THIRD_PARTY = {
'embedUrl': 'https://www.youtube.com/', # Can be any valid URL
}
- BASE_CLIENTS = ('android', 'web', 'tv', 'ios', 'mweb')
+ BASE_CLIENTS = ('ios', 'android', 'web', 'tv', 'mweb')
priority = qualities(BASE_CLIENTS[::-1])
for client, ytcfg in tuple(INNERTUBE_CLIENTS.items()):
'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW,
'BADGE_STYLE_TYPE_VERIFIED': BadgeType.VERIFIED,
- 'BADGE_STYLE_TYPE_VERIFIED_ARTIST': BadgeType.VERIFIED
+ 'BADGE_STYLE_TYPE_VERIFIED_ARTIST': BadgeType.VERIFIED,
}
label_map = {
'live': BadgeType.LIVE_NOW,
'premium': BadgeType.AVAILABILITY_PREMIUM,
'verified': BadgeType.VERIFIED,
- 'official artist channel': BadgeType.VERIFIED
+ 'official artist channel': BadgeType.VERIFIED,
}
badges = []
return funcname
return json.loads(js_to_json(self._search_regex(
- rf'var {re.escape(funcname)}\s*=\s*(\[.+?\]);', jscode,
+ rf'var {re.escape(funcname)}\s*=\s*(\[.+?\])[,;]', jscode,
f'Initial JS player n function list ({funcname}.{idx})')))[int(idx)]
def _extract_n_function_code(self, video_id, player_url):
info['author_is_uploader'] = author_is_uploader
comment_abr = traverse_obj(
- comment_renderer, ('actionsButtons', 'commentActionButtonsRenderer'), expected_type=dict)
+ comment_renderer, ('actionButtons', 'commentActionButtonsRenderer'), expected_type=dict)
if comment_abr is not None:
info['is_favorited'] = 'creatorHeart' in comment_abr
# Pinned comments may appear a second time in newest first sort
# See: https://github.com/yt-dlp/yt-dlp/issues/6712
continue
- self.report_warning('Detected YouTube comments looping. Stopping comment extraction as we probably cannot get any more.')
+ self.report_warning(
+ 'Detected YouTube comments looping. Stopping comment extraction '
+ f'{"for this thread" if parent else ""} as we probably cannot get any more.')
yield
else:
tracker['seen_comment_ids'].add(comment['id'])
# Ignore incomplete data error for replies if retries didn't work.
# This is to allow any other parent comments and comment threads to be downloaded.
# See: https://github.com/yt-dlp/yt-dlp/issues/4669
- if 'incomplete data' in str(e).lower() and parent and self.get_param('ignoreerrors') is True:
- self.report_warning(
- 'Received incomplete data for a comment reply thread and retrying did not help. '
- 'Ignoring to let other comments be downloaded.')
- else:
- raise
+ if 'incomplete data' in str(e).lower() and parent:
+ if self.get_param('ignoreerrors') in (True, 'only_download'):
+ self.report_warning(
+ 'Received incomplete data for a comment reply thread and retrying did not help. '
+ 'Ignoring to let other comments be downloaded. Pass --no-ignore-errors to not ignore.')
+ return
+ else:
+ raise ExtractorError(
+ 'Incomplete data received for comment reply thread. '
+ 'Pass --ignore-errors to ignore and allow rest of comments to download.',
+ expected=True)
+ raise
is_forced_continuation = False
continuation = None
for continuation_items in traverse_obj(response, continuation_items_path, expected_type=list, default=[]):
def _is_unplayable(player_response):
return traverse_obj(player_response, ('playabilityStatus', 'status')) == 'UNPLAYABLE'
- _STORY_PLAYER_PARAMS = '8AEB'
+ _PLAYER_PARAMS = 'CgIQBg=='
def _extract_player_response(self, client, video_id, master_ytcfg, player_ytcfg, player_url, initial_pr, smuggled_data):
'videoId': video_id,
}
if smuggled_data.get('is_story') or _split_innertube_client(client)[0] == 'android':
- yt_query['params'] = self._STORY_PLAYER_PARAMS
+ yt_query['params'] = self._PLAYER_PARAMS
yt_query.update(self._generate_player_context(sts))
return self._extract_response(
def _get_requested_clients(self, url, smuggled_data):
requested_clients = []
- default = ['android', 'web']
+ default = ['ios', 'android', 'web']
allowed_clients = sorted(
(client for client in INNERTUBE_CLIENTS.keys() if client[:1] != '_'),
key=lambda client: INNERTUBE_CLIENTS[client]['priority'], reverse=True)
def _needs_live_processing(self, live_status, duration):
if (live_status == 'is_live' and self.get_param('live_from_start')
- or live_status == 'post_live' and (duration or 0) > 4 * 3600):
+ or live_status == 'post_live' and (duration or 0) > 2 * 3600):
return live_status
def _extract_formats_and_subtitles(self, streaming_data, video_id, player_url, live_status, duration):
'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres'
])
streaming_formats = traverse_obj(streaming_data, (..., ('formats', 'adaptiveFormats'), ...))
- all_formats = self._configuration_arg('include_duplicate_formats')
+ format_types = self._configuration_arg('formats')
+ all_formats = 'duplicate' in format_types
+ if self._configuration_arg('include_duplicate_formats'):
+ all_formats = True
+ self._downloader.deprecated_feature('[youtube] include_duplicate_formats extractor argument is deprecated. '
+ 'Use formats=duplicate extractor argument instead')
def build_fragments(f):
return LazyList({
f'{video_id}: Some formats are possibly damaged. They will be deprioritized', only_once=True)
client_name = fmt.get(STREAMING_DATA_CLIENT_NAME)
+ name = fmt.get('qualityLabel') or quality.replace('audio_quality_', '') or ''
+ fps = int_or_none(fmt.get('fps')) or 0
dct = {
'asr': int_or_none(fmt.get('audioSampleRate')),
'filesize': int_or_none(fmt.get('contentLength')),
'format_note': join_nonempty(
join_nonempty(audio_track.get('displayName'),
language_preference > 0 and ' (default)', delim=''),
- fmt.get('qualityLabel') or quality.replace('audio_quality_', ''),
- fmt.get('isDrc') and 'DRC',
+ 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',
(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
- 'source_preference': -10 if throttled else -5 if itag == '22' else -1,
- 'fps': int_or_none(fmt.get('fps')) or None,
+ 'source_preference': ((-10 if throttled else -5 if itag == '22' else -1)
+ + (100 if 'Premium' in name else 0)),
+ 'fps': fps if fps > 1 else None, # For some formats, fps is wrongly returned as 1
'audio_channels': fmt.get('audioChannels'),
'height': height,
'quality': q(quality) - bool(fmt.get('isDrc')) / 2,
if single_stream and dct.get('ext'):
dct['container'] = dct['ext'] + '_dash'
- if all_formats and dct['filesize']:
+ if (all_formats or 'dashy' in format_types) and dct['filesize']:
yield {
**dct,
'format_id': f'{dct["format_id"]}-dashy' if all_formats else dct['format_id'],
'protocol': 'http_dash_segments',
'fragments': build_fragments(dct),
}
- dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
- yield dct
+ if all_formats or 'dashy' not in format_types:
+ dct['downloader_options'] = {'http_chunk_size': CHUNK_SIZE}
+ yield dct
needs_live_processing = self._needs_live_processing(live_status, duration)
- skip_bad_formats = not self._configuration_arg('include_incomplete_formats')
+ skip_bad_formats = 'incomplete' not in format_types
+ if self._configuration_arg('include_incomplete_formats'):
+ skip_bad_formats = False
+ self._downloader.deprecated_feature('[youtube] include_incomplete_formats extractor argument is deprecated. '
+ 'Use formats=incomplete extractor argument instead')
skip_manifests = set(self._configuration_arg('skip'))
if (not self.get_param('youtube_include_hls_manifest', True)
skip_manifests.add('dash')
if self._configuration_arg('include_live_dash'):
self._downloader.deprecated_feature('[youtube] include_live_dash extractor argument is deprecated. '
- 'Use include_incomplete_formats extractor argument instead')
+ 'Use formats=incomplete extractor argument instead')
elif skip_bad_formats and live_status == 'is_live' and needs_live_processing != 'is_live':
skip_manifests.add('dash')
elif itag:
f['format_id'] = itag
+ if itag in ('616', '235'):
+ f['format_note'] = join_nonempty(f.get('format_note'), 'Premium', delim=' ')
+ f['source_preference'] = (f.get('source_preference') or -1) + 100
+
f['quality'] = q(itag_qualities.get(try_get(f, lambda f: f['format_id'].split('-')[0]), -1))
if f['quality'] == -1 and f.get('height'):
f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
- if self.get_param('verbose'):
+ if self.get_param('verbose') or all_formats:
f['format_note'] = join_nonempty(f.get('format_note'), client_name, delim=', ')
+ if f.get('fps') and f['fps'] <= 1:
+ del f['fps']
return True
subtitles = {}
webpage = None
if 'webpage' not in self._configuration_arg('player_skip'):
query = {'bpctr': '9999999999', 'has_verified': '1'}
- if smuggled_data.get('is_story'):
- query['pp'] = self._STORY_PLAYER_PARAMS
+ if smuggled_data.get('is_story'): # XXX: Deprecated
+ query['pp'] = self._PLAYER_PARAMS
webpage = self._download_webpage(
webpage_url, video_id, fatal=False, query=query)
for fmt in filter(is_bad_format, formats):
fmt['preference'] = (fmt.get('preference') or -1) - 10
- fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 4 hours)', delim=' ')
+ fmt['format_note'] = join_nonempty(fmt.get('format_note'), '(Last 2 hours)', delim=' ')
if needs_live_processing:
self._prepare_live_from_start_formats(
continue
trans_code += f'-{lang_code}'
trans_name += format_field(lang_name, None, ' from %s')
- # Add an "-orig" label to the original language so that it can be distinguished.
- # The subs are returned without "-orig" as well for compatibility
if lang_code == f'a-{orig_trans_code}':
+ # Set audio language based on original subtitles
+ for f in formats:
+ if f.get('acodec') != 'none' and not f.get('language'):
+ f['language'] = orig_trans_code
+ # Add an "-orig" label to the original language so that it can be distinguished.
+ # The subs are returned without "-orig" as well for compatibility
process_language(
automatic_captions, base_url, f'{trans_code}-orig', f'{trans_name} (Original)', {})
# Setting tlang=lang returns damaged subtitles.
info[d_k] = parse_duration(query[k][0])
# Youtube Music Auto-generated description
- if video_description:
+ if (video_description or '').strip().endswith('\nAuto-generated by YouTube.'):
+ # XXX: Causes catastrophic backtracking if description has "·"
+ # E.g. https://www.youtube.com/watch?v=DoPaAxMQoiI
+ # Simulating atomic groups: (?P<a>[^xy]+)x => (?=(?P<a>[^xy]+))(?P=a)x
+ # reduces it, but does not fully fix it. https://regex101.com/r/8Ssf2h/2
mobj = re.search(
r'''(?xs)
- (?P<track>[^·\n]+)·(?P<artist>[^\n]+)\n+
- (?P<album>[^\n]+)
+ (?=(?P<track>[^\n·]+))(?P=track)·
+ (?=(?P<artist>[^\n]+))(?P=artist)\n+
+ (?=(?P<album>[^\n]+))(?P=album)\n
(?:.+?℗\s*(?P<release_year>\d{4})(?!\d))?
(?:.+?Released on\s*:\s*(?P<release_date>\d{4}-\d{2}-\d{2}))?
- (.+?\nArtist\s*:\s*(?P<clean_artist>[^\n]+))?
- .+\nAuto-generated\ by\ YouTube\.\s*$
+ (.+?\nArtist\s*:\s*
+ (?=(?P<clean_artist>[^\n]+))(?P=clean_artist)\n
+ )?.+\nAuto-generated\ by\ YouTube\.\s*$
''', video_description)
if mobj:
release_year = mobj.group('release_year')
and 'no-youtube-prefer-utc-upload-date' not in self.get_param('compat_opts', [])
):
upload_date = strftime_or_none(
- self._parse_time_text(self._get_text(vpir, 'dateText')), '%Y%m%d') or upload_date
+ self._parse_time_text(self._get_text(vpir, 'dateText'))) or upload_date
info['upload_date'] = upload_date
for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]:
'videoRenderer': lambda x: [self._video_entry(x)],
'playlistRenderer': lambda x: self._grid_entries({'items': [{'playlistRenderer': x}]}),
'channelRenderer': lambda x: self._grid_entries({'items': [{'channelRenderer': x}]}),
- 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)]
+ 'hashtagTileRenderer': lambda x: [self._hashtag_tile_entry(x)],
+ 'richGridRenderer': lambda x: self._extract_entries(x, continuation_list),
}
for key, renderer in isr_content.items():
if key not in known_renderers:
last_updated_unix = self._parse_time_text(
self._get_text(playlist_stats, 2) # deprecated, remove when old layout discontinued
or self._get_text(playlist_header_renderer, ('byline', 1, 'playlistBylineRenderer', 'text')))
- info['modified_date'] = strftime_or_none(last_updated_unix, '%Y%m%d')
+ info['modified_date'] = strftime_or_none(last_updated_unix)
info['view_count'] = self._get_count(playlist_stats, 1)
if info['view_count'] is None: # 0 is allowed
'channel_is_verified': True,
},
'playlist_mincount': 10,
+ }, {
+ # Playlist with only shorts, shown as reel renderers
+ # FIXME: future: YouTube currently doesn't give continuation for this,
+ # may do in future.
+ 'url': 'https://www.youtube.com/playlist?list=UUxqPAgubo4coVn9Lx1FuKcg',
+ 'info_dict': {
+ 'id': 'UUxqPAgubo4coVn9Lx1FuKcg',
+ 'channel_url': 'https://www.youtube.com/channel/UCxqPAgubo4coVn9Lx1FuKcg',
+ 'view_count': int,
+ 'uploader_id': '@BangyShorts',
+ 'description': '',
+ 'uploader_url': 'https://www.youtube.com/@BangyShorts',
+ 'channel_id': 'UCxqPAgubo4coVn9Lx1FuKcg',
+ 'channel': 'Bangy Shorts',
+ 'uploader': 'Bangy Shorts',
+ 'tags': [],
+ 'availability': 'public',
+ 'modified_date': '20230626',
+ 'title': 'Uploads from Bangy Shorts',
+ },
+ 'playlist_mincount': 100,
+ 'expected_warnings': [r'[Uu]navailable videos (are|will be) hidden'],
}]
@classmethod
projects / yt-dlp.git / blobdiff