Merge 45c034e262d2e7590cf3183a5e1c3bb2f565ef55 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/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/extractor/zoom.py

@@ -1,68 +1,215 @@
 # coding: utf-8
 from __future__ import unicode_literals
 
-import re
-
 from .common import InfoExtractor
 from ..utils import (
     ExtractorError,
     int_or_none,
     js_to_json,
+    merge_dicts,
     parse_filesize,
+    parse_resolution,
+    strip_or_none,
+    T,
+    traverse_obj,
+    txt_or_none,
+    url_basename,
+    url_or_none,
     urlencode_postdata,
+    urljoin,
 )
 
+from functools import partial
+k_int_or_none = partial(int_or_none, scale=1000)
+
 
 class ZoomIE(InfoExtractor):
     IE_NAME = 'zoom'
-    _VALID_URL = r'(?P<base_url>https?://(?:[^.]+\.)?zoom.us/)rec(?:ording)?/(?:play|share)/(?P<id>[A-Za-z0-9_.-]+)'
+    _VALID_URL = r'(?P<base_url>https?://(?:[^.]+\.)?zoom\.us/)rec(?:ording)?/(?P<type>play|share)/(?P<id>[\w.-]+)'
-    _TEST = {
+    _TESTS = [{
         'url': 'https://economist.zoom.us/rec/play/dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5',
         'md5': 'ab445e8c911fddc4f9adc842c2c5d434',
         'info_dict': {
             'id': 'dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5',
             'ext': 'mp4',
             'title': 'China\'s "two sessions" and the new five-year plan',
+        },
+        'skip': 'Recording requires email authentication to access',
+    }, {
+        'url': 'https://us06web.zoom.us/rec/play/W1ctyErikzJ2CxtwlsTW3xNbiMHze6ZkU1adqeshzivi58DHEJ-7HX2Z8-nqK80a8d4CWHAhrSpsl9mG.OaL6JvfC1gAa1EvZ?canPlayFromShare=true&from=share_recording_detail&continueMode=true&componentName=rec-play&originRequestUrl=https%3A%2F%2Fus06web.zoom.us%2Frec%2Fshare%2F60THDorqjAyUm_IXKS88Z4KgfYRAER3wIG20jgrLqaSFBWJW14qBVBRkfHylpFrk.KXJxuNLN0sRBXyvf',
+        'md5': 'b180e7773a878e4799f194f2280648d5',
+        'info_dict': {
+            'id': 'W1ctyErikzJ2CxtwlsTW3xNbiMHze6ZkU1adqeshzivi58DHEJ-7HX2Z8-nqK80a8d4CWHAhrSpsl9mG.OaL6JvfC1gAa1EvZ',
+            'ext': 'mp4',
+            'title': 'Chipathon Bi-Weekly Meeting',
+            'description': 'Shared screen with speaker view',
+            'timestamp': 1715263581,
+            'upload_date': '20240509',
         }
-    }
+    }, {
+        # play URL
+        'url': 'https://ffgolf.zoom.us/rec/play/qhEhXbrxq1Zoucx8CMtHzq1Z_2YZRPVCqWK_K-2FkEGRsSLDeOX8Tu4P6jtjZcRry8QhIbvKZdtr4UNo.QcPn2debFskI9whJ',
+        'md5': '2c4b1c4e5213ebf9db293e88d9385bee',
+        'info_dict': {
+            'id': 'qhEhXbrxq1Zoucx8CMtHzq1Z_2YZRPVCqWK_K-2FkEGRsSLDeOX8Tu4P6jtjZcRry8QhIbvKZdtr4UNo.QcPn2debFskI9whJ',
+            'ext': 'mp4',
+            'title': 'Prépa AF2023 - Séance 5 du 11 avril - R20/VM/GO',
+        },
+        'skip': 'Recording expired',
+    }, {
+        # share URL
+        'url': 'https://us02web.zoom.us/rec/share/hkUk5Zxcga0nkyNGhVCRfzkA2gX_mzgS3LpTxEEWJz9Y_QpIQ4mZFOUx7KZRZDQA.9LGQBdqmDAYgiZ_8',
+        'md5': '90fdc7cfcaee5d52d1c817fc03c43c9b',
+        'info_dict': {
+            'id': 'hkUk5Zxcga0nkyNGhVCRfzkA2gX_mzgS3LpTxEEWJz9Y_QpIQ4mZFOUx7KZRZDQA.9LGQBdqmDAYgiZ_8',
+            'ext': 'mp4',
+            'title': 'Timea Andrea Lelik\'s Personal Meeting Room',
+        },
+        'skip': 'This recording has expired',
+    }, {
+        # view_with_share URL
+        'url': 'https://cityofdetroit.zoom.us/rec/share/VjE-5kW3xmgbEYqR5KzRgZ1OFZvtMtiXk5HyRJo5kK4m5PYE6RF4rF_oiiO_9qaM.UTAg1MI7JSnF3ZjX',
+        'md5': 'bdc7867a5934c151957fb81321b3c024',
+        'info_dict': {
+            'id': 'VjE-5kW3xmgbEYqR5KzRgZ1OFZvtMtiXk5HyRJo5kK4m5PYE6RF4rF_oiiO_9qaM.UTAg1MI7JSnF3ZjX',
+            'ext': 'mp4',
+            'title': 'February 2022 Detroit Revenue Estimating Conference',
+            'description': 'Speaker view',
+            'timestamp': 1645200510,
+            'upload_date': '20220218',
+            'duration': 7299,
+            'formats': 'mincount:3',
+        },
+    }, {
+        # ytdl-org/youtube-dl#32094
+        'url': 'https://us02web.zoom.us/rec/share/9pdVT4f2XWBaEOSqDJagSsvI0Yu2ixXW0YcJGIVhfV19Zr7E1q5gf0wTMZHnqrvq.Yoq1dDHeeKjaVcv3',
+        'md5': 'fdb6f8df7f5ee0c07ced5fae55c0ced4',
+        'info_dict': {
+            'id': '9pdVT4f2XWBaEOSqDJagSsvI0Yu2ixXW0YcJGIVhfV19Zr7E1q5gf0wTMZHnqrvq.Yoq1dDHeeKjaVcv3',
+            'ext': 'mp4',
+            'title': 'Untersuchungskurs Gruppe V',
+            'description': 'Shared screen with speaker view',
+            'timestamp': 1681889972,
+            'upload_date': '20230419',
+        },
+    }]
 
-    def _real_extract(self, url):
+    def _get_page_data(self, webpage, video_id):
-        base_url, play_id = re.match(self._VALID_URL, url).groups()
+        return self._search_json(
-        webpage = self._download_webpage(url, play_id)
+            r'window\.__data__\s*=', webpage, 'data', video_id, transform_source=js_to_json)
 
+    def _get_real_webpage(self, url, base_url, video_id, url_type):
+        webpage = self._download_webpage(url, video_id, note='Downloading {0} webpage'.format(url_type))
         try:
             form = self._form_hidden_inputs('password_form', webpage)
         except ExtractorError:
-            form = None
+            return webpage
-        if form:
-            password = self._downloader.params.get('videopassword')
-            if not password:
-                raise ExtractorError(
-                    'This video is protected by a passcode, use the --video-password option', expected=True)
-            is_meeting = form.get('useWhichPasswd') == 'meeting'
-            validation = self._download_json(
-                base_url + 'rec/validate%s_passwd' % ('_meet' if is_meeting else ''),
-                play_id, 'Validating passcode', 'Wrong passcode', data=urlencode_postdata({
-                    'id': form[('meet' if is_meeting else 'file') + 'Id'],
-                    'passwd': password,
-                    'action': form.get('action'),
-                }))
-            if not validation.get('status'):
-                raise ExtractorError(validation['errorMessage'], expected=True)
-            webpage = self._download_webpage(url, play_id)
 
-        data = self._parse_json(self._search_regex(
+        password = self.get_param('videopassword')
-            r'(?s)window\.__data__\s*=\s*({.+?});',
+        if not password:
-            webpage, 'data'), play_id, js_to_json)
+            raise ExtractorError(
+                'This video is protected by a passcode: use the --video-password option', expected=True)
+        is_meeting = form.get('useWhichPasswd') == 'meeting'
+        validation = self._download_json(
+            base_url + 'rec/validate%s_passwd' % ('_meet' if is_meeting else ''),
+            video_id, 'Validating passcode', 'Wrong passcode', data=urlencode_postdata({
+                'id': form[('meet' if is_meeting else 'file') + 'Id'],
+                'passwd': password,
+                'action': form.get('action'),
+            }))
+        if not validation.get('status'):
+            raise ExtractorError(validation['errorMessage'], expected=True)
+        return self._download_webpage(url, video_id, note='Re-downloading {0} webpage'.format(url_type))
 
-        return {
+    def _real_extract(self, url):
-            'id': play_id,
+        base_url, url_type, video_id = self._match_valid_url(url).group('base_url', 'type', 'id')
-            'title': data['topic'],
+        query = {}
-            'url': data['viewMp4Url'],
+
-            'width': int_or_none(data.get('viewResolvtionsWidth')),
+        if url_type == 'share':
-            'height': int_or_none(data.get('viewResolvtionsHeight')),
+            webpage = self._get_real_webpage(url, base_url, video_id, 'share')
+            meeting_id = self._get_page_data(webpage, video_id)['meetingId']
+            share_info = self._download_json(
+                '{0}nws/recording/1.0/play/share-info/{1}'.format(base_url, meeting_id),
+                video_id, note='Downloading share info JSON', fatal=False)
+            url = traverse_obj(share_info, (
+                'result', 'redirectUrl', T(lambda u: urljoin(base_url, u))))
+            if not url:
+                raise ExtractorError(traverse_obj(
+                    share_info, 'errorMessage') or 'No video found from share link')
+            query['continueMode'] = 'true'
+
+        webpage = self._get_real_webpage(url, base_url, video_id, 'play')
+        file_id = traverse_obj(webpage, T(lambda x: txt_or_none(self._get_page_data(x, video_id)['fileId'])))
+        if not file_id:
+            # When things go wrong, file_id can be empty string
+            raise ExtractorError('Unable to extract file ID')
+
+        data = self._download_json(
+            '{0}nws/recording/1.0/play/info/{1}'.format(base_url, file_id), video_id, query=query,
+            note='Downloading play info JSON')['result']
+
+        formats = []
+        subtitles = dict(
+            (s_type, [{'url': s_url, 'ext': 'vtt', }])
+            for s_type, s_url in traverse_obj(
+                ('transcript', 'cc', 'chapter'),
+                (Ellipsis,
+                 T(lambda t: (t, urljoin(base_url, txt_or_none(data['%sUrl' % (t,)])))),
+                 T(lambda x: x if x[1] else None)))) or None
+
+        def if_url(f):
+            return lambda x: f(x) if x.get('url') else None
+
+        formats.extend(traverse_obj(data, ((
+            ({
+                'url': ('viewMp4Url', T(url_or_none)),
+                'width': ('viewResolvtions', 0, T(int_or_none)),
+                'height': ('viewResolvtions', 1, T(int_or_none)),
+                'format_id': ('recording', 'id', T(txt_or_none)),
+                'filesize_approx': ('recording', 'fileSizeInMB', T(parse_filesize)),
+            }, T(if_url(lambda x: merge_dicts({
+                'format_note': 'Camera stream',
+                'ext': 'mp4',
+                'preference': 0,
+            }, x)))),
+            ({
+                'url': ('shareMp4Url', T(url_or_none)),
+                'width': ('shareResolvtions', 0, T(int_or_none)),
+                'height': ('shareResolvtions', 1, T(int_or_none)),
+                'format_id': ('shareVideo', 'id', T(txt_or_none)),
+                'filesize_approx': ('recording', 'fileSizeInMB', T(parse_filesize)),
+            }, T(if_url(lambda x: merge_dicts({
+                'format_note': 'Screen share stream',
+                'ext': 'mp4',
+                'preference': -1,
+            }, x)))),
+            ({
+                'url': ('viewMp4WithshareUrl', T(url_or_none)),
+            }, T(if_url(lambda x: merge_dicts({
+                'format_note': 'Screen share with camera',
+                'format_id': 'view_with_share',
+                'ext': 'mp4',
+                'preference': 1,
+            }, parse_resolution(self._search_regex(
+                r'_(\d+x\d+)\.mp4', url_basename(x['url']),
+                'resolution', default=None)), x)
+            )))), all)
+        ))
+
+        if not formats and data.get('message'):
+            raise ExtractorError('No media found; %s said "%s"' % (self.IE_NAME, data['message'],), expected=True)
+        self._sort_formats(formats)
+
+        return merge_dicts(traverse_obj(data, {
+            'title': ('meet', 'topic', T(strip_or_none)),
+            'description': ('recording', 'displayFileName', T(strip_or_none)),
+            'duration': ('duration', T(int_or_none)),
+            'timestamp': (('clipStartTime', 'fileStartTime'), T(k_int_or_none), any),
+        }), {
+            'id': video_id,
+            'subtitles': subtitles,
+            'formats': formats,
             'http_headers': {
                 'Referer': base_url,
             },
-            'filesize_approx': parse_filesize(data.get('fileSize')),
+        })
-        }

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: