlex.split('中文')
     assert (isinstance(args, list)
@@ -2969,6 +3024,7 @@ except (AssertionError, UnicodeEncodeError):
         return list(map(lambda s: s.decode('utf-8'), shlex.split(s, comments, posix)))
 
 
+# compat_ord
 def compat_ord(c):
     if isinstance(c, int):
         return c
@@ -2976,6 +3032,7 @@ def compat_ord(c):
         return ord(c)
 
 
+# compat_getenv, compat_os_path_expanduser, compat_setenv
 if sys.version_info >= (3, 0):
     compat_getenv = os.getenv
     compat_expanduser = os.path.expanduser
@@ -3063,6 +3120,7 @@ else:
 compat_os_path_expanduser = compat_expanduser
 
 
+# compat_os_path_realpath
 if compat_os_name == 'nt' and sys.version_info < (3, 8):
     # os.path.realpath on Windows does not follow symbolic links
     # prior to Python 3.8 (see https://bugs.python.org/issue9949)
@@ -3076,6 +3134,7 @@ else:
 compat_os_path_realpath = compat_realpath
 
 
+# compat_print
 if sys.version_info < (3, 0):
     def compat_print(s):
         from .utils import preferredencoding
@@ -3086,6 +3145,7 @@ else:
         print(s)
 
 
+# compat_getpass_getpass
 if sys.version_info < (3, 0) and sys.platform == 'win32':
     def compat_getpass(prompt, *args, **kwargs):
         if isinstance(prompt, compat_str):
@@ -3098,22 +3158,22 @@ else:
 compat_getpass_getpass = compat_getpass
 
 
+# compat_input
 try:
     compat_input = raw_input
 except NameError:  # Python 3
     compat_input = input
 
 
+# compat_kwargs
 # Python < 2.6.5 require kwargs to be bytes
 try:
-    def _testfunc(x):
+    (lambda x: x)(**{'x': 0})
-        pass
-    _testfunc(**{'x': 0})
 except TypeError:
     def compat_kwargs(kwargs):
         return dict((bytes(k), v) for k, v in kwargs.items())
 else:
-    compat_kwargs = lambda kwargs: kwargs
+    compat_kwargs = _IDENTITY
 
 
 # compat_numeric_types
@@ -3132,6 +3192,8 @@ except NameError:  # Python 3
 # compat_int
 compat_int = compat_integer_types[-1]
 
+
+# compat_socket_create_connection
 if sys.version_info < (2, 7):
     def compat_socket_create_connection(address, timeout, source_address=None):
         host, port = address
@@ -3158,6 +3220,7 @@ else:
     compat_socket_create_connection = socket.create_connection
 
 
+# compat_contextlib_suppress
 try:
     from contextlib import suppress as compat_contextlib_suppress
 except ImportError:
@@ -3200,12 +3263,12 @@ except AttributeError:
                         # repeated .close() is OK, but just in case
                         with compat_contextlib_suppress(EnvironmentError):
                             f.close()
-                popen.wait()
+            popen.wait()
 
 
 # Fix https://github.com/ytdl-org/youtube-dl/issues/4223
 # See http://bugs.python.org/issue9161 for what is broken
-def workaround_optparse_bug9161():
+def _workaround_optparse_bug9161():
     op = optparse.OptionParser()
     og = optparse.OptionGroup(op, 'foo')
     try:
@@ -3224,9 +3287,10 @@ def workaround_optparse_bug9161():
         optparse.OptionGroup.add_option = _compat_add_option
 
 
-if hasattr(shutil, 'get_terminal_size'):  # Python >= 3.3
+# compat_shutil_get_terminal_size
-    compat_get_terminal_size = shutil.get_terminal_size
+try:
-else:
+    from shutil import get_terminal_size as compat_get_terminal_size  # Python >= 3.3
+except ImportError:
     _terminal_size = collections.namedtuple('terminal_size', ['columns', 'lines'])
 
     def compat_get_terminal_size(fallback=(80, 24)):
@@ -3256,27 +3320,33 @@ else:
                 columns = _columns
             if lines is None or lines <= 0:
                 lines = _lines
+
         return _terminal_size(columns, lines)
 
+compat_shutil_get_terminal_size = compat_get_terminal_size
 
+
+# compat_itertools_count
 try:
-    itertools.count(start=0, step=1)
+    type(itertools.count(start=0, step=1))
     compat_itertools_count = itertools.count
-except TypeError:  # Python 2.6
+except TypeError:  # Python 2.6 lacks step
     def compat_itertools_count(start=0, step=1):
         while True:
             yield start
             start += step
 
 
+# compat_tokenize_tokenize
 if sys.version_info >= (3, 0):
     from tokenize import tokenize as compat_tokenize_tokenize
 else:
     from tokenize import generate_tokens as compat_tokenize_tokenize
 
 
+# compat_struct_pack, compat_struct_unpack, compat_Struct
 try:
-    struct.pack('!I', 0)
+    type(struct.pack('!I', 0))
 except TypeError:
     # In Python 2.6 and 2.7.x < 2.7.7, struct requires a bytes argument
     # See https://bugs.python.org/issue19099
@@ -3308,8 +3378,10 @@ else:
         compat_Struct = struct.Struct
 
 
-# compat_map/filter() returning an iterator, supposedly the
+# builtins returning an iterator
-# same versioning as for zip below
+
+# compat_map, compat_filter
+# supposedly the same versioning as for zip below
 try:
     from future_builtins import map as compat_map
 except ImportError:
@@ -3326,6 +3398,7 @@ except ImportError:
     except ImportError:
         compat_filter = filter
 
+# compat_zip
 try:
     from future_builtins import zip as compat_zip
 except ImportError:  # not 2.6+ or is 3.x
@@ -3335,6 +3408,7 @@ except ImportError:  # not 2.6+ or is 3.x
         compat_zip = zip
 
 
+# compat_itertools_zip_longest
 # method renamed between Py2/3
 try:
     from itertools import zip_longest as compat_itertools_zip_longest
@@ -3342,7 +3416,8 @@ except ImportError:
     from itertools import izip_longest as compat_itertools_zip_longest
 
 
-# new class in collections
+# compat_collections_chain_map
+# collections.ChainMap: new class
 try:
     from collections import ChainMap as compat_collections_chain_map
     # Py3.3's ChainMap is deficient
@@ -3398,19 +3473,22 @@ except ImportError:
         def new_child(self, m=None, **kwargs):
             m = m or {}
             m.update(kwargs)
-            return compat_collections_chain_map(m, *self.maps)
+            # support inheritance !
+            return type(self)(m, *self.maps)
 
         @property
         def parents(self):
-            return compat_collections_chain_map(*(self.maps[1:]))
+            return type(self)(*(self.maps[1:]))
 
 
+# compat_re_Pattern, compat_re_Match
 # Pythons disagree on the type of a pattern (RegexObject, _sre.SRE_Pattern, Pattern, ...?)
 compat_re_Pattern = type(re.compile(''))
 # and on the type of a match
 compat_re_Match = type(re.match('a', 'a'))
 
 
+# compat_base64_b64decode
 if sys.version_info < (3, 3):
     def compat_b64decode(s, *args, **kwargs):
         if isinstance(s, compat_str):
@@ -3422,6 +3500,7 @@ else:
 compat_base64_b64decode = compat_b64decode
 
 
+# compat_ctypes_WINFUNCTYPE
 if platform.python_implementation() == 'PyPy' and sys.pypy_version_info < (5, 4, 0):
     # PyPy2 prior to version 5.4.0 expects byte strings as Windows function
     # names, see the original PyPy issue [1] and the youtube-dl one [2].
@@ -3440,6 +3519,7 @@ else:
         return ctypes.WINFUNCTYPE(*args, **kwargs)
 
 
+# compat_open
 if sys.version_info < (3, 0):
     # open(file, mode='r', buffering=- 1, encoding=None, errors=None, newline=None, closefd=True) not: opener=None
     def compat_open(file_, *args, **kwargs):
@@ -3467,18 +3547,28 @@ except AttributeError:
     def compat_datetime_timedelta_total_seconds(td):
         return (td.microseconds + (td.seconds + td.days * 24 * 3600) * 10**6) / 10**6
 
+
 # optional decompression packages
+# compat_brotli
 # PyPi brotli package implements 'br' Content-Encoding
 try:
     import brotli as compat_brotli
 except ImportError:
     compat_brotli = None
+# compat_ncompress
 # PyPi ncompress package implements 'compress' Content-Encoding
 try:
     import ncompress as compat_ncompress
 except ImportError:
     compat_ncompress = None
 
+# compat_zstandard
+# PyPi zstandard package implements 'zstd' Content-Encoding (RFC 8878 7.2)
+try:
+    import zstandard as compat_zstandard
+except ImportError:
+    compat_zstandard = None
+
 
 legacy = [
     'compat_HTMLParseError',
@@ -3495,6 +3585,7 @@ legacy = [
     'compat_getpass',
     'compat_parse_qs',
     'compat_realpath',
+    'compat_shlex_split',
     'compat_urllib_parse_parse_qs',
     'compat_urllib_parse_unquote',
     'compat_urllib_parse_unquote_plus',
@@ -3508,8 +3599,6 @@ legacy = [
 
 
 __all__ = [
-    'compat_html_parser_HTMLParseError',
-    'compat_html_parser_HTMLParser',
     'compat_Struct',
     'compat_base64_b64decode',
     'compat_basestring',
@@ -3518,13 +3607,9 @@ __all__ = [
     'compat_chr',
     'compat_collections_abc',
     'compat_collections_chain_map',
-    'compat_datetime_timedelta_total_seconds',
-    'compat_http_cookiejar',
-    'compat_http_cookiejar_Cookie',
-    'compat_http_cookies',
-    'compat_http_cookies_SimpleCookie',
     'compat_contextlib_suppress',
     'compat_ctypes_WINFUNCTYPE',
+    'compat_datetime_timedelta_total_seconds',
     'compat_etree_fromstring',
     'compat_etree_iterfind',
     'compat_filter',
@@ -3533,6 +3618,12 @@ __all__ = [
     'compat_getpass_getpass',
     'compat_html_entities',
     'compat_html_entities_html5',
+    'compat_html_parser_HTMLParseError',
+    'compat_html_parser_HTMLParser',
+    'compat_http_cookiejar',
+    'compat_http_cookiejar_Cookie',
+    'compat_http_cookies',
+    'compat_http_cookies_SimpleCookie',
     'compat_http_client',
     'compat_http_server',
     'compat_input',
@@ -3555,7 +3646,7 @@ __all__ = [
     'compat_register_utf8',
     'compat_setenv',
     'compat_shlex_quote',
-    'compat_shlex_split',
+    'compat_shutil_get_terminal_size',
     'compat_socket_create_connection',
     'compat_str',
     'compat_struct_pack',
@@ -3575,5 +3666,5 @@ __all__ = [
     'compat_xml_etree_register_namespace',
     'compat_xpath',
     'compat_zip',
-    'workaround_optparse_bug9161',
+    'compat_zstandard',
 ]

youtube_dl/extractor/common.py

@@ -3323,7 +3323,11 @@ class InfoExtractor(object):
         if (self._downloader.params.get('mark_watched', False)
                 and (self._get_login_info()[0] is not None
                      or self._downloader.params.get('cookiefile') is not None)):
-            self._mark_watched(*args, **kwargs)
+            # extractors apart from YoutubeIE can mark: be more lenient
+            try:
+                self._mark_watched(*args, **kwargs)
+            except NotImplementedError:
+                self.report_warning('Marking as watched is not supported')
 
     def _mark_watched(self, *args, **kwargs):
         raise NotImplementedError('This method must be implemented by subclasses')

youtube_dl/extractor/extractors.py

@@ -917,6 +917,11 @@ from .palcomp3 import (
     PalcoMP3VideoIE,
 )
 from .pandoratv import PandoraTVIE
+from .panopto import (
+    PanoptoIE,
+    PanoptoListIE,
+    PanoptoPlaylistIE,
+ )
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE

youtube_dl/extractor/generic.py

@@ -132,6 +132,7 @@ from .kinja import KinjaEmbedIE
 from .arcpublishing import ArcPublishingIE
 from .medialaan import MedialaanIE
 from .simplecast import SimplecastIE
+from .panopto import PanoptoIE
 
 
 class GenericIE(InfoExtractor):
@@ -2340,6 +2341,15 @@ class GenericIE(InfoExtractor):
             },
             'expected_warnings': ['uploader id'],
         },
+        {
+            # Panopto embeds
+            'url': 'https://www.monash.edu/learning-teaching/teachhq/learning-technologies/panopto/how-to/insert-a-quiz-into-a-panopto-video',
+            'info_dict': {
+                'title': 'Insert a quiz into a Panopto video',
+                'id': 'insert-a-quiz-into-a-panopto-video'
+            },
+            'playlist_count': 1
+        },
     ]
 
     def report_following_redirect(self, new_url):
@@ -3518,6 +3528,9 @@ class GenericIE(InfoExtractor):
             return self.playlist_from_matches(
                 zype_urls, video_id, video_title, ie=ZypeIE.ie_key())
 
+        panopto_entries = PanoptoIE._extract_from_webpage(url, webpage)
+        if panopto_entries:
+            return self.playlist_result(panopto_entries, video_id, video_title)
         # Look for HTML5 media
         entries = self._parse_html5_media_entries(url, webpage, video_id, m3u8_id='hls')
         if entries:

youtube_dl/extractor/panopto.py

@@ -0,0 +1,663 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import calendar
+from datetime import datetime
+import functools
+import json
+import itertools
+from random import random
+import re
+
+from .common import InfoExtractor
+from ..compat import (
+    compat_map as map,
+    compat_parse_qs as parse_qs,
+    compat_str,
+    compat_urllib_parse_urlparse,
+)
+
+from ..utils import (
+    bug_reports_message,
+    ExtractorError,
+    get_first,
+    int_or_none,
+    LazyList,
+    merge_dicts,
+    OnDemandPagedList,
+    orderedSet,
+    srt_subtitles_timecode,
+    traverse_obj,
+    try_get,
+    update_url_query,
+)
+
+import inspect
+if len(try_get(InfoExtractor.report_warning,
+               (lambda x: inspect.getfullargspec(x).FullArgs,
+                lambda x: inspect.getargspec(x).Args, ), list) or []) <= 2:
+
+    BaseInfoExtractor = InfoExtractor
+
+    class InfoExtractor(BaseInfoExtractor):
+
+        def report_warning(self, warning, only_once=True, _memo=set()):
+            from hashlib import md5
+            if only_once:
+                w_hash = md5(warning).hexdigest()
+                if w_hash in _memo:
+                    return
+                _memo.add(w_hash)
+            super(InfoExtractor, self).report_warning(self, warning)
+
+        @classmethod
+        def _match_valid_url(cls, url):
+            return re.match(cls._VALID_URL, url)
+
+        @staticmethod
+        def _merge_subtitle_items(subtitle_list1, subtitle_list2):
+            """ Merge subtitle items for one language. Items with duplicated URLs/data
+            will be dropped. """
+            list1_data = {(item.get('url'), item.get('data')) for item in subtitle_list1}
+            ret = list(subtitle_list1)
+            ret.extend(item for item in subtitle_list2 if (item.get('url'), item.get('data')) not in list1_data)
+            return ret
+
+
+class PanoptoBaseIE(InfoExtractor):
+    BASE_URL_RE = r'(?P<base_url>https?://[\w.-]+\.panopto.(?:com|eu)/Panopto)'
+
+    # see panopto core.js
+    _SUB_LANG_MAPPING = {
+        0: 'en-US',
+        1: 'en-GB',
+        2: 'es-MX',
+        3: 'es-ES',
+        4: 'de-DE',
+        5: 'fr-FR',
+        6: 'nl-NL',
+        7: 'th-TH',
+        8: 'zh-CN',
+        9: 'zh-TW',
+        10: 'ko-KR',
+        11: 'ja-JP',
+        12: 'ru-RU',
+        13: 'pt-PT',
+        14: 'pl-PL',
+        15: 'en-AU',
+        16: 'da-DK',
+        17: 'fi-FI',
+        18: 'hu-HU',
+        19: 'nb-NO',
+        20: 'sv-SE',
+        21: 'it-IT'
+    }
+
+    def _call_api(self, base_url, path, video_id, data=None, fatal=True, **kwargs):
+        response = self._download_json(
+            base_url + path, video_id, data=json.dumps(data).encode('utf8') if data else None,
+            fatal=fatal, headers={'accept': 'application/json', 'content-type': 'application/json'}, **kwargs)
+        if not response:
+            return
+        error_code = traverse_obj(response, 'ErrorCode')
+        if error_code == 2:
+            self.raise_login_required(method='cookies')
+        elif error_code is not None:
+            msg = '%s said: %s' % (self.IE_NAME, response.get('ErrorMessage') or '[no message]')
+            if fatal:
+                raise ExtractorError(msg, video_id=video_id, expected=True)
+            else:
+                self.report_warning(msg, video_id=video_id)
+        return response
+
+    @staticmethod
+    def _parse_fragment(url):
+        return dict((k, json.loads(v[0])) for k, v in parse_qs(compat_urllib_parse_urlparse(url).fragment).items())
+
+
+class PanoptoIE(PanoptoBaseIE):
+    _VALID_URL = PanoptoBaseIE.BASE_URL_RE + r'/Pages/(Viewer|Embed)\.aspx.*(?:\?|&)id=(?P<id>[a-f0-9-]+)'
+    _EMBED_REGEX = [
+        r'''<iframe\b[^>]+\bsrc\s*=\s*(["'])(?P<url>%s/Pages/(?:Viewer|Embed|Sessions/List)\.aspx(?:(?!\1)[\w\W])+)'''
+        % (PanoptoBaseIE.BASE_URL_RE, )]
+    _TESTS = [
+        {
+            'url': 'https://demo.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=26b3ae9e-4a48-4dcc-96ba-0befba08a0fb',
+            'info_dict': {
+                'id': '26b3ae9e-4a48-4dcc-96ba-0befba08a0fb',
+                'title': 'Panopto for Business - Use Cases',
+                'timestamp': 1459184200,
+                'thumbnail': r're:https://demo\.hosted\.panopto\.com/.+',
+                'upload_date': '20160328',
+                'ext': 'mp4',
+                'cast': [],
+                'chapters': [],
+                'duration': 88.17099999999999,
+                'average_rating': int,
+                'uploader_id': '2db6b718-47a0-4b0b-9e17-ab0b00f42b1e',
+                'channel_id': 'e4c6a2fc-1214-4ca0-8fb7-aef2e29ff63a',
+                'channel': 'Showcase Videos'
+            },
+        },
+        {
+            'url': 'https://demo.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=ed01b077-c9e5-4c7b-b8ff-15fa306d7a59',
+            'info_dict': {
+                'id': 'ed01b077-c9e5-4c7b-b8ff-15fa306d7a59',
+                'title': 'Overcoming Top 4 Challenges of Enterprise Video',
+                'uploader': 'Panopto Support',
+                'timestamp': 1449409251,
+                'thumbnail': r're:https://demo\.hosted\.panopto\.com/.+',
+                'upload_date': '20151206',
+                'ext': 'mp4',
+                'chapters': 'count:12',
+                'cast': ['Panopto Support'],
+                'uploader_id': 'a96d1a31-b4de-489b-9eee-b4a5b414372c',
+                'average_rating': int,
+                'description': 'md5:4391837802b3fc856dadf630c4b375d1',
+                'duration': 1088.2659999999998,
+                'channel_id': '9f3c1921-43bb-4bda-8b3a-b8d2f05a8546',
+                'channel': 'Webcasts',
+            },
+        },
+        {
+            # Extra params in URL
+            'url': 'https://howtovideos.hosted.panopto.com/Panopto/Pages/Viewer.aspx?randomparam=thisisnotreal&id=5fa74e93-3d87-4694-b60e-aaa4012214ed&advance=true',
+            'info_dict': {
+                'id': '5fa74e93-3d87-4694-b60e-aaa4012214ed',
+                'ext': 'mp4',
+                'duration': 129.513,
+                'cast': ['Kathryn Kelly'],
+                'uploader_id': '316a0a58-7fa2-4cd9-be1c-64270d284a56',
+                'timestamp': 1569845768,
+                'tags': ['Viewer', 'Enterprise'],
+                'chapters': [],
+                'upload_date': '20190930',
+                'thumbnail': r're:https://howtovideos\.hosted\.panopto\.com/.+',
+                'description': 'md5:2d844aaa1b1a14ad0e2601a0993b431f',
+                'title': 'Getting Started: View a Video',
+                'average_rating': int,
+                'uploader': 'Kathryn Kelly',
+                'channel_id': 'fb93bc3c-6750-4b80-a05b-a921013735d3',
+                'channel': 'Getting Started',
+            }
+        },
+        {
+            # Does not allow normal Viewer.aspx. AUDIO livestream has no url, so should be skipped and only give one stream.
+            'url': 'https://unisa.au.panopto.com/Panopto/Pages/Embed.aspx?id=9d9a0fa3-e99a-4ebd-a281-aac2017f4da4',
+            'info_dict': {
+                'id': '9d9a0fa3-e99a-4ebd-a281-aac2017f4da4',
+                'ext': 'mp4',
+                'cast': ['LTS CLI Script'],
+                'chapters': [],
+                'duration': 2178.45,
+                'description': 'md5:ee5cf653919f55b72bce2dbcf829c9fa',
+                'channel_id': 'b23e673f-c287-4cb1-8344-aae9005a69f8',
+                'average_rating': int,
+                'uploader_id': '38377323-6a23-41e2-9ff6-a8e8004bf6f7',
+                'uploader': 'LTS CLI Script',
+                'timestamp': 1572458134,
+                'title': 'WW2 Vets Interview 3 Ronald Stanley George',
+                'thumbnail': r're:https://unisa\.au\.panopto\.com/.+',
+                'channel': 'World War II Veteran Interviews',
+                'upload_date': '20191030',
+            },
+        },
+        {
+            # Slides/storyboard
+            'url': 'https://demo.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=a7f12f1d-3872-4310-84b0-f8d8ab15326b',
+            'info_dict': {
+                'id': 'a7f12f1d-3872-4310-84b0-f8d8ab15326b',
+                'ext': 'mhtml',
+                'timestamp': 1448798857,
+                'duration': 4712.681,
+                'title': 'Cache Memory - CompSci 15-213, Lecture 12',
+                'channel_id': 'e4c6a2fc-1214-4ca0-8fb7-aef2e29ff63a',
+                'uploader_id': 'a96d1a31-b4de-489b-9eee-b4a5b414372c',
+                'upload_date': '20151129',
+                'average_rating': 0,
+                'uploader': 'Panopto Support',
+                'channel': 'Showcase Videos',
+                'description': 'md5:55e51d54233ddb0e6c2ed388ca73822c',
+                'cast': ['ISR Videographer', 'Panopto Support'],
+                'chapters': 'count:28',
+                'thumbnail': r're:https://demo\.hosted\.panopto\.com/.+',
+            },
+            'params': {'format': 'mhtml', 'skip_download': True},
+            'skip': 'Not yet implemented',
+        },
+        {
+            'url': 'https://na-training-1.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=8285224a-9a2b-4957-84f2-acb0000c4ea9',
+            'info_dict': {
+                'id': '8285224a-9a2b-4957-84f2-acb0000c4ea9',
+                'ext': 'mp4',
+                'chapters': [],
+                'title': 'Company Policy',
+                'average_rating': 0,
+                'timestamp': 1615058901,
+                'channel': 'Human Resources',
+                'tags': ['HumanResources'],
+                'duration': 1604.243,
+                'thumbnail': r're:https://na-training-1\.hosted\.panopto\.com/.+',
+                'uploader_id': '8e8ba0a3-424f-40df-a4f1-ab3a01375103',
+                'uploader': 'Cait M.',
+                'upload_date': '20210306',
+                'cast': ['Cait M.'],
+                # 'subtitles': {'en-US': [{'ext': 'srt', 'data': 'md5:a3f4d25963fdeace838f327097c13265'}],
+                #               'es-ES': [{'ext': 'srt', 'data': 'md5:57e9dad365fd0fbaf0468eac4949f189'}]},
+            },
+            'params': {'writesubtitles': True, 'skip_download': True}
+        }, {
+            # On Panopto there are two subs: "Default" and en-US. en-US is blank and should be skipped.
+            'url': 'https://na-training-1.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=940cbd41-f616-4a45-b13e-aaf1000c915b',
+            'info_dict': {
+                'id': '940cbd41-f616-4a45-b13e-aaf1000c915b',
+                'ext': 'mp4',
+                'subtitles': 'count:1',
+                'title': 'HR Benefits Review Meeting*',
+                'cast': ['Panopto Support'],
+                'chapters': [],
+                'timestamp': 1575024251,
+                'thumbnail': r're:https://na-training-1\.hosted\.panopto\.com/.+',
+                'channel': 'Zoom',
+                'description': 'md5:04f90a9c2c68b7828144abfb170f0106',
+                'uploader': 'Panopto Support',
+                'average_rating': 0,
+                'duration': 409.34499999999997,
+                'uploader_id': 'b6ac04ad-38b8-4724-a004-a851004ea3df',
+                'upload_date': '20191129',
+
+            },
+            'params': {'writesubtitles': True, 'skip_download': True}
+        },
+        {
+            'url': 'https://ucc.cloud.panopto.eu/Panopto/Pages/Viewer.aspx?id=0e8484a4-4ceb-4d98-a63f-ac0200b455cb',
+            'only_matching': True
+        },
+        {
+            'url': 'https://brown.hosted.panopto.com/Panopto/Pages/Embed.aspx?id=0b3ff73b-36a0-46c5-8455-aadf010a3638',
+            'only_matching': True
+        },
+    ]
+
+    @classmethod
+    def suitable(cls, url):
+        return False if PanoptoPlaylistIE.suitable(url) else super(PanoptoIE, cls).suitable(url)
+
+    @classmethod
+    def _extract_from_webpage(cls, url, webpage):
+        return LazyList(map(
+            lambda u: cls.url_result(u, cls.ie_key()),
+            orderedSet(m.group('url') for m in itertools.chain(
+                *(re.finditer(embed_re, webpage) for embed_re in cls._EMBED_REGEX)))))
+
+    def _mark_watched(self, base_url, video_id, delivery_info):
+        duration = traverse_obj(delivery_info, ('Delivery', 'Duration'), expected_type=float)
+        invocation_id = delivery_info.get('InvocationId')
+        stream_id = traverse_obj(delivery_info, ('Delivery', 'Streams', Ellipsis, 'PublicID'), get_all=False, expected_type=compat_str)
+        if invocation_id and stream_id and duration:
+            timestamp_str = '/Date(%s000)/' % (calendar.timegm(datetime.utcnow().timetuple()), )
+            data = {
+                'streamRequests': [
+                    {
+                        'ClientTimeStamp': timestamp_str,
+                        'ID': 0,
+                        'InvocationID': invocation_id,
+                        'PlaybackSpeed': 1,
+                        'SecondsListened': duration - 1,
+                        'SecondsRejected': 0,
+                        'StartPosition': 0,
+                        'StartReason': 2,
+                        'StopReason': None,
+                        'StreamID': stream_id,
+                        'TimeStamp': timestamp_str,
+                        'UpdatesRejected': 0
+                    },
+                ]}
+
+            self._download_webpage(
+                base_url + '/Services/Analytics.svc/AddStreamRequests', video_id,
+                fatal=False, data=json.dumps(data).encode('utf8'), headers={'content-type': 'application/json'},
+                note='Marking watched', errnote='Unable to mark watched')
+
+    @staticmethod
+    def _extract_chapters(timestamps):
+        chapters = []
+        for timestamp in timestamps or []:
+            caption = timestamp.get('Caption')
+            start, duration = int_or_none(timestamp.get('Time')), int_or_none(timestamp.get('Duration'))
+            if not caption or start is None or duration is None:
+                continue
+            chapters.append({
+                'start_time': start,
+                'end_time': start + duration,
+                'title': caption
+            })
+        return chapters
+
+    @staticmethod
+    def _extract_mhtml_formats(base_url, timestamps):
+        image_frags = {}
+        for timestamp in timestamps or []:
+            duration = timestamp.get('Duration')
+            obj_id, obj_sn = timestamp.get('ObjectIdentifier'), timestamp.get('ObjectSequenceNumber'),
+            if timestamp.get('EventTargetType') == 'PowerPoint' and obj_id is not None and obj_sn is not None:
+                image_frags.setdefault('slides', []).append({
+                    'url': update_url_query(
+                        base_url + '/Pages/Viewer/Image.aspx', {
+                            'id': obj_id,
+                            'number': obj_sn,
+                        }),
+                    'duration': duration
+                })
+
+            obj_pid, session_id, abs_time = timestamp.get('ObjectPublicIdentifier'), timestamp.get('SessionID'), timestamp.get('AbsoluteTime')
+            if None not in (obj_pid, session_id, abs_time):
+                image_frags.setdefault('chapter', []).append({
+                    'url': update_url_query(
+                        base_url + '/Pages/Viewer/Thumb.aspx?isPrimary=false', {
+                            'eventTargetPID': obj_pid,
+                            'sessionPID': session_id,
+                            'number': obj_sn,
+                            'absoluteTime': abs_time,
+                        }),
+                    'duration': duration,
+                })
+        for name, fragments in image_frags.items():
+            yield {
+                'format_id': name,
+                'ext': 'mhtml',
+                'protocol': 'mhtml',
+                'acodec': 'none',
+                'vcodec': 'none',
+                'url': 'about:invalid',
+                'fragments': fragments
+            }
+
+    @staticmethod
+    def _json2srt(data, delivery):
+        SRT_CAPTION_FMT = '{0}\n{1} --> {2}\n{3}'
+
+        def gen_lines(dat, deliv):
+            for i, line in enumerate(dat):
+                start_time = line['Time']
+                duration = line.get('Duration')
+                if duration:
+                    end_time = start_time + duration
+                else:
+                    end_time = traverse_obj(dat, (i + 1, 'Time')) or deliv['Duration']
+                yield SRT_CAPTION_FMT.format(
+                    i + 1, srt_subtitles_timecode(start_time), srt_subtitles_timecode(end_time), line['Caption'])
+        return '\n\n'.join(gen_lines(data, delivery))
+
+    def _get_subtitles(self, base_url, video_id, delivery):
+        subtitles = {}
+        for lang in delivery.get('AvailableLanguages') or []:
+            response = self._call_api(
+                base_url, '/Pages/Viewer/DeliveryInfo.aspx', video_id, fatal=False,
+                note='Downloading captions JSON metadata', query={
+                    'deliveryId': video_id,
+                    'getCaptions': True,
+                    'language': compat_str(lang),
+                    'responseType': 'json'
+                }
+            )
+            if not isinstance(response, list):
+                continue
+            subtitles.setdefault(self._SUB_LANG_MAPPING.get(lang) or 'default', []).append({
+                'ext': 'srt',
+                'data': self._json2srt(response, delivery),
+            })
+        return subtitles
+
+    def _extract_streams_formats_and_subtitles(self, video_id, streams, **fmt_kwargs):
+        formats = []
+        subtitles = {}
+        for stream in streams or []:
+            stream_formats = []
+            http_stream_url = stream.get('StreamHttpUrl')
+            stream_url = stream.get('StreamUrl')
+
+            if http_stream_url:
+                stream_formats.append({'url': http_stream_url})
+
+            if stream_url:
+                media_type = stream.get('ViewerMediaFileTypeName')
+                if media_type in ('hls', ):
+                    # m3u8_formats, stream_subtitles = self._extract_m3u8_formats_and_subtitles(stream_url, video_id)
+                    m3u8_formats = self._extract_m3u8_formats(stream_url, video_id)
+                    stream_formats.extend(m3u8_formats)
+                    # subtitles = self._merge_subtitles(subtitles, stream_subtitles)
+                else:
+                    stream_formats.append({
+                        'url': stream_url
+                    })
+            for fmt in stream_formats:
+                fmt.update({'format_note': stream.get('Tag'), })
+                fmt.update(fmt_kwargs)
+            formats.extend(stream_formats)
+
+        return formats, subtitles
+
+    def _real_extract(self, url):
+        base_url, video_id = self._match_valid_url(url).group('base_url', 'id')
+        delivery_info = self._call_api(
+            base_url, '/Pages/Viewer/DeliveryInfo.aspx', video_id,
+            query={
+                'deliveryId': video_id,
+                'invocationId': '',
+                'isLiveNotes': 'false',
+                'refreshAuthCookie': 'true',
+                'isActiveBroadcast': 'false',
+                'isEditing': 'false',
+                'isKollectiveAgentInstalled': 'false',
+                'isEmbed': 'false',
+                'responseType': 'json',
+            }
+        )
+
+        delivery = delivery_info['Delivery']
+        session_start_time = int_or_none(delivery.get('SessionStartTime'))
+        timestamps = delivery.get('Timestamps')
+
+        # Podcast stream is usually the combined streams. We will prefer that by default.
+        podcast_formats, podcast_subtitles = self._extract_streams_formats_and_subtitles(
+            video_id, delivery.get('PodcastStreams'), format_note='PODCAST')
+
+        streams_formats, streams_subtitles = self._extract_streams_formats_and_subtitles(
+            video_id, delivery.get('Streams'), preference=-10)
+
+        formats = podcast_formats + streams_formats
+        formats.extend(self._extract_mhtml_formats(base_url, timestamps))
+        subtitles = self._merge_subtitles(podcast_subtitles, streams_subtitles)
+        subtitles = self._merge_subtitles(subtitles, self.extract_subtitles(base_url, video_id, delivery))
+
+        self._sort_formats(formats)
+        self.mark_watched(base_url, video_id, delivery_info)
+
+        return {
+            'id': video_id,
+            'title': delivery.get('SessionName'),
+            'cast': traverse_obj(delivery, ('Contributors', Ellipsis, 'DisplayName'), default=[], expected_type=lambda x: x or None),
+            'timestamp': session_start_time - 11640000000 if session_start_time else None,
+            'duration': delivery.get('Duration'),
+            'thumbnail': update_url_query(
+                base_url + '/Services/FrameGrabber.svc/FrameRedirect?mode=Delivery', {
+                    'objectId': video_id,
+                    'random': random(),
+                }),
+            'average_rating': delivery.get('AverageRating'),
+            'chapters': self._extract_chapters(timestamps),
+            'uploader': delivery.get('OwnerDisplayName') or None,
+            'uploader_id': delivery.get('OwnerId'),
+            'description': delivery.get('SessionAbstract'),
+            'tags': traverse_obj(delivery, ('Tags', Ellipsis, 'Content')),
+            'channel_id': delivery.get('SessionGroupPublicID'),
+            'channel': traverse_obj(delivery, 'SessionGroupLongName', 'SessionGroupShortName', get_all=False),
+            'formats': formats,
+            'subtitles': subtitles
+        }
+
+
+class PanoptoPlaylistIE(PanoptoBaseIE):
+    _VALID_URL = PanoptoBaseIE.BASE_URL_RE + r'/Pages/(Viewer|Embed)\.aspx.*(?:\?|&)pid=(?P<id>[a-f0-9-]+)'
+    _TESTS = [
+        {
+            'url': 'https://howtovideos.hosted.panopto.com/Panopto/Pages/Viewer.aspx?pid=f3b39fcf-882f-4849-93d6-a9f401236d36&id=5fa74e93-3d87-4694-b60e-aaa4012214ed&advance=true',
+            'info_dict': {
+                'title': 'Featured Video Tutorials',
+                'id': 'f3b39fcf-882f-4849-93d6-a9f401236d36',
+            },
+            'playlist_mincount': 34,  # was 36
+        },
+        {
+            'url': 'https://utsa.hosted.panopto.com/Panopto/Pages/Viewer.aspx?pid=e2900555-3ad4-4bdb-854d-ad2401686190',
+            'info_dict': {
+                'title': 'Library Website Introduction Playlist',
+                'id': 'e2900555-3ad4-4bdb-854d-ad2401686190',
+                'description': 'md5:f958bca50a1cbda15fdc1e20d32b3ecb',
+            },
+            'playlist_mincount': 4
+        },
+    ]
+
+    def _entries(self, base_url, playlist_id, session_list_id):
+        session_list_info = self._call_api(
+            base_url,
+            '/Api/SessionLists/%s?collections[0].maxCount=500&collections[0].name=items' % (session_list_id, ),
+            playlist_id)
+
+        items = session_list_info['Items']
+        for item in items:
+            if item.get('TypeName') != 'Session':
+                self.report_warning('Got an item in the playlist that is not a Session' + bug_reports_message(), only_once=True)
+                continue
+            yield merge_dicts(
+                self.url_result(item.get('ViewerUri'), item.get('Id'), item.get('Name')), {
+                    'description': item.get('Description'),
+                    'duration': item.get('Duration'),
+                    'channel': traverse_obj(item, ('Parent', 'Name')),
+                    'channel_id': traverse_obj(item, ('Parent', 'Id'))
+                })
+
+    def _real_extract(self, url):
+        base_url, playlist_id = self._match_valid_url(url).group('base_url', 'id')
+
+        video_id = get_first(parse_qs(url), 'id')
+        if video_id:
+            if self.get_param('noplaylist'):
+                self.to_screen('Downloading just video %s because of --no-playlist' % video_id)
+                return self.url_result(update_url_query(base_url + '/Pages/Viewer.aspx', {'id': video_id}), ie_key=PanoptoIE.ie_key(), video_id=video_id)
+            else:
+                self.to_screen('Downloading playlist {playlist_id}; add --no-playlist to just download video {video_id}'.format(**locals()))
+
+        playlist_info = self._call_api(base_url, '/Api/Playlists/' + playlist_id, playlist_id)
+        return self.playlist_result(
+            self._entries(base_url, playlist_id, playlist_info['SessionListId']),
+            playlist_id=playlist_id, playlist_title=playlist_info.get('Name'),
+            playlist_description=playlist_info.get('Description'))
+
+
+class PanoptoListIE(PanoptoBaseIE):
+    _VALID_URL = PanoptoBaseIE.BASE_URL_RE + r'/Pages/Sessions/List\.aspx'
+    _PAGE_SIZE = 250
+    _TESTS = [
+        {
+            'url': 'https://demo.hosted.panopto.com/Panopto/Pages/Sessions/List.aspx#folderID=%22e4c6a2fc-1214-4ca0-8fb7-aef2e29ff63a%22',
+            'info_dict': {
+                'id': 'e4c6a2fc-1214-4ca0-8fb7-aef2e29ff63a',
+                'title': 'Showcase Videos'
+            },
+            'playlist_mincount': 140
+
+        },
+        {
+            'url': 'https://demo.hosted.panopto.com/Panopto/Pages/Sessions/List.aspx#view=2&maxResults=250',
+            'info_dict': {
+                'id': 'panopto_list',
+                'title': 'panopto_list'
+            },
+            'playlist_mincount': 300
+        },
+        {
+            # Folder that contains 8 folders and a playlist
+            'url': 'https://howtovideos.hosted.panopto.com/Panopto/Pages/Sessions/List.aspx?noredirect=true#folderID=%224b9de7ae-0080-4158-8496-a9ba01692c2e%22',
+            'info_dict': {
+                'id': '4b9de7ae-0080-4158-8496-a9ba01692c2e',
+                'title': 'Video Tutorials'
+            },
+            'playlist_mincount': 9
+        }
+
+    ]
+
+    def _fetch_page(self, base_url, query_params, display_id, page):
+        params = merge_dicts({
+            'page': page,
+            'maxResults': self._PAGE_SIZE,
+        }, query_params, {
+            'sortColumn': 1,
+            'getFolderData': True,
+            'includePlaylists': True,
+        })
+
+        response = self._call_api(
+            base_url, '/Services/Data.svc/GetSessions', '%s page %d' % (display_id, page + 1),
+            data={'queryParameters': params}, fatal=False)
+
+        for result in get_first(response, 'Results', default=[]):
+            # This could be a video, playlist (or maybe something else)
+            item_id = result.get('DeliveryID')
+            yield merge_dicts(
+                self.url_result(
+                    traverse_obj(result, 'ViewerUrl', 'EmbedUrl', get_all=False)
+                    or update_url_query(base_url + '/Pages/Viewer.aspx', {'id': item_id}),
+                    item_id, result.get('SessionName')), {
+                    'duration': result.get('Duration'),
+                    'channel': result.get('FolderName'),
+                    'channel_id': result.get('FolderID'),
+                })
+
+        for folder in get_first(response, 'Subfolders', default=[]):
+            folder_id = folder.get('ID')
+            yield self.url_result(
+                '%s/Pages/Sessions/List.aspx#folderID=%s' % (base_url, folder_id),
+                ie_key=PanoptoListIE.ie_key(), video_id=folder_id, title=folder.get('Name'))
+
+    def _extract_folder_metadata(self, base_url, folder_id):
+        response = self._call_api(
+            base_url, '/Services/Data.svc/GetFolderInfo', folder_id,
+            data={'folderID': folder_id}, fatal=False)
+        return {
+            'title': get_first(response, 'Name', default=[])
+        }
+
+    def _real_extract(self, url):
+        mobj = self._match_valid_url(url)
+        base_url = mobj.group('base_url')
+
+        query_params = self._parse_fragment(url)
+        folder_id, display_id = query_params.get('folderID'), 'panopto_list'
+
+        if query_params.get('isSubscriptionsPage'):
+            display_id = 'subscriptions'
+            if not query_params.get('subscribableTypes'):
+                query_params['subscribableTypes'] = [0, 1, 2]
+        elif query_params.get('isSharedWithMe'):
+            display_id = 'sharedwithme'
+        elif folder_id:
+            display_id = folder_id
+
+        query = query_params.get('query')
+        if query:
+            display_id += ': query "%s"' % (query, )
+
+        info = self.playlist_result(
+            OnDemandPagedList(
+                functools.partial(self._fetch_page, base_url, query_params, display_id), self._PAGE_SIZE),
+            playlist_id=display_id,
+            playlist_title=display_id)
+
+        if folder_id:
+            info.update(self._extract_folder_metadata(base_url, folder_id))
+
+        return info

youtube_dl/extractor/youtube.py

@@ -27,6 +27,7 @@ from ..compat import (
 )
 from ..jsinterp import JSInterpreter
 from ..utils import (
+    bug_reports_message,
     clean_html,
     dict_get,
     error_to_compat_str,
@@ -65,6 +66,7 @@ from ..utils import (
     url_or_none,
     urlencode_postdata,
     urljoin,
+    variadic,
 )
 
 
@@ -89,12 +91,12 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'IOS',
-                    'clientVersion': '19.45.4',
+                    'clientVersion': '20.10.4',
                     'deviceMake': 'Apple',
                     'deviceModel': 'iPhone16,2',
-                    'userAgent': 'com.google.ios.youtube/19.45.4 (iPhone16,2; U; CPU iOS 18_1_0 like Mac OS X;)',
+                    'userAgent': 'com.google.ios.youtube/20.10.4 (iPhone16,2; U; CPU iOS 18_3_2 like Mac OS X;)',
                     'osName': 'iPhone',
-                    'osVersion': '18.1.0.22B83',
+                    'osVersion': '18.3.2.22D82',
                 },
             },
             'INNERTUBE_CONTEXT_CLIENT_NAME': 5,
@@ -107,7 +109,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'MWEB',
-                    'clientVersion': '2.20241202.07.00',
+                    'clientVersion': '2.20250311.03.00',
                     # mweb previously did not require PO Token with this UA
                     'userAgent': 'Mozilla/5.0 (iPad; CPU OS 16_7_10 like Mac OS X) AppleWebKit/605.1.15 (KHTML, like Gecko) Version/16.6 Mobile/15E148 Safari/604.1,gzip(gfe)',
                 },
@@ -120,7 +122,8 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'TVHTML5',
-                    'clientVersion': '7.20241201.18.00',
+                    'clientVersion': '7.20250312.16.00',
+                    'userAgent': 'Mozilla/5.0 (ChromiumStylePlatform) Cobalt/Version',
                 },
             },
             'INNERTUBE_CONTEXT_CLIENT_NAME': 7,
@@ -130,7 +133,7 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'INNERTUBE_CONTEXT': {
                 'client': {
                     'clientName': 'WEB',
-                    'clientVersion': '2.20241126.01.00',
+                    'clientVersion': '2.20250312.04.00',
                 },
             },
             'INNERTUBE_CONTEXT_CLIENT_NAME': 1,
@@ -460,6 +463,26 @@ class YoutubeBaseInfoExtractor(InfoExtractor):
             'uploader': uploader,
         }
 
+    @staticmethod
+    def _extract_thumbnails(data, *path_list, **kw_final_key):
+        """
+        Extract thumbnails from thumbnails dict
+        @param path_list: path list to level that contains 'thumbnails' key
+        """
+        final_key = kw_final_key.get('final_key', 'thumbnails')
+
+        return traverse_obj(data, ((
+            tuple(variadic(path) + (final_key, Ellipsis)
+                  for path in path_list or [()])), {
+            'url': ('url', T(url_or_none),
+                    # Sometimes youtube gives a wrong thumbnail URL. See:
+                    # https://github.com/yt-dlp/yt-dlp/issues/233
+                    # https://github.com/ytdl-org/youtube-dl/issues/28023
+                    T(lambda u: update_url(u, query=None) if u and 'maxresdefault' in u else u)),
+            'height': ('height', T(int_or_none)),
+            'width': ('width', T(int_or_none)),
+        }, T(lambda t: t if t.get('url') else None)))
+
     def _search_results(self, query, params):
         data = {
             'context': {
@@ -669,7 +692,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         'invidious': '|'.join(_INVIDIOUS_SITES),
     }
     _PLAYER_INFO_RE = (
-        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player',
+        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})//(?:tv-)?player',
         r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
         r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
     )
@@ -1829,12 +1852,22 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
         if func_code:
             return jsi, player_id, func_code
+        return self._extract_n_function_code_jsi(video_id, jsi, player_id)
 
-        func_name = self._extract_n_function_name(jscode)
+    def _extract_n_function_code_jsi(self, video_id, jsi, player_id=None):
+
+        var_ay = self._search_regex(
+            r'(?:[;\s]|^)\s*(var\s*[\w$]+\s*=\s*"(?:\\"|[^"])+"\s*\.\s*split\("\W+"\))(?=\s*[,;])',
+            jsi.code, 'useful values', default='')
+
+        func_name = self._extract_n_function_name(jsi.code)
 
         func_code = jsi.extract_function_code(func_name)
+        if var_ay:
+            func_code = (func_code[0], ';\n'.join((var_ay, func_code[1])))
 
-        self.cache.store('youtube-nsig', player_id, func_code)
+        if player_id:
+            self.cache.store('youtube-nsig', player_id, func_code)
         return jsi, player_id, func_code
 
     def _extract_n_function_from_code(self, jsi, func_code):
@@ -3183,8 +3216,12 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
             expected_type=txt_or_none)
 
     def _grid_entries(self, grid_renderer):
-        for item in grid_renderer['items']:
+        for item in traverse_obj(grid_renderer, ('items', Ellipsis, T(dict))):
-            if not isinstance(item, dict):
+            lockup_view_model = traverse_obj(item, ('lockupViewModel', T(dict)))
+            if lockup_view_model:
+                entry = self._extract_lockup_view_model(lockup_view_model)
+                if entry:
+                    yield entry
                 continue
             renderer = self._extract_grid_item_renderer(item)
             if not isinstance(renderer, dict):
@@ -3268,6 +3305,25 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor):
                 continue
             yield self._extract_video(renderer)
 
+    def _extract_lockup_view_model(self, view_model):
+        content_id = view_model.get('contentId')
+        if not content_id:
+            return
+        content_type = view_model.get('contentType')
+        if content_type not in ('LOCKUP_CONTENT_TYPE_PLAYLIST', 'LOCKUP_CONTENT_TYPE_PODCAST'):
+            self.report_warning(
+                'Unsupported lockup view model content type "{0}"{1}'.format(content_type, bug_reports_message()), only_once=True)
+            return
+        return merge_dicts(self.url_result(
+            update_url_query('https://www.youtube.com/playlist', {'list': content_id}),
+            ie=YoutubeTabIE.ie_key(), video_id=content_id), {
+                'title': traverse_obj(view_model, (
+                    'metadata', 'lockupMetadataViewModel', 'title', 'content', T(compat_str))),
+                'thumbnails': self._extract_thumbnails(view_model, (
+                    'contentImage', 'collectionThumbnailViewModel', 'primaryThumbnail',
+                    'thumbnailViewModel', 'image'), final_key='sources'),
+        })
+
     def _video_entry(self, video_renderer):
         video_id = video_renderer.get('videoId')
         if video_id:

youtube_dl/jsinterp.py

@@ -1,10 +1,12 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
+import calendar
 import itertools
 import json
 import operator
 import re
+import time
 
 from functools import update_wrapper, wraps
 
@@ -12,8 +14,10 @@ from .utils import (
     error_to_compat_str,
     ExtractorError,
     float_or_none,
+    int_or_none,
     js_to_json,
     remove_quotes,
+    str_or_none,
     unified_timestamp,
     variadic,
     write_string,
@@ -150,6 +154,7 @@ def _js_to_primitive(v):
     )
 
 
+# more exact: yt-dlp/yt-dlp#12110
 def _js_toString(v):
     return (
         'undefined' if v is JS_Undefined
@@ -158,7 +163,7 @@ def _js_toString(v):
         else 'null' if v is None
         # bool <= int: do this first
         else ('false', 'true')[v] if isinstance(v, bool)
-        else '{0:.7f}'.format(v).rstrip('.0') if isinstance(v, compat_numeric_types)
+        else re.sub(r'(?<=\d)\.?0*$', '', '{0:.7f}'.format(v)) if isinstance(v, compat_numeric_types)
         else _js_to_primitive(v))
 
 
@@ -404,6 +409,7 @@ class JSInterpreter(object):
     class Exception(ExtractorError):
         def __init__(self, msg, *args, **kwargs):
             expr = kwargs.pop('expr', None)
+            msg = str_or_none(msg, default='"None"')
             if expr is not None:
                 msg = '{0} in: {1!r:.100}'.format(msg.rstrip(), expr)
             super(JSInterpreter.Exception, self).__init__(msg, *args, **kwargs)
@@ -431,6 +437,7 @@ class JSInterpreter(object):
                 flags, _ = self.regex_flags(flags)
             # First, avoid https://github.com/python/cpython/issues/74534
             self.__self = None
+            pattern_txt = str_or_none(pattern_txt) or '(?:)'
             self.__pattern_txt = pattern_txt.replace('[[', r'[\[')
             self.__flags = flags
 
@@ -475,6 +482,73 @@ class JSInterpreter(object):
                 flags |= cls.RE_FLAGS[ch]
             return flags, expr[idx + 1:]
 
+    class JS_Date(object):
+        _t = None
+
+        @staticmethod
+        def __ymd_etc(*args, **kw_is_utc):
+            # args: year, monthIndex, day, hours, minutes, seconds, milliseconds
+            is_utc = kw_is_utc.get('is_utc', False)
+
+            args = list(args[:7])
+            args += [0] * (9 - len(args))
+            args[1] += 1  # month 0..11 -> 1..12
+            ms = args[6]
+            for i in range(6, 9):
+                args[i] = -1  # don't know
+            if is_utc:
+                args[-1] = 1
+            # TODO: [MDN] When a segment overflows or underflows its expected
+            # range, it usually "carries over to" or "borrows from" the higher segment.
+            try:
+                mktime = calendar.timegm if is_utc else time.mktime
+                return mktime(time.struct_time(args)) * 1000 + ms
+            except (OverflowError, ValueError):
+                return None
+
+        @classmethod
+        def UTC(cls, *args):
+            t = cls.__ymd_etc(*args, is_utc=True)
+            return _NaN if t is None else t
+
+        @staticmethod
+        def parse(date_str, **kw_is_raw):
+            is_raw = kw_is_raw.get('is_raw', False)
+
+            t = unified_timestamp(str_or_none(date_str), False)
+            return int(t * 1000) if t is not None else t if is_raw else _NaN
+
+        @staticmethod
+        def now(**kw_is_raw):
+            is_raw = kw_is_raw.get('is_raw', False)
+
+            t = time.time()
+            return int(t * 1000) if t is not None else t if is_raw else _NaN
+
+        def __init__(self, *args):
+            if not args:
+                args = [self.now(is_raw=True)]
+            if len(args) == 1:
+                if isinstance(args[0], JSInterpreter.JS_Date):
+                    self._t = int_or_none(args[0].valueOf(), default=None)
+                else:
+                    arg_type = _js_typeof(args[0])
+                    if arg_type == 'string':
+                        self._t = self.parse(args[0], is_raw=True)
+                    elif arg_type == 'number':
+                        self._t = int(args[0])
+            else:
+                self._t = self.__ymd_etc(*args)
+
+        def toString(self):
+            try:
+                return time.strftime('%a %b %0d %Y %H:%M:%S %Z%z', self._t).rstrip()
+            except TypeError:
+                return "Invalid Date"
+
+        def valueOf(self):
+            return _NaN if self._t is None else self._t
+
     @classmethod
     def __op_chars(cls):
         op_chars = set(';,[')
@@ -599,14 +673,15 @@ class JSInterpreter(object):
         except Exception as e:
             raise self.Exception('Failed to evaluate {left_val!r:.50} {op} {right_val!r:.50}'.format(**locals()), expr, cause=e)
 
-    def _index(self, obj, idx, allow_undefined=True):
+    def _index(self, obj, idx, allow_undefined=None):
         if idx == 'length' and isinstance(obj, list):
             return len(obj)
         try:
             return obj[int(idx)] if isinstance(obj, list) else obj[compat_str(idx)]
         except (TypeError, KeyError, IndexError) as e:
-            if allow_undefined:
+            # allow_undefined is None gives correct behaviour
-                # when is not allowed?
+            if allow_undefined or (
+                    allow_undefined is None and not isinstance(e, TypeError)):
                 return JS_Undefined
             raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
 
@@ -715,7 +790,7 @@ class JSInterpreter(object):
 
         new_kw, _, obj = expr.partition('new ')
         if not new_kw:
-            for klass, konstr in (('Date', lambda x: int(unified_timestamp(x, False) * 1000)),
+            for klass, konstr in (('Date', lambda *x: self.JS_Date(*x).valueOf()),
                                   ('RegExp', self.JS_RegExp),
                                   ('Error', self.Exception)):
                 if not obj.startswith(klass + '('):
@@ -1034,6 +1109,7 @@ class JSInterpreter(object):
                     'String': compat_str,
                     'Math': float,
                     'Array': list,
+                    'Date': self.JS_Date,
                 }
                 obj = local_vars.get(variable)
                 if obj in (JS_Undefined, None):
@@ -1086,6 +1162,8 @@ class JSInterpreter(object):
                         assertion(len(argvals) == 2, 'takes two arguments')
                         return argvals[0] ** argvals[1]
                     raise self.Exception('Unsupported Math method ' + member, expr=expr)
+                elif obj is self.JS_Date:
+                    return getattr(obj, member)(*argvals)
 
                 if member == 'split':
                     assertion(len(argvals) <= 2, 'takes at most two arguments')

youtube_dl/options.py

@@ -186,11 +186,11 @@ def parseOpts(overrideArguments=None):
     general.add_option(
         '--mark-watched',
         action='store_true', dest='mark_watched', default=False,
-        help='Mark videos watched (YouTube only)')
+        help='Mark videos watched (if supported for site)')
     general.add_option(
         '--no-mark-watched',
         action='store_false', dest='mark_watched', default=False,
-        help='Do not mark videos watched (YouTube only)')
+        help='Do not mark videos watched')
     general.add_option(
         '--no-color', '--no-colors',
         action='store_true', dest='no_color',