Merge 48d0994b5b3ef6714dba9510656808883e40df40 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
-        test_id = m.group('id')
+        assert m, '{0!r} should follow URL format'.format(url)
+        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/extractors.py

@@ -371,7 +371,6 @@ from .esri import EsriVideoIE
 from .europa import EuropaIE
 from .expotv import ExpoTVIE
 from .expressen import ExpressenIE
-from .extremetube import ExtremeTubeIE
 from .eyedotv import EyedoTVIE
 from .facebook import (
     FacebookIE,
@@ -562,7 +561,6 @@ from .kaltura import KalturaIE
 from .kankan import KankanIE
 from .karaoketv import KaraoketvIE
 from .karrierevideos import KarriereVideosIE
-from .keezmovies import KeezMoviesIE
 from .ketnet import KetnetIE
 from .khanacademy import (
     KhanAcademyIE,
@@ -713,10 +711,6 @@ from .mlb import (
 )
 from .mnet import MnetIE
 from .moevideo import MoeVideoIE
-from .mofosex import (
-    MofosexIE,
-    MofosexEmbedIE,
-)
 from .mojvideo import MojvideoIE
 from .morningstar import MorningstarIE
 from .motherless import (

youtube_dl/extractor/extremetube.py

@@ -1,50 +0,0 @@
-from __future__ import unicode_literals
-
-from ..utils import str_to_int
-from .keezmovies import KeezMoviesIE
-
-
-class ExtremeTubeIE(KeezMoviesIE):
-    _VALID_URL = r'https?://(?:www\.)?extremetube\.com/(?:[^/]+/)?video/(?P<id>[^/#?&]+)'
-    _TESTS = [{
-        'url': 'http://www.extremetube.com/video/music-video-14-british-euro-brit-european-cumshots-swallow-652431',
-        'md5': '92feaafa4b58e82f261e5419f39c60cb',
-        'info_dict': {
-            'id': 'music-video-14-british-euro-brit-european-cumshots-swallow-652431',
-            'ext': 'mp4',
-            'title': 'Music Video 14 british euro brit european cumshots swallow',
-            'uploader': 'anonim',
-            'view_count': int,
-            'age_limit': 18,
-        }
-    }, {
-        'url': 'http://www.extremetube.com/gay/video/abcde-1234',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.extremetube.com/video/latina-slut-fucked-by-fat-black-dick',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.extremetube.com/video/652431',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        webpage, info = self._extract_info(url)
-
-        if not info['title']:
-            info['title'] = self._search_regex(
-                r'<h1[^>]+title="([^"]+)"[^>]*>', webpage, 'title')
-
-        uploader = self._html_search_regex(
-            r'Uploaded by:\s*</[^>]+>\s*<a[^>]+>(.+?)</a>',
-            webpage, 'uploader', fatal=False)
-        view_count = str_to_int(self._search_regex(
-            r'Views:\s*</[^>]+>\s*<[^>]+>([\d,\.]+)</',
-            webpage, 'view count', fatal=False))
-
-        info.update({
-            'uploader': uploader,
-            'view_count': view_count,
-        })
-
-        return info

youtube_dl/extractor/generic.py

@@ -66,7 +66,6 @@ from .tnaflix import TNAFlixNetworkEmbedIE
 from .drtuber import DrTuberIE
 from .redtube import RedTubeIE
 from .tube8 import Tube8IE
-from .mofosex import MofosexEmbedIE
 from .spankwire import SpankwireIE
 from .youporn import YouPornIE
 from .vimeo import (
@@ -3044,11 +3043,6 @@ class GenericIE(InfoExtractor):
         if tube8_urls:
             return self.playlist_from_matches(tube8_urls, video_id, video_title, ie=Tube8IE.ie_key())
 
-        # Look for embedded Mofosex player
-        mofosex_urls = MofosexEmbedIE._extract_urls(webpage)
-        if mofosex_urls:
-            return self.playlist_from_matches(mofosex_urls, video_id, video_title, ie=MofosexEmbedIE.ie_key())
-
         # Look for embedded Spankwire player
         spankwire_urls = SpankwireIE._extract_urls(webpage)
         if spankwire_urls:

youtube_dl/extractor/keezmovies.py

@@ -1,133 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..aes import aes_decrypt_text
-from ..compat import compat_urllib_parse_unquote
-from ..utils import (
-    determine_ext,
-    ExtractorError,
-    int_or_none,
-    str_to_int,
-    strip_or_none,
-    url_or_none,
-)
-
-
-class KeezMoviesIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?keezmovies\.com/video/(?:(?P<display_id>[^/]+)-)?(?P<id>\d+)'
-    _TESTS = [{
-        'url': 'https://www.keezmovies.com/video/arab-wife-want-it-so-bad-i-see-she-thirsty-and-has-tiny-money-18070681',
-        'md5': '2ac69cdb882055f71d82db4311732a1a',
-        'info_dict': {
-            'id': '18070681',
-            'display_id': 'arab-wife-want-it-so-bad-i-see-she-thirsty-and-has-tiny-money',
-            'ext': 'mp4',
-            'title': 'Arab wife want it so bad I see she thirsty and has tiny money.',
-            'thumbnail': None,
-            'view_count': int,
-            'age_limit': 18,
-        }
-    }, {
-        'url': 'http://www.keezmovies.com/video/18070681',
-        'only_matching': True,
-    }]
-
-    def _extract_info(self, url, fatal=True):
-        mobj = re.match(self._VALID_URL, url)
-        video_id = mobj.group('id')
-        display_id = (mobj.group('display_id')
-                      if 'display_id' in mobj.groupdict()
-                      else None) or mobj.group('id')
-
-        webpage = self._download_webpage(
-            url, display_id, headers={'Cookie': 'age_verified=1'})
-
-        formats = []
-        format_urls = set()
-
-        title = None
-        thumbnail = None
-        duration = None
-        encrypted = False
-
-        def extract_format(format_url, height=None):
-            format_url = url_or_none(format_url)
-            if not format_url or not format_url.startswith(('http', '//')):
-                return
-            if format_url in format_urls:
-                return
-            format_urls.add(format_url)
-            tbr = int_or_none(self._search_regex(
-                r'[/_](\d+)[kK][/_]', format_url, 'tbr', default=None))
-            if not height:
-                height = int_or_none(self._search_regex(
-                    r'[/_](\d+)[pP][/_]', format_url, 'height', default=None))
-            if encrypted:
-                format_url = aes_decrypt_text(
-                    video_url, title, 32).decode('utf-8')
-            formats.append({
-                'url': format_url,
-                'format_id': '%dp' % height if height else None,
-                'height': height,
-                'tbr': tbr,
-            })
-
-        flashvars = self._parse_json(
-            self._search_regex(
-                r'flashvars\s*=\s*({.+?});', webpage,
-                'flashvars', default='{}'),
-            display_id, fatal=False)
-
-        if flashvars:
-            title = flashvars.get('video_title')
-            thumbnail = flashvars.get('image_url')
-            duration = int_or_none(flashvars.get('video_duration'))
-            encrypted = flashvars.get('encrypted') is True
-            for key, value in flashvars.items():
-                mobj = re.search(r'quality_(\d+)[pP]', key)
-                if mobj:
-                    extract_format(value, int(mobj.group(1)))
-            video_url = flashvars.get('video_url')
-            if video_url and determine_ext(video_url, None):
-                extract_format(video_url)
-
-        video_url = self._html_search_regex(
-            r'flashvars\.video_url\s*=\s*(["\'])(?P<url>http.+?)\1',
-            webpage, 'video url', default=None, group='url')
-        if video_url:
-            extract_format(compat_urllib_parse_unquote(video_url))
-
-        if not formats:
-            if 'title="This video is no longer available"' in webpage:
-                raise ExtractorError(
-                    'Video %s is no longer available' % video_id, expected=True)
-
-        try:
-            self._sort_formats(formats)
-        except ExtractorError:
-            if fatal:
-                raise
-
-        if not title:
-            title = self._html_search_regex(
-                r'<h1[^>]*>([^<]+)', webpage, 'title')
-
-        return webpage, {
-            'id': video_id,
-            'display_id': display_id,
-            'title': strip_or_none(title),
-            'thumbnail': thumbnail,
-            'duration': duration,
-            'age_limit': 18,
-            'formats': formats,
-        }
-
-    def _real_extract(self, url):
-        webpage, info = self._extract_info(url, fatal=False)
-        if not info['formats']:
-            return self.url_result(url, 'Generic')
-        info['view_count'] = str_to_int(self._search_regex(
-            r'<b>([\d,.]+)</b> Views?', webpage, 'view count', fatal=False))
-        return info

youtube_dl/extractor/mofosex.py

@@ -1,79 +0,0 @@
-from __future__ import unicode_literals
-
-import re
-
-from .common import InfoExtractor
-from ..utils import (
-    int_or_none,
-    str_to_int,
-    unified_strdate,
-)
-from .keezmovies import KeezMoviesIE
-
-
-class MofosexIE(KeezMoviesIE):
-    _VALID_URL = r'https?://(?:www\.)?mofosex\.com/videos/(?P<id>\d+)/(?P<display_id>[^/?#&.]+)\.html'
-    _TESTS = [{
-        'url': 'http://www.mofosex.com/videos/318131/amateur-teen-playing-and-masturbating-318131.html',
-        'md5': '558fcdafbb63a87c019218d6e49daf8a',
-        'info_dict': {
-            'id': '318131',
-            'display_id': 'amateur-teen-playing-and-masturbating-318131',
-            'ext': 'mp4',
-            'title': 'amateur teen playing and masturbating',
-            'thumbnail': r're:^https?://.*\.jpg$',
-            'upload_date': '20121114',
-            'view_count': int,
-            'like_count': int,
-            'dislike_count': int,
-            'age_limit': 18,
-        }
-    }, {
-        # This video is no longer available
-        'url': 'http://www.mofosex.com/videos/5018/japanese-teen-music-video.html',
-        'only_matching': True,
-    }]
-
-    def _real_extract(self, url):
-        webpage, info = self._extract_info(url)
-
-        view_count = str_to_int(self._search_regex(
-            r'VIEWS:</span>\s*([\d,.]+)', webpage, 'view count', fatal=False))
-        like_count = int_or_none(self._search_regex(
-            r'id=["\']amountLikes["\'][^>]*>(\d+)', webpage,
-            'like count', fatal=False))
-        dislike_count = int_or_none(self._search_regex(
-            r'id=["\']amountDislikes["\'][^>]*>(\d+)', webpage,
-            'like count', fatal=False))
-        upload_date = unified_strdate(self._html_search_regex(
-            r'Added:</span>([^<]+)', webpage, 'upload date', fatal=False))
-
-        info.update({
-            'view_count': view_count,
-            'like_count': like_count,
-            'dislike_count': dislike_count,
-            'upload_date': upload_date,
-            'thumbnail': self._og_search_thumbnail(webpage),
-        })
-
-        return info
-
-
-class MofosexEmbedIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?mofosex\.com/embed/?\?.*?\bvideoid=(?P<id>\d+)'
-    _TESTS = [{
-        'url': 'https://www.mofosex.com/embed/?videoid=318131&referrer=KM',
-        'only_matching': True,
-    }]
-
-    @staticmethod
-    def _extract_urls(webpage):
-        return re.findall(
-            r'<iframe[^>]+\bsrc=["\']((?:https?:)?//(?:www\.)?mofosex\.com/embed/?\?.*?\bvideoid=\d+)',
-            webpage)
-
-    def _real_extract(self, url):
-        video_id = self._match_id(url)
-        return self.url_result(
-            'http://www.mofosex.com/videos/{0}/{0}.html'.format(video_id),
-            ie=MofosexIE.ie_key(), video_id=video_id)

youtube_dl/extractor/tube8.py

@@ -3,32 +3,33 @@ from __future__ import unicode_literals
 import re
 
 from ..utils import (
+    determine_ext,
+    ExtractorError,
     int_or_none,
     str_to_int,
+    url_or_none,
 )
-from .keezmovies import KeezMoviesIE
+from .common import InfoExtractor
+from ..aes import aes_decrypt_text
+from ..compat import compat_urllib_parse_unquote
 
 
-class Tube8IE(KeezMoviesIE):
+class Tube8IE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?tube8\.com/(?:[^/]+/)+(?P<display_id>[^/]+)/(?P<id>\d+)'
     _TESTS = [{
-        'url': 'http://www.tube8.com/teen/kasia-music-video/229795/',
-        'md5': '65e20c48e6abff62ed0c3965fff13a39',
+        'url': 'https://www.tube8.com/erotic/playtime/81807731/',
+        'md5': 'fefa69ff76debaa63aa59374bfc51c95',
         'info_dict': {
-            'id': '229795',
-            'display_id': 'kasia-music-video',
+            'id': '81807731',
+            'display_id': 'playtime',
             'ext': 'mp4',
-            'description': 'hot teen Kasia grinding',
-            'uploader': 'unknown',
-            'title': 'Kasia music video',
+            'uploader': 'kikkijay-ph',
+            'title': 'Playtime',
             'age_limit': 18,
-            'duration': 230,
-            'categories': ['Teen'],
-            'tags': ['dancing'],
+            'duration': 988,
+            'categories': ['Erotic'],
+            'tags': ['adult toys', 'big boobs', 'butt', 'masturbate'],
         },
-    }, {
-        'url': 'http://www.tube8.com/shemale/teen/blonde-cd-gets-kidnapped-by-two-blacks-and-punished-for-being-a-slutty-girl/19569151/',
-        'only_matching': True,
     }]
 
     @staticmethod
@@ -38,49 +39,121 @@ class Tube8IE(KeezMoviesIE):
             webpage)
 
     def _real_extract(self, url):
-        webpage, info = self._extract_info(url)
+        mobj = re.match(self._VALID_URL, url)
+        video_id = mobj.group('id')
+        display_id = mobj.group('display_id')
 
-        if not info['title']:
-            info['title'] = self._html_search_regex(
-                r'videoTitle\s*=\s*"([^"]+)', webpage, 'title')
+        webpage = self._download_webpage(
+            url, display_id, headers={'Cookie': 'age_verified=1'})
 
-        description = self._html_search_regex(
-            r'(?s)Description:</dt>\s*<dd>(.+?)</dd>', webpage, 'description', fatal=False)
-        uploader = self._html_search_regex(
-            r'<span class="username">\s*(.+?)\s*<',
-            webpage, 'uploader', fatal=False)
+        formats = []
+        format_urls = set()
+
+        title = None
+        thumbnail = None
+        duration = None
+        encrypted = False
+
+        def extract_format(format_url, height=None):
+            format_url = url_or_none(format_url)
+            if not format_url or not format_url.startswith(('http', '//')):
+                return
+            if format_url in format_urls:
+                return
+            format_urls.add(format_url)
+            tbr = int_or_none(self._search_regex(
+                r'[/_](\d+)[kK][/_]', format_url, 'tbr', default=None))
+            if not height:
+                height = int_or_none(self._search_regex(
+                    r'[/_](\d+)[pP][/_]', format_url, 'height', default=None))
+            if encrypted:
+                format_url = aes_decrypt_text(
+                    video_url, title, 32).decode('utf-8')
+            formats.append({
+                'url': format_url,
+                'format_id': '%dp' % height if height else None,
+                'height': height,
+                'tbr': tbr,
+            })
+
+        flashvars = self._parse_json(
+            self._search_regex(
+                r'flashvars\s*=\s*({.+?});', webpage,
+                'flashvars', default='{}'),
+            display_id, fatal=False)
+
+        if flashvars:
+            title = flashvars.get('video_title')
+            thumbnail = flashvars.get('image_url')
+            duration = int_or_none(flashvars.get('video_duration'))
+            encrypted = flashvars.get('encrypted') is True
+            uploader = flashvars.get('sponsor')
+            for key, value in flashvars.items():
+                mobj = re.search(r'quality_(\d+)[pP]', key)
+                if mobj:
+                    extract_format(value, int(mobj.group(1)))
+            video_url = flashvars.get('video_url')
+            for key, value in flashvars.items():
+                mobj = re.search(r'quality_(\d+)[pP]', key)
+                if mobj:
+                    extract_format(value, int(mobj.group(1)))
+            video_url = flashvars.get('video_url')
+            if video_url and determine_ext(video_url, None):
+                extract_format(video_url)
+
+        video_url = self._html_search_regex(
+            r'flashvars\.video_url\s*=\s*(["\'])(?P<url>http.+?)\1',
+            webpage, 'video url', default=None, group='url')
+        if video_url:
+            extract_format(compat_urllib_parse_unquote(video_url))
+
+        if not formats:
+            if 'title="This video is no longer available"' in webpage:
+                raise ExtractorError(
+                    'Video %s is no longer available' % video_id, expected=True)
+
+        self._sort_formats(formats)
+
+        if not title:
+            title = self._html_search_regex([
+                r'<h1[^>]*>([^<]+)'
+                r'videoTitle\s*=\s*"([^"]+)'], webpage, 'title', default=display_id.capitalize())
 
         like_count = int_or_none(self._search_regex(
-            r'rupVar\s*=\s*"(\d+)"', webpage, 'like count', fatal=False))
+            r'rupVar\s*=\s*(\d+);', webpage, 'like count', fatal=False))
         dislike_count = int_or_none(self._search_regex(
-            r'rdownVar\s*=\s*"(\d+)"', webpage, 'dislike count', fatal=False))
+            r'rdownVar\s*=\s*(\d+);', webpage, 'dislike count', fatal=False))
         view_count = str_to_int(self._search_regex(
             r'Views:\s*</dt>\s*<dd>([\d,\.]+)',
             webpage, 'view count', fatal=False))
         comment_count = str_to_int(self._search_regex(
-            r'<span id="allCommentsCount">(\d+)</span>',
+            r'<span id="allCommentsCount">\((\d+)\)</span>',
             webpage, 'comment count', fatal=False))
 
         category = self._search_regex(
-            r'Category:\s*</dt>\s*<dd>\s*<a[^>]+href=[^>]+>([^<]+)',
+            r'videoCategoryByName\s*=\s*"([^"]+)";',
             webpage, 'category', fatal=False)
         categories = [category] if category else None
 
         tags_str = self._search_regex(
-            r'(?s)Tags:\s*</dt>\s*<dd>(.+?)</(?!a)',
+            r"(?s)<li\s+class\s*=\s*'video-tag'\s+data-esp-node\s*=\s*'tag'>(.*?)</div>",
             webpage, 'tags', fatal=False)
         tags = [t for t in re.findall(
             r'<a[^>]+href=[^>]+>([^<]+)', tags_str)] if tags_str else None
 
-        info.update({
-            'description': description,
+        return {
             'uploader': uploader,
+            'id': video_id,
+            'display_id': display_id,
+            'title': title,
+            'thumbnail': thumbnail,
+            'duration': duration,
+            'age_limit': 18,
+            'formats': formats,
             'view_count': view_count,
             'like_count': like_count,
             'dislike_count': dislike_count,
             'comment_count': comment_count,
             'categories': categories,
             'tags': tags,
-        })
-
-        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'/(?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$',
-        r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$',
+        r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/(?:tv-)?player',
+        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$',
+        r'\b(?P<id>vfl[a-zA-Z0-9_-]{6,})\b.*?\.js$',
     )
     _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(cls, player_url):
-        for player_re in cls._PLAYER_INFO_RE:
-            id_m = re.search(player_re, player_url)
-            if id_m:
-                break
-        else:
-            raise ExtractorError('Cannot identify player %r' % player_url)
-        return id_m.group('id')
+    def _extract_player_info(self, player_url):
+        try:
+            return self._search_regex(
+                self._PLAYER_INFO_RE, player_url, 'player info', group='id')
+        except ExtractorError as e:
+            raise ExtractorError(
+                'Cannot identify player %r' % (player_url,), cause=e)
 
     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)
-        if var_ay:
-            func_code = (func_code[0], ';\n'.join((var_ay, func_code[1])))
+        func_code = self._extract_sig_fn(jsi, func_name)
 
         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,
-            ('captions', 'playerCaptionsTracklistRenderer', T(dict)))
+            (player_response, api_player_response),
+            (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(
-                        ('' if x in (None, JS_Undefined) else _js_toString(x))
-                        for x in obj)
+                    return (',' if len(argvals) == 0 or argvals[0] in (None, JS_Undefined)
+                            else argvals[0]).join(
+                                ('' 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(',')],