AVAILABILITY_PREMIUM = enum.auto()
AVAILABILITY_SUBSCRIPTION = enum.auto()
LIVE_NOW = enum.auto()
+ VERIFIED = enum.auto()
class YoutubeBaseInfoExtractor(InfoExtractor):
def _extract_and_report_alerts(self, data, *args, **kwargs):
return self._report_alerts(self._extract_alerts(data), *args, **kwargs)
- def _extract_badges(self, renderer: dict):
- privacy_icon_map = {
+ def _extract_badges(self, badge_list: list):
+ """
+ Extract known BadgeType's from a list of badge renderers.
+ @returns [{'type': BadgeType}]
+ """
+ icon_type_map = {
'PRIVACY_UNLISTED': BadgeType.AVAILABILITY_UNLISTED,
'PRIVACY_PRIVATE': BadgeType.AVAILABILITY_PRIVATE,
- 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC
+ 'PRIVACY_PUBLIC': BadgeType.AVAILABILITY_PUBLIC,
+ 'CHECK_CIRCLE_THICK': BadgeType.VERIFIED,
+ 'OFFICIAL_ARTIST_BADGE': BadgeType.VERIFIED,
+ 'CHECK': BadgeType.VERIFIED,
}
badge_style_map = {
'BADGE_STYLE_TYPE_MEMBERS_ONLY': BadgeType.AVAILABILITY_SUBSCRIPTION,
'BADGE_STYLE_TYPE_PREMIUM': BadgeType.AVAILABILITY_PREMIUM,
- 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW
+ 'BADGE_STYLE_TYPE_LIVE_NOW': BadgeType.LIVE_NOW,
+ 'BADGE_STYLE_TYPE_VERIFIED': BadgeType.VERIFIED,
+ 'BADGE_STYLE_TYPE_VERIFIED_ARTIST': BadgeType.VERIFIED
}
label_map = {
'private': BadgeType.AVAILABILITY_PRIVATE,
'members only': BadgeType.AVAILABILITY_SUBSCRIPTION,
'live': BadgeType.LIVE_NOW,
- 'premium': BadgeType.AVAILABILITY_PREMIUM
+ 'premium': BadgeType.AVAILABILITY_PREMIUM,
+ 'verified': BadgeType.VERIFIED,
+ 'official artist channel': BadgeType.VERIFIED
}
badges = []
- for badge in traverse_obj(renderer, ('badges', ..., 'metadataBadgeRenderer')):
+ for badge in traverse_obj(badge_list, (..., lambda key, _: re.search(r'[bB]adgeRenderer$', key))):
badge_type = (
- privacy_icon_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
+ icon_type_map.get(traverse_obj(badge, ('icon', 'iconType'), expected_type=str))
or badge_style_map.get(traverse_obj(badge, 'style'))
)
if badge_type:
continue
# fallback, won't work in some languages
- label = traverse_obj(badge, 'label', expected_type=str, default='')
+ label = traverse_obj(
+ badge, 'label', ('accessibilityData', 'label'), 'tooltip', 'iconTooltip', get_all=False, expected_type=str, default='')
for match, label_badge_type in label_map.items():
if match in label.lower():
- badges.append({'type': badge_type})
- continue
+ badges.append({'type': label_badge_type})
+ break
return badges
overlay_style = traverse_obj(
renderer, ('thumbnailOverlays', ..., 'thumbnailOverlayTimeStatusRenderer', 'style'),
get_all=False, expected_type=str)
- badges = self._extract_badges(renderer)
-
+ badges = self._extract_badges(traverse_obj(renderer, 'badges'))
+ owner_badges = self._extract_badges(traverse_obj(renderer, 'ownerBadges'))
navigation_url = urljoin('https://www.youtube.com/', traverse_obj(
renderer, ('navigationEndpoint', 'commandMetadata', 'webCommandMetadata', 'url'),
expected_type=str)) or ''
needs_subscription=self._has_badge(badges, BadgeType.AVAILABILITY_SUBSCRIPTION) or None,
is_unlisted=self._has_badge(badges, BadgeType.AVAILABILITY_UNLISTED) or None),
view_count_field: view_count,
- 'live_status': live_status
+ 'live_status': live_status,
+ 'channel_is_verified': True if self._has_badge(owner_badges, BadgeType.VERIFIED) else None
}
'uploader': 'Philipp Hagemeister',
'uploader_url': 'https://www.youtube.com/@PhilippHagemeister',
'uploader_id': '@PhilippHagemeister',
+ 'heatmap': 'count:100',
},
'params': {
'skip_download': True,
'uploader': 'The Witcher',
'uploader_url': 'https://www.youtube.com/@thewitcher',
'uploader_id': '@thewitcher',
+ 'comment_count': int,
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
},
},
{
'uploader_url': 'https://www.youtube.com/@FlyingKitty900',
'uploader_id': '@FlyingKitty900',
'comment_count': int,
+ 'channel_is_verified': True,
},
},
{
'uploader': 'Olympics',
'uploader_url': 'https://www.youtube.com/@Olympics',
'uploader_id': '@Olympics',
+ 'channel_is_verified': True,
},
'params': {
'skip_download': 'requires avconv',
'uploader': 'Bernie Sanders',
'uploader_url': 'https://www.youtube.com/@BernieSanders',
'uploader_id': '@BernieSanders',
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
},
'params': {
'skip_download': True,
'uploader': 'Vsauce',
'uploader_url': 'https://www.youtube.com/@Vsauce',
'uploader_id': '@Vsauce',
+ 'comment_count': int,
+ 'channel_is_verified': True,
},
'params': {
'skip_download': True,
'uploader': 'kudvenkat',
'uploader_url': 'https://www.youtube.com/@Csharp-video-tutorialsBlogspot',
'uploader_id': '@Csharp-video-tutorialsBlogspot',
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
},
'params': {
'skip_download': True,
'uploader': 'CBS Mornings',
'uploader_url': 'https://www.youtube.com/@CBSMornings',
'uploader_id': '@CBSMornings',
+ 'comment_count': int,
+ 'channel_is_verified': True,
}
},
{
'uploader': 'colinfurze',
'uploader_url': 'https://www.youtube.com/@colinfurze',
'uploader_id': '@colinfurze',
+ 'comment_count': int,
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
},
'params': {
'format': '17', # 3gp format available on android
'uploader': 'SciShow',
'uploader_url': 'https://www.youtube.com/@SciShow',
'uploader_id': '@SciShow',
+ 'comment_count': int,
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
}, 'params': {'format': 'mhtml', 'skip_download': True}
}, {
# Ensure video upload_date is in UTC timezone (video was uploaded 1641170939)
'uploader': 'Leon Nguyen',
'uploader_url': 'https://www.youtube.com/@LeonNguyen',
'uploader_id': '@LeonNguyen',
+ 'heatmap': 'count:100',
}
}, {
# Same video as above, but with --compat-opt no-youtube-prefer-utc-upload-date
'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']}
}, {
'uploader': 'Quackity',
'uploader_id': '@Quackity',
'uploader_url': 'https://www.youtube.com/@Quackity',
+ 'comment_count': int,
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
}
},
{ # continuous livestream. Microformat upload date should be preferred.
'uploader': 'MrBeast',
'uploader_url': 'https://www.youtube.com/@MrBeast',
'uploader_id': '@MrBeast',
+ 'comment_count': int,
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
},
'params': {'extractor_args': {'youtube': {'player_client': ['ios']}}, 'format': '233-1'},
}, {
'uploader': 'さなちゃんねる',
'uploader_url': 'https://www.youtube.com/@sana_natori',
'uploader_id': '@sana_natori',
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
},
},
{
'thumbnail': r're:^https?://.*\.webp',
'channel_url': 'https://www.youtube.com/channel/UCxzC4EngIsMrPmbm6Nxvb-A',
'playable_in_embed': True,
+ 'comment_count': int,
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
},
'params': {
'extractor_args': {'youtube': {'player_client': ['android'], 'player_skip': ['webpage']}},
'uploader': 'Christopher Sykes',
'uploader_url': 'https://www.youtube.com/@ChristopherSykesDocumentaries',
'uploader_id': '@ChristopherSykesDocumentaries',
+ 'heatmap': 'count:100',
},
'params': {
'skip_download': True,
if comment_abr is not None:
info['is_favorited'] = 'creatorHeart' in comment_abr
- comment_ab_icontype = traverse_obj(
- comment_renderer, ('authorCommentBadge', 'authorCommentBadgeRenderer', 'icon', 'iconType'))
- if comment_ab_icontype is not None:
- info['author_is_verified'] = comment_ab_icontype in ('CHECK_CIRCLE_THICK', 'OFFICIAL_ARTIST_BADGE')
+ badges = self._extract_badges([traverse_obj(comment_renderer, 'authorCommentBadge')])
+ if self._has_badge(badges, BadgeType.VERIFIED):
+ info['author_is_verified'] = True
is_pinned = traverse_obj(comment_renderer, 'pinnedCommentBadge')
if is_pinned:
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,
f['quality'] = q(res_qualities[min(res_qualities, key=lambda x: abs(x - f['height']))])
if self.get_param('verbose'):
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 = {}
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['artist'] = mrr_contents_text
elif mrr_title == 'Song':
info['track'] = mrr_contents_text
+ owner_badges = self._extract_badges(traverse_obj(vsir, ('owner', 'videoOwnerRenderer', 'badges')))
+ if self._has_badge(owner_badges, BadgeType.VERIFIED):
+ info['channel_is_verified'] = True
info.update({
'uploader': info.get('channel'),
if v:
info[d_k] = v
- badges = self._extract_badges(traverse_obj(contents, (..., 'videoPrimaryInfoRenderer'), get_all=False))
+ badges = self._extract_badges(traverse_obj(vpir, 'badges'))
is_private = (self._has_badge(badges, BadgeType.AVAILABILITY_PRIVATE)
or get_first(video_details, 'isPrivate', expected_type=bool))
channel_id = self.ucid_or_none(renderer['channelId'])
title = self._get_text(renderer, 'title')
channel_url = format_field(channel_id, None, 'https://www.youtube.com/channel/%s', default=None)
- # As of 2023-03-01 YouTube doesn't use the channel handles on these renderers yet.
- # However we can expect them to change that in the future.
channel_handle = self.handle_from_url(
traverse_obj(renderer, (
'navigationEndpoint', (('commandMetadata', 'webCommandMetadata', 'url'),
('browseEndpoint', 'canonicalBaseUrl')),
{str}), get_all=False))
+ if not channel_handle:
+ # As of 2023-06-01, YouTube sets subscriberCountText to the handle in search
+ channel_handle = self.handle_or_none(self._get_text(renderer, 'subscriberCountText'))
return {
'_type': 'url',
'url': channel_url,
'title': title,
'uploader_id': channel_handle,
'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
- 'channel_follower_count': self._get_count(renderer, 'subscriberCountText'),
+ # See above. YouTube sets videoCountText to the subscriber text in search channel renderers.
+ # However, in feed/channels this is set correctly to the subscriber count
+ 'channel_follower_count': traverse_obj(
+ renderer, 'subscriberCountText', 'videoCountText', expected_type=self._get_count),
'thumbnails': self._extract_thumbnails(renderer, 'thumbnail'),
- 'playlist_count': self._get_count(renderer, 'videoCountText'),
+ 'playlist_count': (
+ # videoCountText may be the subscriber count
+ self._get_count(renderer, 'videoCountText')
+ if self._get_count(renderer, 'subscriberCountText') is not None else None),
'description': self._get_text(renderer, 'descriptionSnippet'),
+ 'channel_is_verified': True if self._has_badge(
+ self._extract_badges(traverse_obj(renderer, 'ownerBadges')), BadgeType.VERIFIED) else None,
}
def _grid_entries(self, grid_renderer):
'uploader_id': channel_handle,
'uploader_url': format_field(channel_handle, None, 'https://www.youtube.com/%s', default=None),
})
+
+ channel_badges = self._extract_badges(traverse_obj(data, ('header', ..., 'badges'), get_all=False))
+ if self._has_badge(channel_badges, BadgeType.VERIFIED):
+ info['channel_is_verified'] = True
# Playlist stats is a text runs array containing [video count, view count, last updated].
# last updated or (view count and last updated) may be missing.
playlist_stats = get_first(
playlist_header_renderer = traverse_obj(data, ('header', 'playlistHeaderRenderer')) or {}
player_header_privacy = playlist_header_renderer.get('privacy')
- badges = self._extract_badges(sidebar_renderer)
+ badges = self._extract_badges(traverse_obj(sidebar_renderer, 'badges'))
# Personal playlists, when authenticated, have a dropdown visibility selector instead of a badge
privacy_setting_icon = get_first(
'uploader_url': 'https://www.youtube.com/@3blue1brown',
'uploader': '3Blue1Brown',
'tags': ['Mathematics'],
- 'channel_follower_count': int
+ 'channel_follower_count': int,
+ 'channel_is_verified': True,
},
}, {
'note': 'playlists, singlepage',
'uploader_url': 'https://www.youtube.com/@3blue1brown',
'uploader_id': '@3blue1brown',
'uploader': '3Blue1Brown',
+ 'channel_is_verified': True,
},
}, {
'url': 'https://invidio.us/channel/UCmlqkdCBesrv2Lak1mF_MxA',
}, {
'url': 'https://www.youtube.com/channel/UCoMdktPbSTixAyNGwb-UYkQ/live',
'info_dict': {
- 'id': 'AlTsmyW4auo', # This will keep changing
+ 'id': 'hGkQjiJLjWQ', # This will keep changing
'ext': 'mp4',
'title': str,
'upload_date': r're:\d{8}',
'uploader_url': 'https://www.youtube.com/@SkyNews',
'uploader_id': '@SkyNews',
'uploader': 'Sky News',
+ 'channel_is_verified': True,
},
'params': {
'skip_download': True,
'channel': str,
'uploader': str,
'uploader_url': str,
- 'uploader_id': str
+ 'uploader_id': str,
+ 'channel_is_verified': bool, # this will keep changing
}
}],
'params': {'extract_flat': True, 'playlist_items': '1'},
'uploader': 'PewDiePie',
'uploader_url': 'https://www.youtube.com/@PewDiePie',
'uploader_id': '@PewDiePie',
+ 'channel_is_verified': True,
}
}],
'params': {'extract_flat': True},
'uploader_url': 'https://www.youtube.com/@3blue1brown',
'uploader_id': '@3blue1brown',
'uploader': '3Blue1Brown',
+ 'channel_is_verified': True,
},
'playlist_count': 0,
}, {
'description': 'I make music',
'channel_url': 'https://www.youtube.com/channel/UCgFwu-j5-xNJml2FtTrrB3A',
'channel_follower_count': int,
+ 'channel_is_verified': True,
},
'playlist_mincount': 10,
}]
'description': 'md5:4ae48dfa9505ffc307dad26342d06bfc',
'title': 'Kurzgesagt – In a Nutshell',
'channel_id': 'UCsXVk37bltHxD1rDPwtNM8Q',
- 'playlist_count': int, # XXX: should have a way of saying > 1
+ # No longer available for search as it is set to the handle.
+ # 'playlist_count': int,
'channel_url': 'https://www.youtube.com/channel/UCsXVk37bltHxD1rDPwtNM8Q',
'thumbnails': list,
'uploader_id': '@kurzgesagt',
'uploader_url': 'https://www.youtube.com/@kurzgesagt',
'uploader': 'Kurzgesagt – In a Nutshell',
+ 'channel_is_verified': True,
+ 'channel_follower_count': int,
}
}],
'params': {'extract_flat': True, 'playlist_items': '1'},
'live_status': 'not_live',
'channel_follower_count': int,
'chapters': 'count:20',
+ 'comment_count': int,
+ 'heatmap': 'count:100',
}
}]
'channel': 'さなちゃんねる',
'description': 'md5:6aebf95cc4a1d731aebc01ad6cc9806d',
'uploader': 'さなちゃんねる',
+ 'channel_is_verified': True,
+ 'heatmap': 'count:100',
},
'add_ie': ['Youtube'],
'params': {'skip_download': 'Youtube'},