Merge 5abcd988a2bbbccbd8496ec255dd771f02bfef27 into 3eb8d22ddb8982ca4fb56bb7a8d6517538bf14c6

* thx seproDev, yt-dlp/yt-dlp/pull/12694/commits/ae69e3c
* also add documenting comment in `InfoExtractor`

youtube_dl/extractor/bokecc.py

@@ -32,7 +32,7 @@ class BokeCCBaseIE(InfoExtractor):
 
 
 class BokeCCIE(BokeCCBaseIE):
-    _IE_DESC = 'CC视频'
+    IE_DESC = 'CC视频'
     _VALID_URL = r'https?://union\.bokecc\.com/playvideo\.bo\?(?P<query>.*)'
 
     _TESTS = [{

youtube_dl/extractor/cloudy.py

@@ -9,7 +9,7 @@ from ..utils import (
 
 
 class CloudyIE(InfoExtractor):
-    _IE_DESC = 'cloudy.ec'
+    IE_DESC = 'cloudy.ec'
     _VALID_URL = r'https?://(?:www\.)?cloudy\.ec/(?:v/|embed\.php\?.*?\bid=)(?P<id>[A-Za-z0-9]+)'
     _TESTS = [{
         'url': 'https://www.cloudy.ec/v/af511e2527aac',

youtube_dl/extractor/common.py

@@ -422,6 +422,8 @@ class InfoExtractor(object):
     _GEO_COUNTRIES = None
     _GEO_IP_BLOCKS = None
     _WORKING = True
+    # supply this in public subclasses: used in supported sites list, etc
+    # IE_DESC = 'short description of IE'
 
     def __init__(self, downloader=None):
         """Constructor. Receives an optional downloader."""

youtube_dl/extractor/cozytv.py

@@ -0,0 +1,172 @@
+# coding: utf-8
+from __future__ import unicode_literals
+
+import re
+
+from .common import InfoExtractor
+from ..utils import (
+    float_or_none,
+    int_or_none,
+    unified_timestamp,
+    ExtractorError,
+)
+
+
+class CozyTVBaseIE(InfoExtractor):
+    _API_BASE = 'https://api.cozy.tv'
+
+    def _join_url(self, *parts):
+        return '/'.join(parts).rstrip('/')
+
+    def _user_status(self, user, video_id):
+        url = self._join_url(self._API_BASE, 'cache', user, 'status')
+        return self._download_json(url, video_id)
+
+    def _extend_formats(self, formats, cdn_url, video_id, live=False):
+        formats.extend(self._extract_m3u8_formats(
+            cdn_url + '/index.m3u8',
+            video_id,
+            ext='mp4',
+            m3u8_id='hls',
+            preference=-1,
+            live=live))
+        formats.extend(self._extract_m3u8_formats(
+            cdn_url + '/index-ts-audio.m3u8',
+            video_id,
+            ext='mp4',
+            m3u8_id='hls-audio',
+            preference=-2,
+            fatal=False,
+            live=live))
+        self._sort_formats(formats)
+
+
+class CozyTVStreamIE(CozyTVBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?cozy\.tv/(?P<user>[^/?#]+)'
+    _TEST = {
+        'url': 'https://cozy.tv/alexjones',
+        'info_dict': {
+            'id': 'alexjones',
+            'title': 'The Alex Jones Show - Infowars.com',
+            'ext': 'mp4',
+            'uploader': 'AlexJones',
+            'timestamp': 1658422904,
+            'upload_date': '20220721',
+            'uploader_id': 'alexjones',
+            'uploader_url': 'https://cozy.tv/alexjones',
+            'view_count': int,
+            'is_live': True,
+        },
+        'params': {
+            'skip_download': True,
+        },
+    }
+
+    def _real_extract(self, url):
+        video_id, user = re.match(self._VALID_URL, url).group('user', 'user')
+        user_status = self._user_status(user, video_id)
+        is_live = user_status['isLive']
+
+        if is_live is None:
+            raise ExtractorError('%s is offline' % user, expected=True)
+
+        cdn_url = self._join_url(user_status['livecdns'][0], 'live', user)
+
+        info = {
+            'id': video_id,
+            'title': user_status['title'],
+        }
+
+        timestamp = unified_timestamp(is_live)
+        formats = []
+        self._extend_formats(formats, cdn_url, video_id, live=True)
+
+        info.update({
+            'display_id': video_id,
+            'formats': formats,
+            'timestamp': timestamp,
+            'uploader': user_status.get('displayName') or user,
+            'uploader_id': user,
+            'uploader_url': 'https://cozy.tv/' + user,
+            'view_count': int_or_none(user_status.get('viewers')),
+            'is_live': True,
+        })
+
+        return info
+
+
+class CozyTVReplayIE(CozyTVBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?cozy\.tv/(?P<user>[^/]+)/replays/(?P<id>\d{4}-(?:0[1-9]|1[0-2])-(?:0[1-9]|[12][0-9]|3[01])(?:_\d+)?)'
+    _TESTS = [
+        {
+            'url': 'https://cozy.tv/althype/replays/2022-05-24_2',
+            'info_dict': {
+                'id': '628d070e3f2f1a99756b6c2f',
+                'title': 'Secret Stream: Making Video',
+                'ext': 'mp4',
+                'display_id': '2022-05-24_2',
+                'thumbnail': r're:^https?://.*/replays/althype/2022-05-24_2/thumb\.webp$',
+                'uploader': 'AltHype',
+                'timestamp': 1653409506,
+                'upload_date': '20220524',
+                'uploader_id': 'althype',
+                'uploader_url': 'https://cozy.tv/althype',
+                'duration': 9303.0,
+                'view_count': 217,
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+        {
+            'url': 'https://cozy.tv/nick/replays/2022-07-28',
+            'info_dict': {
+                'id': '62e23d1c4a22de6fb37b83f5',
+                'title': 'GAY PANDEMIC: Biden Admin To Declare MONKEYPOX EMERGENCY | America First Ep. 1037',
+                'ext': 'mp4',
+                'display_id': '2022-07-28',
+                'thumbnail': r're:^https?://.*/replays/nick/2022-07-28/thumb\.webp$',
+                'uploader': 'Nick',
+                'timestamp': 1658993895,
+                'upload_date': '20220728',
+                'uploader_id': 'nick',
+                'uploader_url': 'https://cozy.tv/nick',
+                'duration': 16410.0,
+                'view_count': 6281,
+            },
+            'params': {
+                'skip_download': True,
+            },
+        },
+    ]
+
+    def _real_extract(self, url):
+        video_id, user = re.match(self._VALID_URL, url).group('id', 'user')
+        user_status = self._user_status(user, video_id)
+        data_url = self._join_url(self._API_BASE, 'cache', user, 'replay', video_id)
+        data = self._download_json(data_url, video_id)
+
+        cdn_url = self._join_url(data['cdns'][0], 'replays', user, video_id)
+
+        info = {
+            'id': data.get('_id') or video_id,
+            'title': data['title'],
+        }
+
+        timestamp = unified_timestamp(data.get('date'))
+        formats = []
+        self._extend_formats(formats, cdn_url, video_id)
+
+        info.update({
+            'display_id': video_id,
+            'duration': float_or_none(data.get('duration')),
+            'formats': formats,
+            'thumbnail': cdn_url + '/thumb.webp',
+            'timestamp': timestamp,
+            'uploader': user_status.get('displayName') or user,
+            'uploader_id': user,
+            'uploader_url': 'https://cozy.tv/' + user,
+            'view_count': int_or_none(data.get('peakViewers')),
+        })
+
+        return info

youtube_dl/extractor/extractors.py

@@ -248,6 +248,10 @@ from .cnn import (
     CNNArticleIE,
 )
 from .coub import CoubIE
+from .cozytv import (
+    CozyTVReplayIE,
+    CozyTVStreamIE,
+)
 from .comedycentral import (
     ComedyCentralIE,
     ComedyCentralTVIE,

youtube_dl/extractor/itv.py

@@ -35,15 +35,6 @@ from ..utils import (
 
 class ITVBaseIE(InfoExtractor):
 
