aboutsummaryrefslogtreecommitdiffstats
path: root/youtube_dl/extractor
diff options
context:
space:
mode:
authorJames Taylor <user234683@users.noreply.github.com>2019-08-09 22:01:04 -0700
committerJames Taylor <user234683@users.noreply.github.com>2019-08-09 22:01:04 -0700
commit2e75c6d9603f8a5edf6495f8d4fb3115a67d823c (patch)
tree8fb2d1bec2cf0e50c5fce6bc718f755485419db0 /youtube_dl/extractor
parentcc9283ad5332f59a69a91d9d0fab299779de513c (diff)
parentadc40bc760345a23678a01f27d7697dfd3811914 (diff)
downloadyt-local-2e75c6d9603f8a5edf6495f8d4fb3115a67d823c.tar.lz
yt-local-2e75c6d9603f8a5edf6495f8d4fb3115a67d823c.tar.xz
yt-local-2e75c6d9603f8a5edf6495f8d4fb3115a67d823c.zip
Merge flask framework and other stuff from master
Diffstat (limited to 'youtube_dl/extractor')
-rw-r--r--youtube_dl/extractor/youtube.py339
-rw-r--r--youtube_dl/extractor/youtube_unmodified_reference.py334
2 files changed, 502 insertions, 171 deletions
diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py
index 52c8731..18b240a 100644
--- a/youtube_dl/extractor/youtube.py
+++ b/youtube_dl/extractor/youtube.py
@@ -17,6 +17,7 @@ from ..jsinterp import JSInterpreter
from ..swfinterp import SWFInterpreter
from ..compat import (
compat_chr,
+ compat_HTTPError,
compat_kwargs,
compat_parse_qs,
compat_urllib_parse_unquote,
@@ -28,6 +29,7 @@ from ..compat import (
)
from ..utils import (
clean_html,
+ dict_get,
error_to_compat_str,
ExtractorError,
float_or_none,
@@ -289,10 +291,25 @@ class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
if not mobj:
break
- more = self._download_json(
- 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
- 'Downloading page #%s' % page_num,
- transform_source=uppercase_escape)
+ count = 0
+ retries = 3
+ while count <= retries:
+ try:
+ # Downloading page may result in intermittent 5xx HTTP error
+ # that is usually worked around with a retry
+ more = self._download_json(
+ 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
+ 'Downloading page #%s%s'
+ % (page_num, ' (retry #%d)' % count if count else ''),
+ transform_source=uppercase_escape)
+ break
+ except ExtractorError as e:
+ if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
+ count += 1
+ if count <= retries:
+ continue
+ raise
+
content_html = more['content_html']
if not content_html.strip():
# Some webpages show a "Load more" button but they don't
@@ -353,7 +370,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
(?:www\.)?hooktube\.com/|
(?:www\.)?yourepeat\.com/|
tube\.majestyc\.net/|
- (?:www\.)?invidio\.us/|
+ (?:(?:www|dev)\.)?invidio\.us/|
+ (?:www\.)?invidiou\.sh/|
+ (?:www\.)?invidious\.snopyta\.org/|
+ (?:www\.)?invidious\.kabi\.tk/|
+ (?:www\.)?vid\.wxzm\.sx/|
youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
(?:.*?\#/)? # handle anchor (#/) redirect urls
(?: # the various things that can precede the ID:
@@ -429,7 +450,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
'136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
'137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
- '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
+ '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
'160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
'212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
'264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
@@ -481,8 +502,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# RTMP (unnamed)
'_rtmp': {'protocol': 'rtmp'},
+
+ # av01 video only formats sometimes served with "unknown" codecs
+ '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
+ '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
+ '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
+ '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
}
- _SUBTITLE_FORMATS = ('ttml', 'vtt')
+ _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
_GEO_BYPASS = False
@@ -694,7 +721,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'age_limit': 18,
},
},
- # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
+ # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
# YouTube Red ad is not captured for creator
{
'url': '__2ABJjxzNo',
@@ -715,7 +742,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'DASH manifest missing',
]
},
- # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
+ # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
{
'url': 'lqQg6PlCWgI',
'info_dict': {
@@ -766,7 +793,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
},
'skip': 'This live event has ended.',
},
- # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
+ # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
{
'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
'info_dict': {
@@ -869,7 +896,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'skip': 'This video is not available.',
},
{
- # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
+ # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
'info_dict': {
'id': 'gVfLd0zydlo',
@@ -887,10 +914,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'only_matching': True,
},
{
- # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
+ # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
# Also tests cut-off URL expansion in video description (see
- # https://github.com/rg3/youtube-dl/issues/1892,
- # https://github.com/rg3/youtube-dl/issues/8164)
+ # https://github.com/ytdl-org/youtube-dl/issues/1892,
+ # https://github.com/ytdl-org/youtube-dl/issues/8164)
'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
'info_dict': {
'id': 'lsguqyKfVQg',
@@ -906,13 +933,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
'track': 'Dark Walk - Position Music',
'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
+ 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
},
'params': {
'skip_download': True,
},
},
{
- # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
+ # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
'only_matching': True,
},
@@ -976,7 +1004,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'only_matching': True,
},
{
- # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
+ # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
'only_matching': True,
},
@@ -1084,7 +1112,95 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'skip_download': True,
'youtube_include_dash_manifest': False,
},
- }
+ },
+ {
+ # Youtube Music Auto-generated description
+ 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
+ 'info_dict': {
+ 'id': 'MgNrAu2pzNs',
+ 'ext': 'mp4',
+ 'title': 'Voyeur Girl',
+ 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
+ 'upload_date': '20190312',
+ 'uploader': 'Various Artists - Topic',
+ 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
+ 'artist': 'Stephen',
+ 'track': 'Voyeur Girl',
+ 'album': 'it\'s too much love to know my dear',
+ 'release_date': '20190313',
+ 'release_year': 2019,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ # Retrieve 'artist' field from 'Artist:' in video description
+ # when it is present on youtube music video
+ 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
+ 'info_dict': {
+ 'id': 'k0jLE7tTwjY',
+ 'ext': 'mp4',
+ 'title': 'Latch Feat. Sam Smith',
+ 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
+ 'upload_date': '20150110',
+ 'uploader': 'Various Artists - Topic',
+ 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
+ 'artist': 'Disclosure',
+ 'track': 'Latch Feat. Sam Smith',
+ 'album': 'Latch Featuring Sam Smith',
+ 'release_date': '20121008',
+ 'release_year': 2012,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ # handle multiple artists on youtube music video
+ 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
+ 'info_dict': {
+ 'id': '74qn0eJSjpA',
+ 'ext': 'mp4',
+ 'title': 'Eastside',
+ 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
+ 'upload_date': '20180710',
+ 'uploader': 'Benny Blanco - Topic',
+ 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
+ 'artist': 'benny blanco, Halsey, Khalid',
+ 'track': 'Eastside',
+ 'album': 'Eastside',
+ 'release_date': '20180713',
+ 'release_year': 2018,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ # handle youtube music video with release_year and no release_date
+ 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
+ 'info_dict': {
+ 'id': '-hcAI0g-f5M',
+ 'ext': 'mp4',
+ 'title': 'Put It On Me',
+ 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
+ 'upload_date': '20180426',
+ 'uploader': 'Matt Maeson - Topic',
+ 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
+ 'artist': 'Matt Maeson',
+ 'track': 'Put It On Me',
+ 'album': 'The Hearse',
+ 'release_date': None,
+ 'release_year': 2018,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
]
def __init__(self, *args, **kwargs):
@@ -1198,11 +1314,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _parse_sig_js(self, jscode):
funcname = self._search_regex(
- (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
+ # Obsolete patterns
+ r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
- r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?(?P<sig>[a-zA-Z0-9$]+)\(',
- r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
- r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
+ r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
jscode, 'Initial JS player signature function name', group='sig')
jsi = JSInterpreter(jscode)
@@ -1282,8 +1405,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# regex won't capture the whole JSON. Yet working around by trying more
# concrete regex first keeping in mind proper quoted string handling
# to be implemented in future that will replace this workaround (see
- # https://github.com/rg3/youtube-dl/issues/7468,
- # https://github.com/rg3/youtube-dl/pull/7599)
+ # https://github.com/ytdl-org/youtube-dl/issues/7468,
+ # https://github.com/ytdl-org/youtube-dl/pull/7599)
r';ytplayer\.config\s*=\s*({.+?});ytplayer',
r';ytplayer\.config\s*=\s*({.+?});',
)
@@ -1467,8 +1590,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
return video_id
def _extract_annotations(self, video_id):
- url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
- return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
+ return self._download_webpage(
+ 'https://www.youtube.com/annotations_invideo', video_id,
+ note='Downloading annotations',
+ errnote='Unable to download video annotations', fatal=False,
+ query={
+ 'features': 1,
+ 'legacy': 1,
+ 'video_id': video_id,
+ })
@staticmethod
def _extract_chapters(description, duration):
@@ -1563,12 +1693,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def extract_view_count(v_info):
return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
+ def extract_token(v_info):
+ return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
+
player_response = {}
# Is it unlisted?
- unlisted = (self._search_regex('''<meta itemprop="unlisted" content="(\w*)">''', video_webpage, 'is_unlisted', default='False') == "True")
+ unlisted = ('<span id="watch-privacy-icon"' in video_webpage)
# Related videos
related_vid_info = self._search_regex(r"""'RELATED_PLAYER_ARGS':\s*(\{.*?\})""", video_webpage, "related_player_args", default='')
@@ -1623,8 +1756,6 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
-
-
# Get video info
embed_webpage = None
if re.search(r'player-age-gate-content">', video_webpage) is not None:
@@ -1660,7 +1791,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
add_dash_mpd(video_info)
# Rental video is not rented but preview is available (e.g.
# https://www.youtube.com/watch?v=yYr8q0y5Jfg,
- # https://github.com/rg3/youtube-dl/issues/10532)
+ # https://github.com/ytdl-org/youtube-dl/issues/10532)
if not video_info and args.get('ypc_vid'):
return self.url_result(
args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
@@ -1680,9 +1811,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
# manifest pointed by get_video_info's dashmpd).
# The general idea is to take a union of itags of both DASH manifests (for example
- # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
+ # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
self.report_video_info_webpage_download(video_id)
- for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
+ for el in ('embedded', 'detailpage', 'vevo', ''):
query = {
'video_id': video_id,
'ps': 'default',
@@ -1712,18 +1843,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
view_count = extract_view_count(get_video_info)
if not video_info:
video_info = get_video_info
- get_token = get_video_info.get('token') or get_video_info.get('account_playback_token')
+ get_token = extract_token(get_video_info)
if get_token:
# Different get_video_info requests may report different results, e.g.
# some may report video unavailability, but some may serve it without
- # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
+ # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
# the original webpage as well as el=info and el=embedded get_video_info
# requests report video unavailability due to geo restriction while
# el=detailpage succeeds and returns valid data). This is probably
# due to YouTube measures against IP ranges of hosting providers.
# Working around by preferring the first succeeded video_info containing
# the token if no such video_info yet was found.
- token = video_info.get('token') or video_info.get('account_playback_token')
+ token = extract_token(video_info)
if not token:
video_info = get_video_info
break
@@ -1733,40 +1864,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
video_webpage, 'unavailable message', default=None)
- token = video_info.get('token') or video_info.get('account_playback_token')
- if not token:
- if 'reason' in video_info:
- if 'The uploader has not made this video available in your country.' in video_info['reason']:
- regions_allowed = self._html_search_meta(
- 'regionsAllowed', video_webpage, default=None)
- countries = regions_allowed.split(',') if regions_allowed else None
- self.raise_geo_restricted(
- msg=video_info['reason'][0], countries=countries)
- reason = video_info['reason'][0]
- if 'Invalid parameters' in reason:
- unavailable_message = extract_unavailable_message()
- if unavailable_message:
- reason = unavailable_message
- raise YoutubeError(
- 'YouTube said: %s' % reason,
- expected=True, video_id=video_id)
- else:
- raise ExtractorError(
- '"token" parameter not in video info for unknown reason',
- video_id=video_id)
-
- if video_info.get('license_info'):
- raise ExtractorError('This video is DRM protected.', expected=True)
+ if not video_info:
+ unavailable_message = extract_unavailable_message()
+ if not unavailable_message:
+ unavailable_message = 'Unable to extract video data'
+ raise ExtractorError(
+ 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
video_details = try_get(
player_response, lambda x: x['videoDetails'], dict) or {}
- # title
- if 'title' in video_info:
- video_title = video_info['title'][0]
- elif 'title' in player_response:
- video_title = video_details['title']
- else:
+ video_title = video_info.get('title', [None])[0] or video_details.get('title')
+ if not video_title:
self._downloader.report_warning('Unable to extract video title')
video_title = '_'
@@ -1795,11 +1904,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
''', replace_url, video_description)
video_description = clean_html(video_description)
else:
- fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
- if fd_mobj:
- video_description = unescapeHTML(fd_mobj.group(1))
- else:
- video_description = ''
+ video_description = self._html_search_meta('description', video_webpage) or video_details.get('shortDescription')
if not smuggled_data.get('force_singlefeed', False):
if not self._downloader.params.get('noplaylist'):
@@ -1814,7 +1919,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
for feed in multifeed_metadata_list.split(','):
# Unquote should take place before split on comma (,) since textual
# fields may contain comma as well (see
- # https://github.com/rg3/youtube-dl/issues/8536)
+ # https://github.com/ytdl-org/youtube-dl/issues/8536)
feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
entries.append({
'_type': 'url_transparent',
@@ -1839,7 +1944,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# Check for "rental" videos
if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
- raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
+ raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
def _extract_filesize(media_url):
return int_or_none(self._search_regex(
@@ -1856,7 +1961,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
elif not is_live and (len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
if 'rtmpe%3Dyes' in encoded_url_map:
- raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
+ raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
formats_spec = {}
fmt_list = video_info.get('fmt_list', [''])[0]
if fmt_list:
@@ -1893,7 +1998,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
formats = []
for url_data_str in encoded_url_map.split(','):
url_data = compat_parse_qs(url_data_str)
- if 'itag' not in url_data or 'url' not in url_data:
+ if 'itag' not in url_data or 'url' not in url_data or url_data.get('drm_families'):
continue
stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
# Unsupported FORMAT_STREAM_TYPE_OTF
@@ -1953,7 +2058,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
signature = self._decrypt_signature(
encrypted_sig, video_id, player_url, age_gate)
- url += '&signature=' + signature
+ sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
+ url += '&%s=%s' % (sp, signature)
if 'ratebypass' not in url:
url += '&ratebypass=yes'
@@ -1968,7 +2074,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
dct.update(formats_spec[format_id])
# Some itags are not included in DASH manifest thus corresponding formats will
- # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
+ # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
# Trying to extract metadata from url_encoded_fmt_stream_map entry.
mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
@@ -2017,8 +2123,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
url_or_none(try_get(
player_response,
lambda x: x['streamingData']['hlsManifestUrl'],
- compat_str)) or
- url_or_none(try_get(
+ compat_str))
+ or url_or_none(try_get(
video_info, lambda x: x['hlsvp'][0], compat_str)))
if manifest_url:
formats = []
@@ -2038,7 +2144,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True'
formats.append(a_format)
else:
- error_message = extract_unavailable_message()
+ error_message = clean_html(video_info.get('reason', [None])[0])
alt_error_message = clean_html(video_info.get('reason', [None])[0])
print(alt_error_message)
if not error_message:
@@ -2068,8 +2174,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
else:
self._downloader.report_warning('unable to extract uploader nickname')
- channel_id = self._html_search_meta(
- 'channelId', video_webpage, 'channel id')
+ channel_id = (
+ str_or_none(video_details.get('channelId'))
+ or self._html_search_meta(
+ 'channelId', video_webpage, 'channel id', default=None)
+ or self._search_regex(
+ r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
+ video_webpage, 'channel id', default=None, group='id'))
channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
# thumbnail image
@@ -2128,6 +2239,27 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
track = extract_meta('Song')
artist = extract_meta('Artist')
+ album = extract_meta('Album')
+
+ # Youtube Music Auto-generated description
+ release_date = release_year = None
+ if video_description:
+ mobj = re.search(r'(?s)Provided to YouTube by [^\n]+\n+(?P<track>[^·]+)·(?P<artist>[^\n]+)\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]+))?', video_description)
+ if mobj:
+ if not track:
+ track = mobj.group('track').strip()
+ if not artist:
+ artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
+ if not album:
+ album = mobj.group('album'.strip())
+ release_year = mobj.group('release_year')
+ release_date = mobj.group('release_date')
+ if release_date:
+ release_date = release_date.replace('-', '')
+ if not release_year:
+ release_year = int(release_date[:4])
+ if release_year:
+ release_year = int(release_year)
m_episode = re.search(
r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
@@ -2168,6 +2300,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
'view count', default=None))
+ average_rating = (
+ float_or_none(video_details.get('averageRating'))
+ or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
+
# subtitles
video_subtitles = self._get_subtitles(video_id, video_webpage)
automatic_captions = self._get_automatic_captions(video_id, video_webpage)
@@ -2221,7 +2357,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# Remove the formats we found through non-DASH, they
# contain less info and it can be wrong, because we use
# fixed values (for example the resolution). See
- # https://github.com/rg3/youtube-dl/issues/5774 for an
+ # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
# example.
formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
formats.extend(dash_formats.values())
@@ -2241,6 +2377,32 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
if f.get('vcodec') != 'none':
f['stretched_ratio'] = ratio
+ if not formats:
+ token = extract_token(video_info)
+ if not token:
+ if 'reason' in video_info:
+ if 'The uploader has not made this video available in your country.' in video_info['reason']:
+ regions_allowed = self._html_search_meta(
+ 'regionsAllowed', video_webpage, default=None)
+ countries = regions_allowed.split(',') if regions_allowed else None
+ self.raise_geo_restricted(
+ msg=video_info['reason'][0], countries=countries)
+ reason = video_info['reason'][0]
+ if 'Invalid parameters' in reason:
+ unavailable_message = extract_unavailable_message()
+ if unavailable_message:
+ reason = unavailable_message
+ raise YoutubeError(
+ 'YouTube said: %s' % reason,
+ expected=True, video_id=video_id)
+ else:
+ raise ExtractorError(
+ '"token" parameter not in video info for unknown reason',
+ video_id=video_id)
+
+ if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
+ raise ExtractorError('This video is DRM protected.', expected=True)
+
self._sort_formats(formats)
self.mark_watched(video_id, video_info, player_response)
@@ -2271,7 +2433,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'view_count': view_count,
'like_count': like_count,
'dislike_count': dislike_count,
- 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
+ 'average_rating': average_rating,
'formats': formats,
'is_live': is_live,
'start_time': start_time,
@@ -2281,6 +2443,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'episode_number': episode_number,
'track': track,
'artist': artist,
+ 'album': album,
+ 'release_date': release_date,
+ 'release_year': release_year,
'related_vids': related_vids,
'music_list': music_list,
'unlisted': unlisted,
@@ -2482,9 +2647,9 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
title_span = (
- search_title('playlist-title') or
- search_title('title long-title') or
- search_title('title'))
+ search_title('playlist-title')
+ or search_title('title long-title')
+ or search_title('title'))
title = clean_html(title_span)
return self.playlist_result(url_results, playlist_id, title)
@@ -2493,7 +2658,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
url = self._TEMPLATE_URL % playlist_id
page = self._download_webpage(url, playlist_id)
- # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
+ # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
match = match.strip()
# Check if the playlist exists or is private
@@ -2586,7 +2751,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
return playlist
# Some playlist URLs don't actually serve a playlist (see
- # https://github.com/rg3/youtube-dl/issues/10537).
+ # https://github.com/ytdl-org/youtube-dl/issues/10537).
# Fallback to plain video extraction if there is a video id
# along with playlist id.
return self.url_result(video_id, 'Youtube', video_id=video_id)
diff --git a/youtube_dl/extractor/youtube_unmodified_reference.py b/youtube_dl/extractor/youtube_unmodified_reference.py
index c12c417..f76a6e7 100644
--- a/youtube_dl/extractor/youtube_unmodified_reference.py
+++ b/youtube_dl/extractor/youtube_unmodified_reference.py
@@ -16,6 +16,7 @@ from ..jsinterp import JSInterpreter
from ..swfinterp import SWFInterpreter
from ..compat import (
compat_chr,
+ compat_HTTPError,
compat_kwargs,
compat_parse_qs,
compat_urllib_parse_unquote,
@@ -27,6 +28,7 @@ from ..compat import (
)
from ..utils import (
clean_html,
+ dict_get,
error_to_compat_str,
ExtractorError,
float_or_none,
@@ -287,10 +289,25 @@ class YoutubeEntryListBaseInfoExtractor(YoutubeBaseInfoExtractor):
if not mobj:
break
- more = self._download_json(
- 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
- 'Downloading page #%s' % page_num,
- transform_source=uppercase_escape)
+ count = 0
+ retries = 3
+ while count <= retries:
+ try:
+ # Downloading page may result in intermittent 5xx HTTP error
+ # that is usually worked around with a retry
+ more = self._download_json(
+ 'https://youtube.com/%s' % mobj.group('more'), playlist_id,
+ 'Downloading page #%s%s'
+ % (page_num, ' (retry #%d)' % count if count else ''),
+ transform_source=uppercase_escape)
+ break
+ except ExtractorError as e:
+ if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503):
+ count += 1
+ if count <= retries:
+ continue
+ raise
+
content_html = more['content_html']
if not content_html.strip():
# Some webpages show a "Load more" button but they don't
@@ -351,7 +368,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
(?:www\.)?hooktube\.com/|
(?:www\.)?yourepeat\.com/|
tube\.majestyc\.net/|
- (?:www\.)?invidio\.us/|
+ (?:(?:www|dev)\.)?invidio\.us/|
+ (?:www\.)?invidiou\.sh/|
+ (?:www\.)?invidious\.snopyta\.org/|
+ (?:www\.)?invidious\.kabi\.tk/|
+ (?:www\.)?vid\.wxzm\.sx/|
youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains
(?:.*?\#/)? # handle anchor (#/) redirect urls
(?: # the various things that can precede the ID:
@@ -427,7 +448,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'135': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
'136': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264'},
'137': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264'},
- '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/rg3/youtube-dl/issues/4559)
+ '138': {'ext': 'mp4', 'format_note': 'DASH video', 'vcodec': 'h264'}, # Height can vary (https://github.com/ytdl-org/youtube-dl/issues/4559)
'160': {'ext': 'mp4', 'height': 144, 'format_note': 'DASH video', 'vcodec': 'h264'},
'212': {'ext': 'mp4', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'h264'},
'264': {'ext': 'mp4', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'h264'},
@@ -479,8 +500,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# RTMP (unnamed)
'_rtmp': {'protocol': 'rtmp'},
+
+ # av01 video only formats sometimes served with "unknown" codecs
+ '394': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
+ '395': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
+ '396': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
+ '397': {'acodec': 'none', 'vcodec': 'av01.0.05M.08'},
}
- _SUBTITLE_FORMATS = ('ttml', 'vtt')
+ _SUBTITLE_FORMATS = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt')
_GEO_BYPASS = False
@@ -692,7 +719,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'age_limit': 18,
},
},
- # video_info is None (https://github.com/rg3/youtube-dl/issues/4421)
+ # video_info is None (https://github.com/ytdl-org/youtube-dl/issues/4421)
# YouTube Red ad is not captured for creator
{
'url': '__2ABJjxzNo',
@@ -713,7 +740,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'DASH manifest missing',
]
},
- # Olympics (https://github.com/rg3/youtube-dl/issues/4431)
+ # Olympics (https://github.com/ytdl-org/youtube-dl/issues/4431)
{
'url': 'lqQg6PlCWgI',
'info_dict': {
@@ -764,7 +791,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
},
'skip': 'This live event has ended.',
},
- # Extraction from multiple DASH manifests (https://github.com/rg3/youtube-dl/pull/6097)
+ # Extraction from multiple DASH manifests (https://github.com/ytdl-org/youtube-dl/pull/6097)
{
'url': 'https://www.youtube.com/watch?v=FIl7x6_3R5Y',
'info_dict': {
@@ -867,7 +894,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'skip': 'This video is not available.',
},
{
- # Multifeed video with comma in title (see https://github.com/rg3/youtube-dl/issues/8536)
+ # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536)
'url': 'https://www.youtube.com/watch?v=gVfLd0zydlo',
'info_dict': {
'id': 'gVfLd0zydlo',
@@ -885,10 +912,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'only_matching': True,
},
{
- # Title with JS-like syntax "};" (see https://github.com/rg3/youtube-dl/issues/7468)
+ # Title with JS-like syntax "};" (see https://github.com/ytdl-org/youtube-dl/issues/7468)
# Also tests cut-off URL expansion in video description (see
- # https://github.com/rg3/youtube-dl/issues/1892,
- # https://github.com/rg3/youtube-dl/issues/8164)
+ # https://github.com/ytdl-org/youtube-dl/issues/1892,
+ # https://github.com/ytdl-org/youtube-dl/issues/8164)
'url': 'https://www.youtube.com/watch?v=lsguqyKfVQg',
'info_dict': {
'id': 'lsguqyKfVQg',
@@ -904,13 +931,14 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'creator': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
'track': 'Dark Walk - Position Music',
'artist': 'Todd Haberman, Daniel Law Heath and Aaron Kaplan',
+ 'album': 'Position Music - Production Music Vol. 143 - Dark Walk',
},
'params': {
'skip_download': True,
},
},
{
- # Tags with '};' (see https://github.com/rg3/youtube-dl/issues/7468)
+ # Tags with '};' (see https://github.com/ytdl-org/youtube-dl/issues/7468)
'url': 'https://www.youtube.com/watch?v=Ms7iBXnlUO8',
'only_matching': True,
},
@@ -974,7 +1002,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'only_matching': True,
},
{
- # YouTube Red paid video (https://github.com/rg3/youtube-dl/issues/10059)
+ # YouTube Red paid video (https://github.com/ytdl-org/youtube-dl/issues/10059)
'url': 'https://www.youtube.com/watch?v=i1Ko8UG-Tdo',
'only_matching': True,
},
@@ -1082,7 +1110,95 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'skip_download': True,
'youtube_include_dash_manifest': False,
},
- }
+ },
+ {
+ # Youtube Music Auto-generated description
+ 'url': 'https://music.youtube.com/watch?v=MgNrAu2pzNs',
+ 'info_dict': {
+ 'id': 'MgNrAu2pzNs',
+ 'ext': 'mp4',
+ 'title': 'Voyeur Girl',
+ 'description': 'md5:7ae382a65843d6df2685993e90a8628f',
+ 'upload_date': '20190312',
+ 'uploader': 'Various Artists - Topic',
+ 'uploader_id': 'UCVWKBi1ELZn0QX2CBLSkiyw',
+ 'artist': 'Stephen',
+ 'track': 'Voyeur Girl',
+ 'album': 'it\'s too much love to know my dear',
+ 'release_date': '20190313',
+ 'release_year': 2019,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ # Retrieve 'artist' field from 'Artist:' in video description
+ # when it is present on youtube music video
+ 'url': 'https://www.youtube.com/watch?v=k0jLE7tTwjY',
+ 'info_dict': {
+ 'id': 'k0jLE7tTwjY',
+ 'ext': 'mp4',
+ 'title': 'Latch Feat. Sam Smith',
+ 'description': 'md5:3cb1e8101a7c85fcba9b4fb41b951335',
+ 'upload_date': '20150110',
+ 'uploader': 'Various Artists - Topic',
+ 'uploader_id': 'UCNkEcmYdjrH4RqtNgh7BZ9w',
+ 'artist': 'Disclosure',
+ 'track': 'Latch Feat. Sam Smith',
+ 'album': 'Latch Featuring Sam Smith',
+ 'release_date': '20121008',
+ 'release_year': 2012,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ # handle multiple artists on youtube music video
+ 'url': 'https://www.youtube.com/watch?v=74qn0eJSjpA',
+ 'info_dict': {
+ 'id': '74qn0eJSjpA',
+ 'ext': 'mp4',
+ 'title': 'Eastside',
+ 'description': 'md5:290516bb73dcbfab0dcc4efe6c3de5f2',
+ 'upload_date': '20180710',
+ 'uploader': 'Benny Blanco - Topic',
+ 'uploader_id': 'UCzqz_ksRu_WkIzmivMdIS7A',
+ 'artist': 'benny blanco, Halsey, Khalid',
+ 'track': 'Eastside',
+ 'album': 'Eastside',
+ 'release_date': '20180713',
+ 'release_year': 2018,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
+ {
+ # Youtube Music Auto-generated description
+ # handle youtube music video with release_year and no release_date
+ 'url': 'https://www.youtube.com/watch?v=-hcAI0g-f5M',
+ 'info_dict': {
+ 'id': '-hcAI0g-f5M',
+ 'ext': 'mp4',
+ 'title': 'Put It On Me',
+ 'description': 'md5:93c55acc682ae7b0c668f2e34e1c069e',
+ 'upload_date': '20180426',
+ 'uploader': 'Matt Maeson - Topic',
+ 'uploader_id': 'UCnEkIGqtGcQMLk73Kp-Q5LQ',
+ 'artist': 'Matt Maeson',
+ 'track': 'Put It On Me',
+ 'album': 'The Hearse',
+ 'release_date': None,
+ 'release_year': 2018,
+ },
+ 'params': {
+ 'skip_download': True,
+ },
+ },
]
def __init__(self, *args, **kwargs):
@@ -1196,11 +1312,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def _parse_sig_js(self, jscode):
funcname = self._search_regex(
- (r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ (r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*encodeURIComponent\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'(?P<sig>[a-zA-Z0-9$]+)\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)',
+ # Obsolete patterns
+ r'(["\'])signature\1\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
r'\.sig\|\|(?P<sig>[a-zA-Z0-9$]+)\(',
- r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*c\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?(?P<sig>[a-zA-Z0-9$]+)\(',
- r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
- r'\bc\s*&&\s*d\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
+ r'yt\.akamaized\.net/\)\s*\|\|\s*.*?\s*[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?:encodeURIComponent\s*\()?\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\b[cs]\s*&&\s*[adf]\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\b[a-zA-Z0-9]+\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\bc\s*&&\s*a\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\(',
+ r'\bc\s*&&\s*[a-zA-Z0-9]+\.set\([^,]+\s*,\s*\([^)]*\)\s*\(\s*(?P<sig>[a-zA-Z0-9$]+)\('),
jscode, 'Initial JS player signature function name', group='sig')
jsi = JSInterpreter(jscode)
@@ -1280,8 +1403,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# regex won't capture the whole JSON. Yet working around by trying more
# concrete regex first keeping in mind proper quoted string handling
# to be implemented in future that will replace this workaround (see
- # https://github.com/rg3/youtube-dl/issues/7468,
- # https://github.com/rg3/youtube-dl/pull/7599)
+ # https://github.com/ytdl-org/youtube-dl/issues/7468,
+ # https://github.com/ytdl-org/youtube-dl/pull/7599)
r';ytplayer\.config\s*=\s*({.+?});ytplayer',
r';ytplayer\.config\s*=\s*({.+?});',
)
@@ -1465,8 +1588,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
return video_id
def _extract_annotations(self, video_id):
- url = 'https://www.youtube.com/annotations_invideo?features=1&legacy=1&video_id=%s' % video_id
- return self._download_webpage(url, video_id, note='Searching for annotations.', errnote='Unable to download video annotations.')
+ return self._download_webpage(
+ 'https://www.youtube.com/annotations_invideo', video_id,
+ note='Downloading annotations',
+ errnote='Unable to download video annotations', fatal=False,
+ query={
+ 'features': 1,
+ 'legacy': 1,
+ 'video_id': video_id,
+ })
@staticmethod
def _extract_chapters(description, duration):
@@ -1559,6 +1689,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
def extract_view_count(v_info):
return int_or_none(try_get(v_info, lambda x: x['view_count'][0]))
+ def extract_token(v_info):
+ return dict_get(v_info, ('account_playback_token', 'accountPlaybackToken', 'token'))
+
player_response = {}
# Get video info
@@ -1596,7 +1729,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
add_dash_mpd(video_info)
# Rental video is not rented but preview is available (e.g.
# https://www.youtube.com/watch?v=yYr8q0y5Jfg,
- # https://github.com/rg3/youtube-dl/issues/10532)
+ # https://github.com/ytdl-org/youtube-dl/issues/10532)
if not video_info and args.get('ypc_vid'):
return self.url_result(
args['ypc_vid'], YoutubeIE.ie_key(), video_id=args['ypc_vid'])
@@ -1616,9 +1749,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# are missing from DASH manifest pointed by webpage's dashmpd, some - from DASH
# manifest pointed by get_video_info's dashmpd).
# The general idea is to take a union of itags of both DASH manifests (for example
- # video with such 'manifest behavior' see https://github.com/rg3/youtube-dl/issues/6093)
+ # video with such 'manifest behavior' see https://github.com/ytdl-org/youtube-dl/issues/6093)
self.report_video_info_webpage_download(video_id)
- for el in ('info', 'embedded', 'detailpage', 'vevo', ''):
+ for el in ('embedded', 'detailpage', 'vevo', ''):
query = {
'video_id': video_id,
'ps': 'default',
@@ -1648,18 +1781,18 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
view_count = extract_view_count(get_video_info)
if not video_info:
video_info = get_video_info
- get_token = get_video_info.get('token') or get_video_info.get('account_playback_token')
+ get_token = extract_token(get_video_info)
if get_token:
