Merge 11fc1e228aeed255292c2994b0cc779de06d597c 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/cbs.py

@@ -1,3 +1,4 @@
+# coding: utf-8
 from __future__ import unicode_literals
 
 from .theplatform import ThePlatformFeedIE
@@ -8,6 +9,7 @@ from ..utils import (
     xpath_element,
     xpath_text,
     update_url_query,
+    url_or_none,
 )
 
 
@@ -25,68 +27,27 @@ class CBSBaseIE(ThePlatformFeedIE):
                     })
         return subtitles
 
-
-class CBSIE(CBSBaseIE):
-    _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:(?:cbs|paramountplus)\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)'
-
-    _TESTS = [{
-        'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
-        'info_dict': {
-            'id': '_u7W953k6la293J7EPTd9oHkSPs6Xn6_',
-            'ext': 'mp4',
-            'title': 'Connect Chat feat. Garth Brooks',
-            'description': 'Connect with country music singer Garth Brooks, as he chats with fans on Wednesday November 27, 2013. Be sure to tune in to Garth Brooks: Live from Las Vegas, Friday November 29, at 9/8c on CBS!',
-            'duration': 1495,
-            'timestamp': 1385585425,
-            'upload_date': '20131127',
-            'uploader': 'CBSI-NEW',
-        },
-        'params': {
-            # m3u8 download
-            'skip_download': True,
-        },
-        '_skip': 'Blocked outside the US',
-    }, {
-        'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
-        'only_matching': True,
-    }, {
-        'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
-        'only_matching': True,
-    }, {
-        'url': 'https://www.paramountplus.com/shows/all-rise/video/QmR1WhNkh1a_IrdHZrbcRklm176X_rVc/all-rise-space/',
-        'only_matching': True,
-    }]
-
-    def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
-        items_data = self._download_xml(
-            'http://can.cbs.com/thunder/player/videoPlayerService.php',
-            content_id, query={'partner': site, 'contentId': content_id})
-        video_data = xpath_element(items_data, './/item')
-        title = xpath_text(video_data, 'videoTitle', 'title', True)
+    def _extract_common_video_info(self, content_id, asset_types, mpx_acc, extra_info):
         tp_path = 'dJ5BDC/media/guid/%d/%s' % (mpx_acc, content_id)
         tp_release_url = 'http://link.theplatform.com/s/' + tp_path
+        info = self._extract_theplatform_metadata(tp_path, content_id)
 
-        asset_types = []
-        subtitles = {}
-        formats = []
+        formats, subtitles = [], {}
         last_e = None
-        for item in items_data.findall('.//item'):
-            asset_type = xpath_text(item, 'assetType')
-            if not asset_type or asset_type in asset_types or 'HLS_FPS' in asset_type or 'DASH_CENC' in asset_type:
-                continue
-            asset_types.append(asset_type)
-            query = {
-                'mbr': 'true',
-                'assetTypes': asset_type,
-            }
-            if asset_type.startswith('HLS') or asset_type in ('OnceURL', 'StreamPack'):
-                query['formats'] = 'MPEG4,M3U'
-            elif asset_type in ('RTMP', 'WIFI', '3G'):
-                query['formats'] = 'MPEG4,FLV'
+        for asset_type, query in asset_types.items():
             try:
                 tp_formats, tp_subtitles = self._extract_theplatform_smil(
                     update_url_query(tp_release_url, query), content_id,
                     'Downloading %s SMIL data' % asset_type)
+            except ExtractorError as e:
+                last_e = e
+                if asset_type != 'fallback':
+                    continue
+                query['formats'] = ''  # blank query to check if expired
+            try:
+                tp_formats, tp_subtitles = self._extract_theplatform_smil(
+                    update_url_query(tp_release_url, query), content_id,
+                    'Downloading %s SMIL data, trying again with another format' % asset_type)
             except ExtractorError as e:
                 last_e = e
                 continue
@@ -96,20 +57,117 @@ class CBSIE(CBSBaseIE):
             raise last_e
         self._sort_formats(formats)
 
-        info = self._extract_theplatform_metadata(tp_path, content_id)
-        info.update({
+        extra_info.update({
             'id': content_id,
+            'formats': formats,
+            'subtitles': subtitles,
+        })
+        info.update({k: v for k, v in extra_info.items() if v is not None})
+        return info
+
+    def _extract_video_info(self, *args, **kwargs):
+        # Extract assets + metadata and call _extract_common_video_info
+        raise NotImplementedError('This method must be implemented by subclasses')
+
+    def _real_extract(self, url):
+        return self._extract_video_info(self._match_id(url))
+
+
+class CBSIE(CBSBaseIE):
+    _VALID_URL = r'''(?x)
+        (?:
+            cbs:|
+            https?://(?:www\.)?(?:
+                cbs\.com/(?:shows|movies)/(?:video|[^/]+/video|[^/]+)/|
+                colbertlateshow\.com/(?:video|podcasts)/)
+        )(?P<id>[\w-]+)'''
+
+    # All tests are blocked outside US
+    _TESTS = [{
+        'url': 'https://www.cbs.com/shows/video/xrUyNLtl9wd8D_RWWAg9NU2F_V6QpB3R/',
+        'info_dict': {
+            'id': 'xrUyNLtl9wd8D_RWWAg9NU2F_V6QpB3R',
+            'ext': 'mp4',
+            'title': 'Tough As Nails - Dreams Never Die',
+            'description': 'md5:a3535a62531cdd52b0364248a2c1ae33',
+            'duration': 2588,
+            'timestamp': 1639015200,
+            'upload_date': '20211209',
+            'uploader': 'CBSI-NEW',
+        },
+        'params': {
+            # m3u8 download
+            'skip_download': True,
+        },
+        '_skip': 'Blocked outside the US',
+    }, {
+        'url': 'https://www.cbs.com/shows/video/sZH1MGgomIosZgxGJ1l263MFq16oMtW1/',
+        'info_dict': {
+            'id': 'sZH1MGgomIosZgxGJ1l263MFq16oMtW1',
+            'title': 'The Late Show - 3/16/22 (Michael Buble, Rose Matafeo)',
+            'timestamp': 1647488100,
+            'description': 'md5:d0e6ec23c544b7fa8e39a8e6844d2439',
+            'uploader': 'CBSI-NEW',
+            'upload_date': '20220317',
+        },
+        'params': {
+            'ignore_no_formats_error': True,
+            'skip_download': True,
+        },
+        'expected_warnings': [
+            'This content expired on', 'No video formats found', 'Requested format is not available'],
+    }, {
+        'url': 'http://colbertlateshow.com/video/8GmB0oY0McANFvp2aEffk9jZZZ2YyXxy/the-colbeard/',
+        'only_matching': True,
+    }, {
+        'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/',
+        'only_matching': True,
+    }, {
+        # old format
+        'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/',
+        'only_matching': True,
+    }]
+
+    def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517):
+        items_data = self._download_xml(
+            'https://can.cbs.com/thunder/player/videoPlayerService.php',
+            content_id, query={'partner': site, 'contentId': content_id})
+        video_data = xpath_element(items_data, './/item')
+        title = xpath_text(video_data, 'videoTitle', 'title') or xpath_text(video_data, 'videotitle', 'title')
+
+        asset_types = {}
+        has_drm = False
+        for item in items_data.findall('.//item'):
+            asset_type = xpath_text(item, 'assetType')
+            query = {
+                'mbr': 'true',
+                'assetTypes': asset_type,
+            }
+            if not asset_type:
+                # fallback for content_ids that videoPlayerService doesn't return anything for
+                asset_type = 'fallback'
+                query['formats'] = 'M3U+none,MPEG4,M3U+appleHlsEncryption,MP3'
+                del query['assetTypes']
+            if asset_type in asset_types:
+                continue
+            elif any(excluded in asset_type for excluded in ('HLS_FPS', 'DASH_CENC', 'OnceURL')):
+                if 'DASH_CENC' in asset_type:
+                    has_drm = True
+                continue
+            if asset_type.startswith('HLS') or 'StreamPack' in asset_type:
+                query['formats'] = 'MPEG4,M3U'
+            elif asset_type in ('RTMP', 'WIFI', '3G'):
+                query['formats'] = 'MPEG4,FLV'
+            asset_types[asset_type] = query
+
+        if not asset_types and has_drm:
+            raise ExtractorError('Only DRM formats found', video_id=content_id, expected=True)
+
+        return self._extract_common_video_info(content_id, asset_types, mpx_acc, extra_info={
             'title': title,
             'series': xpath_text(video_data, 'seriesTitle'),
             'season_number': int_or_none(xpath_text(video_data, 'seasonNumber')),
             'episode_number': int_or_none(xpath_text(video_data, 'episodeNumber')),
             'duration': int_or_none(xpath_text(video_data, 'videoLength'), 1000),
-            'thumbnail': xpath_text(video_data, 'previewImageURL'),
-            'formats': formats,
-            'subtitles': subtitles,
+            'thumbnail': url_or_none(xpath_text(video_data, 'previewImageURL')),
         })
-        return info
-
-    def _real_extract(self, url):
-        content_id = self._match_id(url)
-        return self._extract_video_info(content_id)

youtube_dl/extractor/extractors.py

@@ -917,6 +917,10 @@ from .palcomp3 import (
     PalcoMP3VideoIE,
 )
 from .pandoratv import PandoraTVIE
+from .paramountplus import (
+     ParamountPlusIE,
+     ParamountPlusSeriesIE,
+)
 from .parliamentliveuk import ParliamentLiveUKIE
 from .patreon import PatreonIE
 from .pbs import PBSIE

youtube_dl/extractor/paramountplus.py

@@ -0,0 +1,159 @@
+# coding: utf-8
+from __future__ import unicode_literals
+import itertools
+
+from .common import InfoExtractor
+from .cbs import CBSBaseIE
+from ..utils import (
+    int_or_none,
+    url_or_none,
+)
+
+
+class ParamountPlusIE(CBSBaseIE):
+    _VALID_URL = r'''(?x)
+        (?:
+            paramountplus:|
+            https?://(?:www\.)?(?:
+                paramountplus\.com/(?:shows|movies)/(?:video|[^/]+/video|[^/]+)/
+        )(?P<id>[\w-]+))'''
+
+    # All tests are blocked outside US
+    _TESTS = [{
+        'url': 'https://www.paramountplus.com/shows/video/Oe44g5_NrlgiZE3aQVONleD6vXc8kP0k/',
+        'info_dict': {
+            'id': 'Oe44g5_NrlgiZE3aQVONleD6vXc8kP0k',
+            'ext': 'mp4',
+            'title': 'CatDog - Climb Every CatDog/The Canine Mutiny',
+            'description': 'md5:7ac835000645a69933df226940e3c859',
+            'duration': 1426,
+            'timestamp': 920264400,
+            'upload_date': '19990301',
+            'uploader': 'CBSI-NEW',
+        },
+        'params': {
+            'skip_download': 'm3u8',
+        },
+    }, {
+        'url': 'https://www.paramountplus.com/shows/video/6hSWYWRrR9EUTz7IEe5fJKBhYvSUfexd/',
+        'info_dict': {
+            'id': '6hSWYWRrR9EUTz7IEe5fJKBhYvSUfexd',
+            'ext': 'mp4',
+            'title': '7/23/21 WEEK IN REVIEW (Rep. Jahana Hayes/Howard Fineman/Sen. Michael Bennet/Sheera Frenkel & Cecilia Kang)',
+            'description': 'md5:f4adcea3e8b106192022e121f1565bae',
+            'duration': 2506,
+            'timestamp': 1627063200,
+            'upload_date': '20210723',
+            'uploader': 'CBSI-NEW',
+        },
+        'params': {
+            'skip_download': 'm3u8',
+        },
+    }, {
+        'url': 'https://www.paramountplus.com/movies/video/vM2vm0kE6vsS2U41VhMRKTOVHyQAr6pC/',
+        'info_dict': {
+            'id': 'vM2vm0kE6vsS2U41VhMRKTOVHyQAr6pC',
+            'ext': 'mp4',
+            'title': 'Daddy\'s Home',
+            'upload_date': '20151225',
+            'description': 'md5:9a6300c504d5e12000e8707f20c54745',
+            'uploader': 'CBSI-NEW',
+            'timestamp': 1451030400,
+        },
+        'params': {
+            'skip_download': 'm3u8',
+            'format': 'bestvideo',
+        },
+        'expected_warnings': ['Ignoring subtitle tracks'],  # TODO: Investigate this
+    }, {
+        'url': 'https://www.paramountplus.com/movies/video/5EKDXPOzdVf9voUqW6oRuocyAEeJGbEc/',
+        'info_dict': {
+            'id': '5EKDXPOzdVf9voUqW6oRuocyAEeJGbEc',
+            'ext': 'mp4',
+            'uploader': 'CBSI-NEW',
+            'description': 'md5:bc7b6fea84ba631ef77a9bda9f2ff911',
+            'timestamp': 1577865600,
+            'title': 'Sonic the Hedgehog',
+            'upload_date': '20200101',
+        },
+        'params': {
+            'skip_download': 'm3u8',
+            'format': 'bestvideo',
+        },
+        'expected_warnings': ['Ignoring subtitle tracks'],
+    }, {
+        'url': 'https://www.paramountplus.com/shows/the-real-world/video/mOVeHeL9ub9yWdyzSZFYz8Uj4ZBkVzQg/the-real-world-reunion/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.paramountplus.com/shows/video/mOVeHeL9ub9yWdyzSZFYz8Uj4ZBkVzQg/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.paramountplus.com/movies/video/W0VyStQqUnqKzJkrpSAIARuCc9YuYGNy/',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.paramountplus.com/movies/paw-patrol-the-movie/W0VyStQqUnqKzJkrpSAIARuCc9YuYGNy/',
+        'only_matching': True,
+    }, {
+        # old format
+        'url': 'https://www.paramountplus.com/shows/catdog/video/Oe44g5_NrlgiZE3aQVONleD6vXc8kP0k/catdog-climb-every-catdog-the-canine-mutiny/',
+        'only_matching': True,
+    }]
+
+    def _extract_video_info(self, content_id, mpx_acc=2198311517):
+        items_data = self._download_json(
+            'https://www.paramountplus.com/apps-api/v2.0/androidtv/video/cid/%s.json' % content_id,
+            content_id, query={'locale': 'en-us', 'at': 'ABCqWNNSwhIqINWIIAG+DFzcFUvF8/vcN6cNyXFFfNzWAIvXuoVgX+fK4naOC7V8MLI='}, headers=self.geo_verification_headers())
+
+        asset_types = {
+            item.get('assetType'): {
+                'format': 'SMIL',
+                'formats': 'MPEG4,M3U',
+            } for item in items_data['itemList']
+        }
+        item = items_data['itemList'][-1]
+        return self._extract_common_video_info(content_id, asset_types, mpx_acc, extra_info={
+            'title': item.get('title'),
+            'series': item.get('seriesTitle'),
+            'season_number': int_or_none(item.get('seasonNum')),
+            'episode_number': int_or_none(item.get('episodeNum')),
+            'duration': int_or_none(item.get('duration')),
+            'thumbnail': url_or_none(item.get('thumbnail')),
+        })
+
+
+class ParamountPlusSeriesIE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.)?paramountplus\.com/shows/(?P<id>[a-zA-Z0-9-_]+)/?(?:[#?]|$)'
+    _TESTS = [{
+        'url': 'https://www.paramountplus.com/shows/drake-josh',
+        'playlist_mincount': 45,
+        'info_dict': {
+            'id': 'drake-josh',
+        }
+    }, {
+        'url': 'https://www.paramountplus.com/shows/hawaii_five_0/',
+        'playlist_mincount': 240,
+        'info_dict': {
+            'id': 'hawaii_five_0',
+        }
+    }, {
+        'url': 'https://www.paramountplus.com/shows/spongebob-squarepants/',
+        'playlist_mincount': 248,
+        'info_dict': {
+            'id': 'spongebob-squarepants',
+        }
+    }]
+
+    def _entries(self, show_name):
+        for page in itertools.count():
+            show_json = self._download_json(
+                'https://www.paramountplus.com/shows/{0}/xhr/episodes/page/{1}/size/50/xs/0/season/0'.format(show_name, page), show_name)
+            if not show_json.get('success'):
+                return
+            for episode in show_json['result']['data']:
+                yield self.url_result(
+                    'https://www.paramountplus.com%s' % episode['url'],
+                    ie=ParamountPlusIE.ie_key(), video_id=episode['content_id'])
+
+    def _real_extract(self, url):
+        show_name = self._match_id(url)
+        return self.playlist_result(self._entries(show_name), playlist_id=show_name)

youtube_dl/extractor/theplatform.py

@@ -20,6 +20,7 @@ from ..utils import (
     float_or_none,
     int_or_none,
     sanitized_Request,
+    try_get,
     unsmuggle_url,
     update_url_query,
     xpath_with_ns,
@@ -34,6 +35,15 @@ _x = lambda p: xpath_with_ns(p, {'smil': default_ns})
 class ThePlatformBaseIE(OnceIE):
     _TP_TLD = 'com'
 
+    @classmethod
+    def _match_valid_url(cls, url):
+        # This does not use has/getattr intentionally - we want to know whether
+        # we have cached the regexp for *this* class, whereas getattr would also
+        # match the superclass
+        if '_VALID_URL_RE' not in cls.__dict__:
+            cls._VALID_URL_RE = re.compile(cls._VALID_URL)
+        return cls._VALID_URL_RE.match(url)
+
     def _extract_theplatform_smil(self, smil_url, video_id, note='Downloading SMIL data'):
         meta = self._download_xml(
             smil_url, video_id, note=note, query={'format': 'SMIL'},
@@ -238,7 +248,7 @@ class ThePlatformIE(ThePlatformBaseIE, AdobePassIE):
             'countries': smuggled_data.get('geo_countries'),
         })
 
-        mobj = re.match(self._VALID_URL, url)
+        mobj = self._match_valid_url(url)
         provider_id = mobj.group('provider_id')
         video_id = mobj.group('id')
 
@@ -338,6 +348,7 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
             'categories': ['MSNBC/Issues/Democrats', 'MSNBC/Issues/Elections/Election 2016'],
             'uploader': 'NBCU-NEWS',
         },
+        'expected_warnings': ('Empty metadata',),
     }, {
         'url': 'http://feed.theplatform.com/f/2E2eJC/nnd_NBCNews?byGuid=nn_netcast_180306.Copy.01',
         'only_matching': True,
@@ -345,7 +356,11 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
 
     def _extract_feed_info(self, provider_id, feed_id, filter_query, video_id, custom_fields=None, asset_types_query={}, account_id=None):
         real_url = self._URL_TEMPLATE % (self.http_scheme(), provider_id, feed_id, filter_query)
-        entry = self._download_json(real_url, video_id)['entries'][0]
+        entry = self._download_json(real_url, video_id)
+        entry = try_get(entry, lambda x: x['entries'][0], dict)
+        if not entry:
+            self.report_warning('Empty metadata', video_id)
+            return None
         main_smil_url = 'http://link.theplatform.com/s/%s/media/guid/%d/%s' % (provider_id, account_id, entry['guid']) if account_id else entry.get('plmedia$publicUrl')
 
         formats = []
@@ -404,7 +419,7 @@ class ThePlatformFeedIE(ThePlatformBaseIE):
         return ret
 
     def _real_extract(self, url):
-        mobj = re.match(self._VALID_URL, url)
+        mobj = self._match_valid_url(url)
 
         video_id = mobj.group('id')
         provider_id = mobj.group('provider_id')

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(',')],