Merge f6c4fc246a41a682adcf0aa3799fede1f0aef708 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/bitchute.py

@@ -1,142 +1,388 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import itertools
+import functools
 import re
 
 from .common import InfoExtractor
 from ..utils import (
+    clean_html,
+    extract_attributes,
+    ExtractorError,
+    get_element_by_class,
+    get_element_by_id,
+    HEADRequest,
+    int_or_none,
+    merge_dicts,
+    OnDemandPagedList,
     orderedSet,
-    unified_strdate,
+    parse_count,
+    parse_duration,
+    traverse_obj,
+    unified_timestamp,
     urlencode_postdata,
+    urljoin,
 )
 
 
-class BitChuteIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?bitchute\.com/(?:video|embed|torrent/[^/]+)/(?P<id>[^/?#&]+)'
+def get_elements_html_by_class(class_name, html):
+    for m in re.finditer(
+            r'(?P<elt><(?P<tag>[\w-]+)(?:\s[^>]+)?>)', html):
+        if get_element_by_class(class_name, '%sa</%s>' % (m.group('elt'), m.group('tag'))):
+            m = re.match(r'%s(?:[\s\S]*?</%s>)?' % (m.group('elt'), m.group('tag')), html[m.start():])
+            yield m.group(0)
+
+
+def get_element_html_by_class(class_name, html):
+    return next(get_elements_html_by_class(class_name, html), None)
+
+
+class BitChuteBaseIE(InfoExtractor):
+    _USER_AGENT = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.57 Safari/537.36'
+    _HEADERS = {
+        'User-Agent': _USER_AGENT,
+        'Referer': 'https://www.bitchute.com/',
+    }
+    _GEO_BYPASS = False
+
+
+class BitChuteIE(BitChuteBaseIE):
+    _VALID_URL = r'https?://(?:(?:www|old)\.)?bitchute\.com/(?:video|embed|torrent/[^/]+)/(?P<id>[^/?#&]+)'
+    _EMBED_REGEX = [r'<(?:script|iframe)[^>]+\bsrc=(["\'])(?P<url>{0})'.format(_VALID_URL)]
     _TESTS = [{
-        'url': 'https://www.bitchute.com/video/szoMrox2JEI/',
-        'md5': '66c4a70e6bfc40dcb6be3eb1d74939eb',
+        'url': 'https://www.bitchute.com/video/UGlrF9o9b-Q/',
+        'md5': '7e427d7ed7af5a75b5855705ec750e2b',
         'info_dict': {
-            'id': 'szoMrox2JEI',
+            'id': 'UGlrF9o9b-Q',
             'ext': 'mp4',
-            'title': 'Fuck bitches get money',
-            'description': 'md5:3f21f6fb5b1d17c3dee9cf6b5fe60b3a',
+            'title': 'This is the first video on #BitChute !',
+            'timestamp': 1483425420,
+            'upload_date': '20170103',
+            'description': 'md5:a0337e7b1fe39e32336974af8173a034',
             'thumbnail': r're:^https?://.*\.jpg$',
-            'uploader': 'Victoria X Rave',
-            'upload_date': '20170813',
+            'uploader': 'BitChute',
+            'uploader_url': 'https://www.bitchute.com/profile/I5NgtHZn9vPj/',
+            'channel': 'BitChute',
+            'channel_url': 'https://www.bitchute.com/channel/bitchute/',
+            'age_limit': None,
         },
+    }, {
+        # test case: video with different channel and uploader
+        'url': 'https://www.bitchute.com/video/Yti_j9A-UZ4/',
+        'md5': 'f10e6a8e787766235946d0868703f1d0',
+        'info_dict': {
+            'id': 'Yti_j9A-UZ4',
+            'ext': 'mp4',
+            'title': 'Israel at War | Full Measure',
+            'description': 'md5:38cf7bc6f42da1a877835539111c69ef',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'sharylattkisson',
+            'timestamp': 1699296060,
+            'upload_date': '20231106',
+            'uploader_url': 'https://www.bitchute.com/profile/9K0kUWA9zmd9/',
+            'channel': 'Full Measure with Sharyl Attkisson',
+            'channel_url': 'https://www.bitchute.com/channel/sharylattkisson/',
+        },
+    }, {
+        # NSFW (#24419)
+        'url': 'https://www.bitchute.com/video/wrTrKp7PmFZC/',
+        'md5': '4ef880ce8d24e322172d41a0cf6f8096',
+        'info_dict': {
+            'id': 'wrTrKp7PmFZC',
+            'ext': 'mp4',
+            'title': "You Can't Stop Progress | Episode 2",
+            'timestamp': 1541476920,
+            'upload_date': '20181106',
+            'description': 'md5:f191b538a2c4d8f57540141a6bfd7eb0',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader': 'lporiginalg',
+            'channel': "You Can't Stop Progress",
+            'age_limit': 18,
+            'channel_url': 'https://www.bitchute.com/channel/ycsp/',
+        },
+    }, {
+        # video not downloadable in browser, but we can recover it
+        'url': 'https://www.bitchute.com/video/2s6B3nZjAk7R/',
+        'md5': '05c12397d5354bf24494885b08d24ed1',
+        'info_dict': {
+            'id': '2s6B3nZjAk7R',
+            'ext': 'mp4',
+            'filesize': 71537926,
+            'title': 'STYXHEXENHAMMER666 - Election Fraud, Clinton 2020, EU Armies, and Gun Control',
+            'description': 'md5:228ee93bd840a24938f536aeac9cf749',
+            'timestamp': 1542130260,
+            'upload_date': '20181113',
+            'uploader': 'BitChute',
+            'thumbnail': r're:^https?://.*\.jpg$',
+            'uploader_url': 'https://www.bitchute.com/profile/I5NgtHZn9vPj/',
+            'channel': 'BitChute',
+            'channel_url': 'https://www.bitchute.com/channel/bitchute/',
+        },
+        'params': {'check_formats': None},
+    }, {
+        # restricted video
+        'url': 'https://www.bitchute.com/video/WEnQU7XGcTdl/',
+        'info_dict': {
+            'id': 'WEnQU7XGcTdl',
+            'ext': 'mp4',
+            'title': 'Impartial Truth - Ein Letzter Appell an die Vernunft',
+            'description': 'md5:32ebe076beaa571b253788465a69c63a',
+            'timestamp': 1609918800.0,
+            'upload_date': '20210106',
+            'uploader': 'Freier_Mann',
+        },
+        'params': {'skip_download': True},
+        'skip': 'Georestricted in DE',
     }, {
         'url': 'https://www.bitchute.com/embed/lbb5G1hjPhw/',
         'only_matching': True,
     }, {
         'url': 'https://www.bitchute.com/torrent/Zee5BE49045h/szoMrox2JEI.webtorrent',
         'only_matching': True,
+    }, {
+        'url': 'https://old.bitchute.com/video/UGlrF9o9b-Q/',
+        'only_matching': True,
     }]
 
+    @staticmethod
+    def extract_urls(webpage):
+        urls = re.finditer(
+            r'''<(?:script|iframe)\s[^>]*\b(?<!-)src\s*=\s*["'](?P<url>%s)''' % (BitChuteIE._VALID_URL, ),
+            webpage)
+        return (mobj.group('url') for mobj in urls)
+
+    def _check_format(self, video_url, video_id):
+        urls = orderedSet(
+            re.sub(r'(^https?://)(seed\d+)(?=\.bitchute\.com)', r'\g<1>{0}'.format(host), video_url)
+            for host in (r'\g<2>', 'seed122', 'seed125', 'seed126', 'seed128',
+                         'seed132', 'seed150', 'seed151', 'seed152', 'seed153',
+                         'seed167', 'seed171', 'seed177', 'seed305', 'seed307',
+                         'seedp29xb', 'zb10-7gsop1v78'))
+        for url in urls:
+            try:
+                response = self._request_webpage(
+                    HEADRequest(url), video_id=video_id, note='Checking {0}'.format(url), headers=self._HEADERS)
+            except ExtractorError as e:
+                self.to_screen('{0}: URL is invalid, skipping: {1}'.format(video_id, e.cause))
+                continue
+            return {
+                'url': url,
+                'filesize': int_or_none(response.headers.get('Content-Length')),
+            }
+
+    def _raise_if_restricted(self, webpage):
+        page_title = clean_html(get_element_by_class('page-title', webpage)) or ''
+        if re.match(r'(?:Channel|Video) Restricted$', page_title):
+            reason = clean_html(get_element_by_id('page-detail', webpage)) or page_title
+            self.raise_geo_restricted(reason)
+
+    @staticmethod
+    def _make_url(html):
+        path = extract_attributes(get_element_html_by_class('spa', html) or '').get('href')
+        return urljoin('https://www.bitchute.com', path)
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        webpage = self._download_webpage(
-            'https://www.bitchute.com/video/%s' % video_id, video_id, headers={
-                'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_14_0) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/69.0.3497.57 Safari/537.36',
-            })
+        def get_error_title(html):
+            return clean_html(get_element_by_class('page-title', html)) or None
 
-        title = self._html_search_regex(
-            (r'<[^>]+\bid=["\']video-title[^>]+>([^<]+)', r'<title>([^<]+)'),
-            webpage, 'title', default=None) or self._html_search_meta(
-            'description', webpage, 'title',
-            default=None) or self._og_search_description(webpage)
+        webpage, urlh = self._download_webpage_handle(
+            'https://old.bitchute.com/video/{0}'.format(video_id), video_id,
+            headers=self._HEADERS, expected_status=404)
 
-        format_urls = []
-        for mobj in re.finditer(
-                r'addWebSeed\s*\(\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage):
-            format_urls.append(mobj.group('url'))
-        format_urls.extend(re.findall(r'as=(https?://[^&"\']+)', webpage))
+        self._raise_if_restricted(webpage)
 
-        formats = [
-            {'url': format_url}
-            for format_url in orderedSet(format_urls)]
+        if urlh.getcode() == 404:
+            raise ExtractorError(get_error_title(webpage) or 'Cannot find video', expected=True)
+
+        entries = self._parse_html5_media_entries(url, webpage, video_id)
+
+        title = (
+            clean_html(get_element_by_id('video-title', webpage))
+            or self._og_search_title(webpage, default=None)
+            or self._html_search_regex(r'(?s)<title(?!-)\b[^>]*>(.*?)</title', webpage, 'title'))
+
+        formats = []
+        for format_ in traverse_obj(entries, (0, 'formats', Ellipsis)):
+            if self.get_param('check_formats') is not False:
+                format_.update(self._check_format(format_.pop('url'), video_id) or {})
+                if 'url' not in format_:
+                    continue
+            formats.append(format_)
 
         if not formats:
-            formats = self._parse_html5_media_entries(
-                url, webpage, video_id)[0]['formats']
+            self.raise_no_formats(
+                'Video is unavailable. Please make sure this video is playable in the browser '
+                'before reporting this issue.', expected=True, video_id=video_id)
 
-        self._check_formats(formats, video_id)
         self._sort_formats(formats)
 
-        description = self._html_search_regex(
-            r'(?s)<div\b[^>]+\bclass=["\']full hidden[^>]+>(.+?)</div>',
-            webpage, 'description', fatal=False)
-        thumbnail = self._og_search_thumbnail(
-            webpage, default=None) or self._html_search_meta(
-            'twitter:image:src', webpage, 'thumbnail')
-        uploader = self._html_search_regex(
-            (r'(?s)<div class=["\']channel-banner.*?<p\b[^>]+\bclass=["\']name[^>]+>(.+?)</p>',
-             r'(?s)<p\b[^>]+\bclass=["\']video-author[^>]+>(.+?)</p>'),
-            webpage, 'uploader', fatal=False)
+        details = get_element_by_class('details', webpage) or ''
+        uploader_html = get_element_html_by_class('creator', details) or ''
+        channel_html = get_element_html_by_class('name', details) or ''
 
-        upload_date = unified_strdate(self._search_regex(
-            r'class=["\']video-publish-date[^>]+>[^<]+ at \d+:\d+ UTC on (.+?)\.',
-            webpage, 'upload date', fatal=False))
+        def more_unified_timestamp(datetime_str):
+            # ... at hh:mm TZ on month nth, yyyy
+            m_ts = re.search(
+                r'\bat\s+(\d+:\d+)\s+([A-Z]+)\s+on\s+(\w+)\s+(\d+)(?:st|nd|rd|th)?[,\s]\s*(\d+)\s*\.*$',
+                (datetime_str or '').strip())
+            if m_ts:
+                new_dt = ' '.join(m_ts.group(4, 3, 5, 1, 2))
+                new_dt = unified_timestamp(new_dt)
+                if new_dt:
+                    return new_dt
+            return unified_timestamp(datetime_str)
+
+        timestamp = more_unified_timestamp(
+            get_element_by_class('video-publish-date', webpage))
 
         return {
             'id': video_id,
             'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'uploader': uploader,
-            'upload_date': upload_date,
+            'description': (
+                self._og_search_description(webpage, default=None)
+                or clean_html(get_element_by_class('full hidden', webpage))),
+            'thumbnail': self._html_search_meta(
+                ('og:image', 'twitter:image:src'), webpage, 'thumbnail',
+                fatal=False),
+            'uploader': clean_html(uploader_html),
+            'uploader_url': self._make_url(uploader_html),
+            'channel': clean_html(channel_html),
+            'channel_url': self._make_url(channel_html),
+            'timestamp': timestamp,
             'formats': formats,
+            'age_limit': 18 if '>This video has been marked as Not Safe For Work' in webpage else None,
         }
 
 
-class BitChuteChannelIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?bitchute\.com/channel/(?P<id>[^/?#&]+)'
-    _TEST = {
-        'url': 'https://www.bitchute.com/channel/victoriaxrave/',
-        'playlist_mincount': 185,
+class BitChuteChannelIE(BitChuteBaseIE):
+    _VALID_URL = r'https?://(?:(?:www|old)\.)?bitchute\.com/(?P<type>channel|playlist)/(?P<id>[^/?#&]+)'
+    _TESTS = [{
+        'url': 'https://www.bitchute.com/channel/bitchute/',
         'info_dict': {
-            'id': 'victoriaxrave',
+            'id': 'bitchute',
+            'title': 'BitChute',
+            'description': 'md5:2134c37d64fc3a4846787c402956adac',
         },
+        'playlist': [
+            {
+                'md5': '7e427d7ed7af5a75b5855705ec750e2b',
+                'info_dict': {
+                    'id': 'UGlrF9o9b-Q',
+                    'ext': 'mp4',
+                    'title': 'This is the first video on #BitChute !',
+                    'description': 'md5:a0337e7b1fe39e32336974af8173a034',
+                    'thumbnail': r're:^https?://.*\.jpg$',
+                    'uploader': 'BitChute',
+                    'timestamp': 1483425420,
+                    'upload_date': '20170103',
+                    'uploader_url': 'https://www.bitchute.com/profile/I5NgtHZn9vPj/',
+                    'channel': 'BitChute',
+                    'channel_url': 'https://www.bitchute.com/channel/bitchute/',
+                    'duration': 16,
+                    'view_count': int,
+                },
+            },
+        ],
+        'params': {
+            'skip_download': True,
+            'playlist_items': '109',  # ie, -1
+        },
+    }, {
+        'url': 'https://www.bitchute.com/playlist/wV9Imujxasw9/',
+        'playlist_mincount': 20,
+        'info_dict': {
+            'id': 'wV9Imujxasw9',
+            'title': 'Bruce MacDonald and "The Light of Darkness"',
+            'description': r're:(?s)Bruce MacDonald co-wrote a book .{803} available to buy at Amazon\.$',
+        },
+    }, {
+        'url': 'https://www.bitchute.com/playlist/g4WTfWTdYEQa/',
+        'playlist_mincount': 1,
+        'info_dict': {
+            'id': 'g4WTfWTdYEQa',
+            'title': 'Podcasts',
+            'description': 'Podcast Playlist',
+        },
+    }, {
+        'url': 'https://old.bitchute.com/playlist/wV9Imujxasw9/',
+        'only_matching': True,
+    }]
+
+    _API_URL = 'https://old.bitchute.com'
+    _TOKEN = 'zyG6tQcGPE5swyAEFLqKUwMuMMuF6IO2DZ6ZDQjGfsL0e4dcTLwqkTTul05Jdve7'
+    PAGE_SIZE = 25
+    HTML_CLASS_NAMES = {
+        'channel': {
+            'container': 'channel-videos-container',
+            'title': 'channel-videos-title',
+            'description': 'channel-videos-text',
+        },
+        'playlist': {
+            'container': 'playlist-video',
+            'title': 'title',
+            'description': 'description',
+        },
+
     }
 
-    _TOKEN = 'zyG6tQcGPE5swyAEFLqKUwMuMMuF6IO2DZ6ZDQjGfsL0e4dcTLwqkTTul05Jdve7'
+    @classmethod
+    def _make_url(cls, playlist_id, playlist_type):
+        return '{0}/{1}/{2}/'.format(cls._API_URL, playlist_type, playlist_id)
 
-    def _entries(self, channel_id):
-        channel_url = 'https://www.bitchute.com/channel/%s/' % channel_id
-        offset = 0
-        for page_num in itertools.count(1):
-            data = self._download_json(
-                '%sextend/' % channel_url, channel_id,
-                'Downloading channel page %d' % page_num,
-                data=urlencode_postdata({
-                    'csrfmiddlewaretoken': self._TOKEN,
-                    'name': '',
-                    'offset': offset,
-                }), headers={
-                    'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
-                    'Referer': channel_url,
-                    'X-Requested-With': 'XMLHttpRequest',
-                    'Cookie': 'csrftoken=%s' % self._TOKEN,
-                })
-            if data.get('success') is False:
-                break
-            html = data.get('html')
-            if not html:
-                break
-            video_ids = re.findall(
-                r'class=["\']channel-videos-image-container[^>]+>\s*<a\b[^>]+\bhref=["\']/video/([^"\'/]+)',
-                html)
-            if not video_ids:
-                break
-            offset += len(video_ids)
-            for video_id in video_ids:
-                yield self.url_result(
-                    'https://www.bitchute.com/video/%s' % video_id,
-                    ie=BitChuteIE.ie_key(), video_id=video_id)
+    def _fetch_page(self, playlist_id, playlist_type, page_num):
+        playlist_url = self._make_url(playlist_id, playlist_type)
+        data = self._download_json(
+            '{0}extend/'.format(playlist_url), playlist_id, 'Downloading page {0}'.format(page_num),
+            data=urlencode_postdata({
+                'csrfmiddlewaretoken': self._TOKEN,
+                'name': '',
+                'offset': page_num * self.PAGE_SIZE,
+            }), headers={
+                'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
+                'Referer': playlist_url,
+                'X-Requested-With': 'XMLHttpRequest',
+                'Cookie': 'csrftoken=%s' % (self._TOKEN,),
+            })
+        if not traverse_obj(data, 'success'):
+            return
+        classes = self.HTML_CLASS_NAMES[playlist_type]
+        for video_html in get_elements_html_by_class(classes['container'], data.get('html')):
+            video_id = self._search_regex(
+                r'<a\s[^>]*\b(?<!-)href\s*=\s*["\']/video/([^"\'/]+)', video_html, 'video id', default=None)
+            if not video_id:
+                continue
+            yield merge_dicts({
+                '_type': 'url_transparent',
+            }, self.url_result(
+                'https://www.bitchute.com/video/{0}'.format(video_id),
+                BitChuteIE.ie_key(), video_id), {
+                'title': clean_html(get_element_by_class(classes['title'], video_html)),
+                'description': clean_html(get_element_by_class(classes['description'], video_html)),
+                'duration': parse_duration(get_element_by_class('video-duration', video_html)),
+                'view_count': parse_count(clean_html(get_element_by_class('video-views', video_html))),
+            })
 
     def _real_extract(self, url):
-        channel_id = self._match_id(url)
-        return self.playlist_result(
-            self._entries(channel_id), playlist_id=channel_id)
+        playlist_type, playlist_id = self._match_valid_url(url).group('type', 'id')
+        webpage = self._download_webpage(self._make_url(playlist_id, playlist_type), playlist_id)
+
+        page_func = functools.partial(self._fetch_page, playlist_id, playlist_type)
+        return merge_dicts(self.playlist_result(
+            OnDemandPagedList(page_func, self.PAGE_SIZE), playlist_id), {
+            'title': self._html_search_regex(
+                r'(?s)<title\b(?<!-)[^>]*>([^<]+)</title>', webpage, 'title', default=None),
+            'description': (
+                clean_html(get_element_by_class('description', webpage))
+                or self._html_search_meta(
+                    ('description', 'og:description', 'twitter:description'), webpage, default=None)),
+            'playlist_count': int_or_none(self._html_search_regex(
+                r'<span>(\d+)\s+videos?</span>', webpage, 'playlist count', default=None)),
+        })
+
+
+__all__ = [
+    'BitChuteIE',
+    'BitChuteChannelIE',
+]

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(',')],