diff options
Diffstat (limited to 'yt_dlp/extractor/youtube.py')
| -rw-r--r-- | yt_dlp/extractor/youtube.py | 244 |
1 files changed, 161 insertions, 83 deletions
diff --git a/yt_dlp/extractor/youtube.py b/yt_dlp/extractor/youtube.py index e553fff..54da4e3 100644 --- a/yt_dlp/extractor/youtube.py +++ b/yt_dlp/extractor/youtube.py @@ -240,6 +240,16 @@ INNERTUBE_CLIENTS = { }, 'INNERTUBE_CONTEXT_CLIENT_NAME': 85 }, + # This client has pre-merged video+audio 720p/1080p streams + 'mediaconnect': { + 'INNERTUBE_CONTEXT': { + 'client': { + 'clientName': 'MEDIA_CONNECT_FRONTEND', + 'clientVersion': '0.1', + }, + }, + 'INNERTUBE_CONTEXT_CLIENT_NAME': 95 + }, } @@ -1171,7 +1181,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 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$', r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$', ) - _formats = { + _formats = { # NB: Used in YoutubeWebArchiveIE and GoogleDriveIE '5': {'ext': 'flv', 'width': 400, 'height': 240, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'}, '6': {'ext': 'flv', 'width': 450, 'height': 270, 'acodec': 'mp3', 'abr': 64, 'vcodec': 'h263'}, '13': {'ext': '3gp', 'acodec': 'aac', 'vcodec': 'mp4v'}, @@ -1315,6 +1325,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister', 'uploader_id': '@PhilippHagemeister', 'heatmap': 'count:100', + 'timestamp': 1349198244, } }, { @@ -1358,6 +1369,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': 'https://www.youtube.com/@PhilippHagemeister', 'uploader_id': '@PhilippHagemeister', 'heatmap': 'count:100', + 'timestamp': 1349198244, }, 'params': { 'skip_download': True, @@ -1444,6 +1456,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'comment_count': int, 'channel_is_verified': True, 'heatmap': 'count:100', + 'timestamp': 1401991663, }, }, { @@ -1503,6 +1516,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'Projekt Melody', 'uploader_url': 'https://www.youtube.com/@ProjektMelody', 'uploader_id': '@ProjektMelody', + 'timestamp': 1577508724, }, }, { @@ -1608,6 +1622,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': 'https://www.youtube.com/@Olympics', 'uploader_id': '@Olympics', 'channel_is_verified': True, + 'timestamp': 1440707674, }, 'params': { 'skip_download': 'requires avconv', @@ -1641,6 +1656,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': '孫ᄋᄅ', 'uploader_url': 'https://www.youtube.com/@AllenMeow', 'uploader_id': '@AllenMeow', + 'timestamp': 1299776999, }, }, # url_encoded_fmt_stream_map is empty string @@ -1784,6 +1800,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }, }], 'params': {'skip_download': True}, + 'skip': 'Not multifeed anymore', }, { # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536) @@ -1892,6 +1909,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'The Berkman Klein Center for Internet & Society', 'uploader_id': '@BKCHarvard', 'uploader_url': 'https://www.youtube.com/@BKCHarvard', + 'timestamp': 1422422076, }, 'params': { 'skip_download': True, @@ -1927,6 +1945,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': '@BernieSanders', 'channel_is_verified': True, 'heatmap': 'count:100', + 'timestamp': 1447987198, }, 'params': { 'skip_download': True, @@ -1990,6 +2009,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': '@Vsauce', 'comment_count': int, 'channel_is_verified': True, + 'timestamp': 1484761047, }, 'params': { 'skip_download': True, @@ -2145,6 +2165,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'l\'Or Vert asbl', 'uploader_url': 'https://www.youtube.com/@ElevageOrVert', 'uploader_id': '@ElevageOrVert', + 'timestamp': 1497343210, }, 'params': { 'skip_download': True, @@ -2183,6 +2204,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': '@Csharp-video-tutorialsBlogspot', 'channel_is_verified': True, 'heatmap': 'count:100', + 'timestamp': 1377976349, }, 'params': { 'skip_download': True, @@ -2265,6 +2287,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': '@CBSMornings', 'comment_count': int, 'channel_is_verified': True, + 'timestamp': 1405513526, } }, { @@ -2282,7 +2305,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'view_count': int, 'channel': 'Walk around Japan', 'tags': ['Ueno Tokyo', 'Okachimachi Tokyo', 'Ameyoko Street', 'Tokyo attraction', 'Travel in Tokyo'], - 'thumbnail': 'https://i.ytimg.com/vi_webp/cBvYw8_A0vQ/hqdefault.webp', + 'thumbnail': 'https://i.ytimg.com/vi/cBvYw8_A0vQ/hqdefault.jpg', 'age_limit': 0, 'availability': 'public', 'channel_url': 'https://www.youtube.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw', @@ -2292,6 +2315,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'Walk around Japan', 'uploader_url': 'https://www.youtube.com/@walkaroundjapan7124', 'uploader_id': '@walkaroundjapan7124', + 'timestamp': 1605884416, }, 'params': { 'skip_download': True, @@ -2343,6 +2367,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'format': '17', # 3gp format available on android 'extractor_args': {'youtube': {'player_client': ['android']}}, }, + 'skip': 'android client broken', }, { # Skip download of additional client configs (remix client config in this case) @@ -2386,6 +2411,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'comment_count': int, 'channel_is_verified': True, 'heatmap': 'count:100', + 'timestamp': 1395685455, }, 'params': {'format': 'mhtml', 'skip_download': True} }, { # Ensure video upload_date is in UTC timezone (video was uploaded 1641170939) @@ -2415,38 +2441,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': 'https://www.youtube.com/@LeonNguyen', 'uploader_id': '@LeonNguyen', 'heatmap': 'count:100', + 'timestamp': 1641170939, } }, { - # Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date - 'url': 'https://www.youtube.com/watch?v=2NUZ8W2llS4', - 'info_dict': { - 'id': '2NUZ8W2llS4', - 'ext': 'mp4', - 'title': 'The NP that test your phone performance 🙂', - 'description': 'md5:144494b24d4f9dfacb97c1bbef5de84d', - 'channel_id': 'UCRqNBSOHgilHfAczlUmlWHA', - 'channel_url': 'https://www.youtube.com/channel/UCRqNBSOHgilHfAczlUmlWHA', - 'duration': 21, - 'view_count': int, - 'age_limit': 0, - 'categories': ['Gaming'], - 'tags': 'count:23', - 'playable_in_embed': True, - 'live_status': 'not_live', - 'upload_date': '20220102', - 'like_count': int, - 'availability': 'public', - 'channel': 'Leon Nguyen', - 'thumbnail': 'https://i.ytimg.com/vi_webp/2NUZ8W2llS4/maxresdefault.webp', - 'comment_count': int, - 'channel_follower_count': int, - 'uploader': 'Leon Nguyen', - 'uploader_url': 'https://www.youtube.com/@LeonNguyen', - 'uploader_id': '@LeonNguyen', - 'heatmap': 'count:100', - }, - 'params': {'compat_opts': ['no-youtube-prefer-utc-upload-date']} - }, { # date text is premiered video, ensure upload date in UTC (published 1641172509) 'url': 'https://www.youtube.com/watch?v=mzZzzBU6lrM', 'info_dict': { @@ -2477,38 +2474,41 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'comment_count': int, 'channel_is_verified': True, 'heatmap': 'count:100', + 'timestamp': 1641172509, } }, - { # continuous livestream. Microformat upload date should be preferred. - # Upload date was 2021-06-19 (not UTC), while stream start is 2021-11-27 - 'url': 'https://www.youtube.com/watch?v=kgx4WGK0oNU', + { # continuous livestream. + # Upload date was 2022-07-12T05:12:29-07:00, while stream start is 2022-07-12T15:59:30+00:00 + 'url': 'https://www.youtube.com/watch?v=jfKfPfyJRdk', 'info_dict': { - 'id': 'kgx4WGK0oNU', - 'title': r're:jazz\/lofi hip hop radio🌱chill beats to relax\/study to \[LIVE 24\/7\] \d{4}-\d{2}-\d{2} \d{2}:\d{2}', + 'id': 'jfKfPfyJRdk', 'ext': 'mp4', - 'channel_id': 'UC84whx2xxsiA1gXHXXqKGOA', - 'availability': 'public', + 'channel_id': 'UCSJ4gkVC6NrvII8umztf0Ow', + 'like_count': int, + 'uploader': 'Lofi Girl', + 'categories': ['Music'], + 'concurrent_view_count': int, + 'playable_in_embed': True, + 'timestamp': 1657627949, + 'release_date': '20220712', + 'channel_url': 'https://www.youtube.com/channel/UCSJ4gkVC6NrvII8umztf0Ow', + 'description': 'md5:13a6f76df898f5674f9127139f3df6f7', 'age_limit': 0, - 'release_timestamp': 1637975704, - 'upload_date': '20210619', - 'channel_url': 'https://www.youtube.com/channel/UC84whx2xxsiA1gXHXXqKGOA', - 'live_status': 'is_live', - 'thumbnail': 'https://i.ytimg.com/vi/kgx4WGK0oNU/maxresdefault.jpg', - 'channel': 'Abao in Tokyo', + 'thumbnail': 'https://i.ytimg.com/vi/jfKfPfyJRdk/maxresdefault.jpg', + 'release_timestamp': 1657641570, + 'uploader_url': 'https://www.youtube.com/@LofiGirl', 'channel_follower_count': int, - 'release_date': '20211127', - 'tags': 'count:39', - 'categories': ['People & Blogs'], - 'like_count': int, + 'channel_is_verified': True, + 'title': r're:^lofi hip hop radio 📚 - beats to relax/study to', 'view_count': int, - 'playable_in_embed': True, - 'description': 'md5:2ef1d002cad520f65825346e2084e49d', - 'concurrent_view_count': int, - 'uploader': 'Abao in Tokyo', - 'uploader_url': 'https://www.youtube.com/@abaointokyo', - 'uploader_id': '@abaointokyo', + 'live_status': 'is_live', + 'tags': 'count:32', + 'channel': 'Lofi Girl', + 'availability': 'public', + 'upload_date': '20220712', + 'uploader_id': '@LofiGirl', }, - 'params': {'skip_download': True} + 'params': {'skip_download': True}, }, { 'url': 'https://www.youtube.com/watch?v=tjjjtzRLHvA', 'info_dict': { @@ -2534,6 +2534,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': '@lesmiscore', 'uploader': 'Lesmiscore', 'uploader_url': 'https://www.youtube.com/@lesmiscore', + 'timestamp': 1648005313, } }, { # Prefer primary title+description language metadata by default @@ -2561,6 +2562,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': 'https://www.youtube.com/@coletdjnz', 'uploader_id': '@coletdjnz', 'uploader': 'cole-dlp-test-acc', + 'timestamp': 1662677394, }, 'params': {'skip_download': True} }, { @@ -2574,7 +2576,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'duration': 5, 'live_status': 'not_live', 'channel_id': 'UCiu-3thuViMebBjw_5nWYrA', - 'upload_date': '20220728', + 'upload_date': '20220729', 'view_count': int, 'categories': ['People & Blogs'], 'thumbnail': r're:^https?://.*\.jpg', @@ -2587,6 +2589,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': 'https://www.youtube.com/@coletdjnz', 'uploader_id': '@coletdjnz', 'uploader': 'cole-dlp-test-acc', + 'timestamp': 1659073275, + 'like_count': int, }, 'params': {'skip_download': True, 'extractor_args': {'youtube': {'lang': ['fr']}}}, 'expected_warnings': [r'Preferring "fr" translated fields'], @@ -2652,6 +2656,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'Projekt Melody', 'uploader_id': '@ProjektMelody', 'uploader_url': 'https://www.youtube.com/@ProjektMelody', + 'timestamp': 1577508724, }, 'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'}, }, @@ -2686,6 +2691,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': '@sana_natori', 'channel_is_verified': True, 'heatmap': 'count:100', + 'timestamp': 1671798112, }, }, { @@ -2720,7 +2726,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'heatmap': 'count:100', }, 'params': { - 'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}}, + 'extractor_args': {'youtube': {'player_client': ['ios'], 'player_skip': ['webpage']}}, }, }, ] @@ -2755,6 +2761,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries', 'uploader_id': '@ChristopherSykesDocumentaries', 'heatmap': 'count:100', + 'timestamp': 1211825920, }, 'params': { 'skip_download': True, @@ -3307,7 +3314,36 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'value': ('intensityScoreNormalized', {float_or_none}), })) or None - def _extract_comment(self, comment_renderer, parent=None): + def _extract_comment(self, entities, parent=None): + comment_entity_payload = get_first(entities, ('payload', 'commentEntityPayload', {dict})) + if not (comment_id := traverse_obj(comment_entity_payload, ('properties', 'commentId', {str}))): + return + + toolbar_entity_payload = get_first(entities, ('payload', 'engagementToolbarStateEntityPayload', {dict})) + time_text = traverse_obj(comment_entity_payload, ('properties', 'publishedTime', {str})) or '' + + return { + 'id': comment_id, + 'parent': parent or 'root', + **traverse_obj(comment_entity_payload, { + 'text': ('properties', 'content', 'content', {str}), + 'like_count': ('toolbar', 'likeCountA11y', {parse_count}), + 'author_id': ('author', 'channelId', {self.ucid_or_none}), + 'author': ('author', 'displayName', {str}), + 'author_thumbnail': ('author', 'avatarThumbnailUrl', {url_or_none}), + 'author_is_uploader': ('author', 'isCreator', {bool}), + 'author_is_verified': ('author', 'isVerified', {bool}), + 'author_url': ('author', 'channelCommand', 'innertubeCommand', ( + ('browseEndpoint', 'canonicalBaseUrl'), ('commandMetadata', 'webCommandMetadata', 'url') + ), {lambda x: urljoin('https://www.youtube.com', x)}), + }, get_all=False), + 'is_favorited': (None if toolbar_entity_payload is None else + toolbar_entity_payload.get('heartState') == 'TOOLBAR_HEART_STATE_HEARTED'), + '_time_text': time_text, # FIXME: non-standard, but we need a way of showing that it is an estimate. + 'timestamp': self._parse_time_text(time_text), + } + + def _extract_comment_old(self, comment_renderer, parent=None): comment_id = comment_renderer.get('commentId') if not comment_id: return @@ -3388,21 +3424,39 @@ class YoutubeIE(YoutubeBaseInfoExtractor): break return _continuation - def extract_thread(contents): + def extract_thread(contents, entity_payloads): if not parent: tracker['current_page_thread'] = 0 for content in contents: if not parent and tracker['total_parent_comments'] >= max_parents: yield comment_thread_renderer = try_get(content, lambda x: x['commentThreadRenderer']) - comment_renderer = get_first( - (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]], - expected_type=dict, default={}) - comment = self._extract_comment(comment_renderer, parent) + # old comment format + if not entity_payloads: + comment_renderer = get_first( + (comment_thread_renderer, content), [['commentRenderer', ('comment', 'commentRenderer')]], + expected_type=dict, default={}) + + comment = self._extract_comment_old(comment_renderer, parent) + + # new comment format + else: + view_model = ( + traverse_obj(comment_thread_renderer, ('commentViewModel', 'commentViewModel', {dict})) + or traverse_obj(content, ('commentViewModel', {dict}))) + comment_keys = traverse_obj(view_model, (('commentKey', 'toolbarStateKey'), {str})) + if not comment_keys: + continue + entities = traverse_obj(entity_payloads, lambda _, v: v['entityKey'] in comment_keys) + comment = self._extract_comment(entities, parent) + if comment: + comment['is_pinned'] = traverse_obj(view_model, ('pinnedText', {str})) is not None + if not comment: continue comment_id = comment['id'] + if comment.get('is_pinned'): tracker['pinned_comment_ids'].add(comment_id) # Sometimes YouTube may break and give us infinite looping comments. @@ -3495,7 +3549,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): check_get_keys = None if not is_forced_continuation and not (tracker['est_total'] == 0 and tracker['running_total'] == 0): check_get_keys = [[*continuation_items_path, ..., ( - 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentRenderer'))]] + 'commentsHeaderRenderer' if is_first_continuation else ('commentThreadRenderer', 'commentViewModel', 'commentRenderer'))]] try: response = self._extract_response( item_id=None, query=continuation, @@ -3519,6 +3573,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): raise is_forced_continuation = False continuation = None + mutations = traverse_obj(response, ('frameworkUpdates', 'entityBatchUpdate', 'mutations', ..., {dict})) for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]): if is_first_continuation: continuation = extract_header(continuation_items) @@ -3527,7 +3582,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): break continue - for entry in extract_thread(continuation_items): + for entry in extract_thread(continuation_items, mutations): if not entry: return yield entry @@ -3604,8 +3659,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor): yt_query = { 'videoId': video_id, } - if _split_innertube_client(client)[0] in ('android', 'android_embedscreen'): - yt_query['params'] = 'CgIIAQ==' pp_arg = self._configuration_arg('player_params', [None], casesense=True)[0] if pp_arg: @@ -3621,19 +3674,24 @@ class YoutubeIE(YoutubeBaseInfoExtractor): def _get_requested_clients(self, url, smuggled_data): requested_clients = [] - default = ['ios', 'android', 'web'] + android_clients = [] + default = ['ios', 'web'] allowed_clients = sorted( (client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'), key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True) for client in self._configuration_arg('player_client'): - if client in allowed_clients: - requested_clients.append(client) - elif client == 'default': + if client == 'default': requested_clients.extend(default) elif client == 'all': requested_clients.extend(allowed_clients) - else: + elif client not in allowed_clients: self.report_warning(f'Skipping unsupported client {client}') + elif client.startswith('android'): + android_clients.append(client) + else: + requested_clients.append(client) + # Force deprioritization of broken Android clients for format de-duplication + requested_clients.extend(android_clients) if not requested_clients: requested_clients = default @@ -3852,6 +3910,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor): f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True) client_name = fmt.get(STREAMING_DATA_CLIENT_NAME) + # Android client formats are broken due to integrity check enforcement + # Ref: https://github.com/yt-dlp/yt-dlp/issues/9554 + is_broken = client_name and client_name.startswith(short_client_name('android')) + if is_broken: + self.report_warning( + f'{video_id}: Android client formats are broken and may yield HTTP Error 403. ' + 'They will be deprioritized', only_once=True) + name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or '' fps = int_or_none(fmt.get('fps')) or 0 dct = { @@ -3864,7 +3930,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): name, fmt.get('isDrc') and 'DRC', try_get(fmt, lambda x: x['projectionType'].replace('RECTANGULAR', '').lower()), try_get(fmt, lambda x: x['spatialAudioType'].replace('SPATIAL_AUDIO_TYPE_', '').lower()), - throttled and 'THROTTLED', is_damaged and 'DAMAGED', + throttled and 'THROTTLED', is_damaged and 'DAMAGED', is_broken and 'BROKEN', (self.get_param('verbose') or all_formats) and client_name, delim=', '), # Format 22 is likely to be damaged. See https://github.com/yt-dlp/yt-dlp/issues/3372 @@ -3882,8 +3948,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'language': join_nonempty(audio_track.get('id', '').split('.')[0], 'desc' if language_preference < -1 else '') or None, 'language_preference': language_preference, - # Strictly de-prioritize damaged and 3gp formats - 'preference': -10 if is_damaged else -2 if itag == '17' else None, + # Strictly de-prioritize broken, damaged and 3gp formats + 'preference': -20 if is_broken else -10 if is_damaged else -2 if itag == '17' else None, } mime_mobj = re.match( r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', fmt.get('mimeType') or '') @@ -4552,19 +4618,31 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': channel_handle, 'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None), }) + + # We only want timestamp IF it has time precision AND a timezone + # Currently the uploadDate in microformats appears to be in US/Pacific timezone. + timestamp = ( + parse_iso8601(get_first(microformats, 'uploadDate'), timezone=NO_DEFAULT) + or parse_iso8601(search_meta('uploadDate'), timezone=NO_DEFAULT) + ) + upload_date = ( + dt.datetime.fromtimestamp(timestamp, dt.timezone.utc).strftime('%Y%m%d') if timestamp else + ( + unified_strdate(get_first(microformats, 'uploadDate')) + or unified_strdate(search_meta('uploadDate')) + )) + + # In the case we cannot get the timestamp: # The upload date for scheduled, live and past live streams / premieres in microformats # may be different from the stream date. Although not in UTC, we will prefer it in this case. # See: https://github.com/yt-dlp/yt-dlp/pull/2223#issuecomment-1008485139 - upload_date = ( - unified_strdate(get_first(microformats, 'uploadDate')) - or unified_strdate(search_meta('uploadDate'))) - if not upload_date or ( - live_status in ('not_live', None) - and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', []) - ): + if not upload_date or (not timestamp and live_status in ('not_live', None)): + # this should be in UTC, as configured in the cookie/client context upload_date = strftime_or_none( self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date + info['upload_date'] = upload_date + info['timestamp'] = timestamp if upload_date and live_status not in ('is_live', 'post_live', 'is_upcoming'): # Newly uploaded videos' HLS formats are potentially problematic and need to be checked |