-    def _search_nextjs_data(self, webpage, video_id, **kw):
-        transform_source = kw.pop('transform_source', None)
-        fatal = kw.pop('fatal', True)
-        return self._parse_json(
-            self._search_regex(
-                r'''<script\b[^>]+\bid=('|")__NEXT_DATA__\1[^>]*>(?P<js>[^<]+)</script>''',
-                webpage, 'next.js data', group='js', fatal=fatal, **kw),
-            video_id, transform_source=transform_source, fatal=fatal)
-
     def __handle_request_webpage_error(self, err, video_id=None, errnote=None, fatal=True):
         if errnote is False:
             return False
@@ -109,7 +100,9 @@ class ITVBaseIE(InfoExtractor):
 
 class ITVIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?:(?P<w>watch)|hub)/[^/]+/(?(w)[\w-]+/)(?P<id>\w+)'
-    _IE_DESC = 'ITVX'
+    IE_DESC = 'ITVX'
+    _WORKING = False
+
     _TESTS = [{
         'note': 'Hub URLs redirect to ITVX',
         'url': 'https://www.itv.com/hub/liar/2a4547a0012',
@@ -270,7 +263,7 @@ class ITVIE(ITVBaseIE):
                 'ext': determine_ext(href, 'vtt'),
             })
 
-        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default='{}')
+        next_data = self._search_nextjs_data(webpage, video_id, fatal=False, default={})
         video_data.update(traverse_obj(next_data, ('props', 'pageProps', ('title', 'episode')), expected_type=dict)[0] or {})
         title = traverse_obj(video_data, 'headerTitle', 'episodeTitle')
         info = self._og_extract(webpage, require_title=not title)
@@ -323,7 +316,7 @@ class ITVIE(ITVBaseIE):
 
 class ITVBTCCIE(ITVBaseIE):
     _VALID_URL = r'https?://(?:www\.)?itv\.com/(?!(?:watch|hub)/)(?:[^/]+/)+(?P<id>[^/?#&]+)'
-    _IE_DESC = 'ITV articles: News, British Touring Car Championship'
+    IE_DESC = 'ITV articles: News, British Touring Car Championship'
     _TESTS = [{
         'note': 'British Touring Car Championship',
         'url': 'https://www.itv.com/btcc/articles/btcc-2018-all-the-action-from-brands-hatch',

youtube_dl/extractor/senateisvp.py

@@ -47,7 +47,7 @@ class SenateISVPIE(InfoExtractor):
         ['vetaff', '76462', 'http://vetaff-f.akamaihd.net'],
         ['arch', '', 'http://ussenate-f.akamaihd.net/']
     ]
-    _IE_NAME = 'senate.gov'
+    IE_NAME = 'senate.gov'
     _VALID_URL = r'https?://(?:www\.)?senate\.gov/isvp/?\?(?P<qs>.+)'
     _TESTS = [{
         'url': 'http://www.senate.gov/isvp/?comm=judiciary&type=live&stt=&filename=judiciary031715&auto_play=false&wmode=transparent&poster=http%3A%2F%2Fwww.judiciary.senate.gov%2Fthemes%2Fjudiciary%2Fimages%2Fvideo-poster-flash-fit.png',

youtube_dl/jsinterp.py

@@ -686,6 +686,8 @@ class JSInterpreter(object):
             raise self.Exception('Cannot get index {idx!r:.100}'.format(**locals()), expr=repr(obj), cause=e)
 
     def _dump(self, obj, namespace):
+        if obj is JS_Undefined:
+            return 'undefined'
         try:
             return json.dumps(obj)
         except TypeError: