Merge b83889a20eb01094a3361cf08f27895fea164049 into 2b4fbfce25902d557b86b003cf48f738129efce4

thx seproDev, bashonly: yt-dlp/yt-dlp#12748

test/test_youtube_signature.py

@@ -84,6 +84,21 @@ _SIG_TESTS = [
         '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
         '0QJ8wRAIgXmPlOPSBkkUs1bYFYlJCfe29xxAj7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJ2OySqa0q',
     ),
+    (
+        'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js',
+        '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
+        'AAOAOq0QJ8wRAIgXmPlOPSBkkUs1bYFYlJCfe29xx8j7vgpDL0QwbdV06sCIEzpWqMGkFR20CFOS21Tp-7vj_EMu-m37KtXJoOy1',
+    ),
+    (
+        'https://www.youtube.com/s/player/363db69b/player_ias.vflset/en_US/base.js',
+        '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
+        '0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpz2ICs6EVdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
+    ),
+    (
+        'https://www.youtube.com/s/player/4fcd6e4a/player_ias.vflset/en_US/base.js',
+        '2aq0aqSyOoJXtK73m-uME_jv7-pT15gOFC02RFkGMqWpzEICs69VdbwQ0LDp1v7j8xx92efCJlYFYb1sUkkBSPOlPmXgIARw8JQ0qOAOAA',
+        'wAOAOq0QJ8ARAIgXmPlOPSBkkUs1bYFYlJCfe29xx8q7v1pDL0QwbdV96sCIEzpWqMGkFR20CFOg51Tp-7vj_EMu-m37KtXJoOySqa0',
+    ),
 ]
 
 _NSIG_TESTS = [
@@ -153,7 +168,7 @@ _NSIG_TESTS = [
     ),
     (
         'https://www.youtube.com/s/player/c57c113c/player_ias.vflset/en_US/base.js',
-        '-Txvy6bT5R6LqgnQNx', 'dcklJCnRUHbgSg',
+        'M92UUMHa8PdvPd3wyM', '3hPqLJsiNZx7yA',
     ),
     (
         'https://www.youtube.com/s/player/5a3b6271/player_ias.vflset/en_US/base.js',
@@ -173,7 +188,7 @@ _NSIG_TESTS = [
     ),
     (
         'https://www.youtube.com/s/player/cfa9e7cb/player_ias.vflset/en_US/base.js',
-        'qO0NiMtYQ7TeJnfFG2', 'k9cuJDHNS5O7kQ',
+        'aCi3iElgd2kq0bxVbQ', 'QX1y8jGb2IbZ0w',
     ),
     (
         'https://www.youtube.com/s/player/8c7583ff/player_ias.vflset/en_US/base.js',
@@ -231,10 +246,6 @@ _NSIG_TESTS = [
         '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',
@@ -259,6 +270,22 @@ _NSIG_TESTS = [
         'https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js',
         'ir9-V6cdbCiyKxhr', '2PL7ZDYAALMfmA',
     ),
+    (
+        'https://www.youtube.com/s/player/643afba4/player_ias.vflset/en_US/base.js',
+        'W9HJZKktxuYoDTqW', 'larxUlagTRAcSw',
+    ),
+    (
+        'https://www.youtube.com/s/player/363db69b/player_ias.vflset/en_US/base.js',
+        'eWYu5d5YeY_4LyEDc', 'XJQqf-N7Xra3gg',
+    ),
+    (
+        'https://www.youtube.com/s/player/4fcd6e4a/player_ias.vflset/en_US/base.js',
+        'o_L251jm8yhZkWtBW', 'lXoxI3XvToqn6A',
+    ),
+    (
+        'https://www.youtube.com/s/player/4fcd6e4a/tv-player-ias.vflset/tv-player-ias.js',
+        'o_L251jm8yhZkWtBW', 'lXoxI3XvToqn6A',
+    ),
 ]
 
 
@@ -271,6 +298,8 @@ class TestPlayerInfo(unittest.TestCase):
             ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-en_US.vflset/base.js', '64dddad9'),
             ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-phone-de_DE.vflset/base.js', '64dddad9'),
             ('https://www.youtube.com/s/player/64dddad9/player-plasma-ias-tablet-en_US.vflset/base.js', '64dddad9'),
+            ('https://www.youtube.com/s/player/e7567ecf/player_ias_tce.vflset/en_US/base.js', 'e7567ecf'),
+            ('https://www.youtube.com/s/player/643afba4/tv-player-ias.vflset/tv-player-ias.js', '643afba4'),
             # obsolete
             ('https://www.youtube.com/yts/jsbin/player_ias-vfle4-e03/en_US/base.js', 'vfle4-e03'),
             ('https://www.youtube.com/yts/jsbin/player_ias-vfl49f_g4/en_US/base.js', 'vfl49f_g4'),
@@ -280,8 +309,9 @@ class TestPlayerInfo(unittest.TestCase):
             ('https://s.ytimg.com/yts/jsbin/html5player-en_US-vflXGBaUN.js', 'vflXGBaUN'),
             ('https://s.ytimg.com/yts/jsbin/html5player-en_US-vflKjOTVq/html5player.js', 'vflKjOTVq'),
         )
+        ie = YoutubeIE(FakeYDL({'cachedir': False}))
         for player_url, expected_player_id in PLAYER_URLS:
-            player_id = YoutubeIE._extract_player_info(player_url)
+            player_id = ie._extract_player_info(player_url)
             self.assertEqual(player_id, expected_player_id)
 
 
@@ -301,8 +331,8 @@ class TestSignature(unittest.TestCase):
 def t_factory(name, sig_func, url_pattern):
     def make_tfunc(url, sig_input, expected_sig):
         m = url_pattern.match(url)
-        assert m, '%r should follow URL format' % url
+        assert m, '{0!r} should follow URL format'.format(url)
-        test_id = m.group('id')
+        test_id = re.sub(r'[/.-]', '_', m.group('id') or m.group('compat_id'))
 
         def test_func(self):
             basename = 'player-{0}-{1}.js'.format(name, test_id)
@@ -335,12 +365,16 @@ def n_sig(jscode, sig_input):
 
 
 make_sig_test = t_factory(
-    'signature', signature, re.compile(r'.*(?:-|/player/)(?P<id>[a-zA-Z0-9_-]+)(?:/.+\.js|(?:/watch_as3|/html5player)?\.[a-z]+)$'))
+    'signature', signature,
+    re.compile(r'''(?x)
+        .+/(?P<h5>html5)?player(?(h5)(?:-en_US)?-|/)(?P<id>[a-zA-Z0-9/._-]+)
+        (?(h5)/(?:watch_as3|html5player))?\.js$
+    '''))
 for test_spec in _SIG_TESTS:
     make_sig_test(*test_spec)
 
 make_nsig_test = t_factory(
-    'nsig', n_sig, re.compile(r'.+/player/(?P<id>[a-zA-Z0-9_-]+)/.+.js$'))
+    'nsig', n_sig, re.compile(r'.+/player/(?P<id>[a-zA-Z0-9_/.-]+)\.js$'))
 for test_spec in _NSIG_TESTS:
     make_nsig_test(*test_spec)
 

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

@@ -692,9 +692,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         'invidious': '|'.join(_INVIDIOUS_SITES),
     }
     _PLAYER_INFO_RE = (
-        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})//(?:tv-)?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$',
+        r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias(?:_tce)?\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$',
-        r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
+        r'\b(?P<id>vfl[a-zA-Z0-9_-]{6,})\b.*?\.js$',
     )
     _SUBTITLE_FORMATS = ('json3', 'srv1', 'srv2', 'srv3', 'ttml', 'vtt')
 
@@ -1626,15 +1626,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         """ Return a string representation of a signature """
         return '.'.join(compat_str(len(part)) for part in example_sig.split('.'))
 
-    @classmethod
+    def _extract_player_info(self, player_url):
-    def _extract_player_info(cls, player_url):
+        try:
-        for player_re in cls._PLAYER_INFO_RE:
+            return self._search_regex(
-            id_m = re.search(player_re, player_url)
+                self._PLAYER_INFO_RE, player_url, 'player info', group='id')
-            if id_m:
+        except ExtractorError as e:
-                break
+            raise ExtractorError(
-        else:
+                'Cannot identify player %r' % (player_url,), cause=e)
-            raise ExtractorError('Cannot identify player %r' % player_url)
-        return id_m.group('id')
 
     def _load_player(self, video_id, player_url, fatal=True, player_id=None):
         if not player_id:
@@ -1711,6 +1709,23 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 '    return %s\n') % (signature_id_tuple, expr_code)
         self.to_screen('Extracted signature function:\n' + code)
 
+    def _extract_sig_fn(self, jsi, funcname):
+        var_ay = self._search_regex(
+            r'''(?x)
+                (?:\*/|\{|\n|^)\s*(?:'[^']+'\s*;\s*)
+                    (var\s*[\w$]+\s*=\s*(?:
+                        ('|")(?:\\\2|(?!\2).)+\2\s*\.\s*split\(\s*('|")\W+\3\s*\)|
+                        \[\s*(?:('|")(?:\\\4|(?!\4).)*\4\s*(?:(?=\])|,\s*))+\]
+                    ))(?=\s*[,;])
+            ''', jsi.code, 'useful values', default='')
+
+        sig_fn = jsi.extract_function_code(funcname)
+
+        if var_ay:
+            sig_fn = (sig_fn[0], ';\n'.join((var_ay, sig_fn[1])))
+
+        return sig_fn
+
     def _parse_sig_js(self, jscode):
         # Examples where `sig` is funcname:
         # sig=function(a){a=a.split(""); ... ;return a.join("")};
@@ -1736,8 +1751,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             jscode, 'Initial JS player signature function name', group='sig')
 
         jsi = JSInterpreter(jscode)
-        initial_function = jsi.extract_function(funcname)
+
-        return lambda s: initial_function([s])
+        initial_function = self._extract_sig_fn(jsi, funcname)
+
+        func = jsi.extract_function_from_code(*initial_function)
+
+        return lambda s: func([s])
 
     def _cached(self, func, *cache_id):
         def inner(*args, **kwargs):
@@ -1856,15 +1875,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
 
     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)
+        func_code = self._extract_sig_fn(jsi, func_name)
-        if var_ay:
-            func_code = (func_code[0], ';\n'.join((var_ay, func_code[1])))
 
         if player_id:
             self.cache.store('youtube-nsig', player_id, func_code)
@@ -2136,7 +2149,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                     video_details = merge_dicts(*traverse_obj(
                         (player_response, api_player_response),
                         (Ellipsis, 'videoDetails', T(dict))))
-                    player_response.update(api_player_response or {})
+                    player_response.update(filter_dict(
+                        api_player_response or {}, cndn=lambda k, _: k != 'captions'))
                     player_response['videoDetails'] = video_details
 
         def is_agegated(playability):
@@ -2566,8 +2580,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
         }
 
         pctr = traverse_obj(
-            player_response,
+            (player_response, api_player_response),
-            ('captions', 'playerCaptionsTracklistRenderer', T(dict)))
+            (Ellipsis, 'captions', 'playerCaptionsTracklistRenderer', T(dict)))
         if pctr:
             def process_language(container, base_url, lang_code, query):
                 lang_subs = []
@@ -2584,20 +2598,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
             def process_subtitles():
                 subtitles = {}
                 for caption_track in traverse_obj(pctr, (
-                        'captionTracks', lambda _, v: v.get('baseUrl'))):
+                        Ellipsis, 'captionTracks', lambda _, v: (
+                            v.get('baseUrl') and v.get('languageCode')))):
                     base_url = self._yt_urljoin(caption_track['baseUrl'])
                     if not base_url:
                         continue
+                    lang_code = caption_track['languageCode']
                     if caption_track.get('kind') != 'asr':
-                        lang_code = caption_track.get('languageCode')
-                        if not lang_code:
-                            continue
                         process_language(
                             subtitles, base_url, lang_code, {})
                         continue
                     automatic_captions = {}
+                    process_language(
+                        automatic_captions, base_url, lang_code, {})
                     for translation_language in traverse_obj(pctr, (
-                            'translationLanguages', lambda _, v: v.get('languageCode'))):
+                            Ellipsis, 'translationLanguages', lambda _, v: v.get('languageCode'))):
                         translation_language_code = translation_language['languageCode']
                         process_language(
                             automatic_captions, base_url, translation_language_code,

youtube_dl/jsinterp.py

@@ -678,7 +678,7 @@ class JSInterpreter(object):
             return len(obj)
         try:
             return obj[int(idx)] if isinstance(obj, list) else obj[compat_str(idx)]
-        except (TypeError, KeyError, IndexError) as e:
+        except (TypeError, KeyError, IndexError, ValueError) as e:
             # allow_undefined is None gives correct behaviour
             if allow_undefined or (
                     allow_undefined is None and not isinstance(e, TypeError)):
@@ -1038,6 +1038,10 @@ class JSInterpreter(object):
                     left_val = self._index(left_val, idx)
             if isinstance(idx, float):
                 idx = int(idx)
+            if isinstance(left_val, list) and len(left_val) <= int_or_none(idx, default=-1):
+                # JS Array is a sparsely assignable list
+                # TODO: handle extreme sparsity without memory bloat, eg using auxiliary dict
+                left_val.extend((idx - len(left_val) + 1) * [JS_Undefined])
             left_val[idx] = self._operator(
                 m.group('op'), self._index(left_val, idx) if m.group('op') else None,
                 m.group('expr'), expr, local_vars, allow_recursion)
@@ -1204,9 +1208,10 @@ class JSInterpreter(object):
                 elif member == 'join':
                     assertion(isinstance(obj, list), 'must be applied on a list')
                     assertion(len(argvals) <= 1, 'takes at most one argument')
-                    return (',' if len(argvals) == 0 else argvals[0]).join(
+                    return (',' if len(argvals) == 0 or argvals[0] in (None, JS_Undefined)
-                        ('' if x in (None, JS_Undefined) else _js_toString(x))
+                            else argvals[0]).join(
-                        for x in obj)
+                                ('' if x in (None, JS_Undefined) else _js_toString(x))
+                                for x in obj)
                 elif member == 'reverse':
                     assertion(not argvals, 'does not take any arguments')
                     obj.reverse()
@@ -1364,19 +1369,21 @@ class JSInterpreter(object):
         code, _ = self._separate_at_paren(func_m.group('code'))  # refine the match
         return self.build_arglist(func_m.group('args')), code
 
-    def extract_function(self, funcname):
+    def extract_function(self, funcname, *global_stack):
         return function_with_repr(
-            self.extract_function_from_code(*self.extract_function_code(funcname)),
+            self.extract_function_from_code(*itertools.chain(
+                self.extract_function_code(funcname), global_stack)),
             'F<%s>' % (funcname,))
 
     def extract_function_from_code(self, argnames, code, *global_stack):
         local_vars = {}
 
+        start = None
         while True:
-            mobj = re.search(r'function\((?P<args>[^)]*)\)\s*{', code)
+            mobj = re.search(r'function\((?P<args>[^)]*)\)\s*{', code[start:])
             if mobj is None:
                 break
-            start, body_start = mobj.span()
+            start, body_start = ((start or 0) + x for x in mobj.span())
             body, remaining = self._separate_at_paren(code[body_start - 1:])
             name = self._named_object(local_vars, self.extract_function_from_code(
                 [x.strip() for x in mobj.group('args').split(',')],

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',