# Different get_video_info requests may report different results, e.g.
# some may report video unavailability, but some may serve it without
- # any complaint (see https://github.com/rg3/youtube-dl/issues/7362,
+ # any complaint (see https://github.com/ytdl-org/youtube-dl/issues/7362,
# the original webpage as well as el=info and el=embedded get_video_info
# requests report video unavailability due to geo restriction while
# el=detailpage succeeds and returns valid data). This is probably
# due to YouTube measures against IP ranges of hosting providers.
# Working around by preferring the first succeeded video_info containing
# the token if no such video_info yet was found.
- token = video_info.get('token') or video_info.get('account_playback_token')
+ token = extract_token(video_info)
if not token:
video_info = get_video_info
break
@@ -1669,44 +1802,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
r'(?s)<h1[^>]+id="unavailable-message"[^>]*>(.+?)</h1>',
video_webpage, 'unavailable message', default=None)
- token = video_info.get('token') or video_info.get('account_playback_token')
- if not token:
- if 'reason' in video_info:
- if 'The uploader has not made this video available in your country.' in video_info['reason']:
- regions_allowed = self._html_search_meta(
- 'regionsAllowed', video_webpage, default=None)
- countries = regions_allowed.split(',') if regions_allowed else None
- self.raise_geo_restricted(
- msg=video_info['reason'][0], countries=countries)
- reason = video_info['reason'][0]
- if 'Invalid parameters' in reason:
- unavailable_message = extract_unavailable_message()
- if unavailable_message:
- reason = unavailable_message
- raise ExtractorError(
- 'YouTube said: %s' % reason,
- expected=True, video_id=video_id)
- else:
- raise ExtractorError(
- '"token" parameter not in video info for unknown reason',
- video_id=video_id)
-
- if video_info.get('license_info'):
- raise ExtractorError('This video is DRM protected.', expected=True)
+ if not video_info:
+ unavailable_message = extract_unavailable_message()
+ if not unavailable_message:
+ unavailable_message = 'Unable to extract video data'
+ raise ExtractorError(
+ 'YouTube said: %s' % unavailable_message, expected=True, video_id=video_id)
video_details = try_get(
player_response, lambda x: x['videoDetails'], dict) or {}
- # title
- if 'title' in video_info:
- video_title = video_info['title'][0]
- elif 'title' in player_response:
- video_title = video_details['title']
- else:
+ video_title = video_info.get('title', [None])[0] or video_details.get('title')
+ if not video_title:
self._downloader.report_warning('Unable to extract video title')
video_title = '_'
- # description
description_original = video_description = get_element_by_id("eow-description", video_webpage)
if video_description:
@@ -1731,11 +1841,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
''', replace_url, video_description)
video_description = clean_html(video_description)
else:
- fd_mobj = re.search(r'<meta name="description" content="([^"]+)"', video_webpage)
- if fd_mobj:
- video_description = unescapeHTML(fd_mobj.group(1))
- else:
- video_description = ''
+ video_description = self._html_search_meta('description', video_webpage) or video_details.get('shortDescription')
if not smuggled_data.get('force_singlefeed', False):
if not self._downloader.params.get('noplaylist'):
@@ -1750,7 +1856,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
for feed in multifeed_metadata_list.split(','):
# Unquote should take place before split on comma (,) since textual
# fields may contain comma as well (see
- # https://github.com/rg3/youtube-dl/issues/8536)
+ # https://github.com/ytdl-org/youtube-dl/issues/8536)
feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed))
entries.append({
'_type': 'url_transparent',
@@ -1775,7 +1881,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# Check for "rental" videos
if 'ypc_video_rental_bar_text' in video_info and 'author' not in video_info:
- raise ExtractorError('"rental" videos not supported. See https://github.com/rg3/youtube-dl/issues/359 for more information.', expected=True)
+ raise ExtractorError('"rental" videos not supported. See https://github.com/ytdl-org/youtube-dl/issues/359 for more information.', expected=True)
def _extract_filesize(media_url):
return int_or_none(self._search_regex(
@@ -1792,7 +1898,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
elif not is_live and (len(video_info.get('url_encoded_fmt_stream_map', [''])[0]) >= 1 or len(video_info.get('adaptive_fmts', [''])[0]) >= 1):
encoded_url_map = video_info.get('url_encoded_fmt_stream_map', [''])[0] + ',' + video_info.get('adaptive_fmts', [''])[0]
if 'rtmpe%3Dyes' in encoded_url_map:
- raise ExtractorError('rtmpe downloads are not supported, see https://github.com/rg3/youtube-dl/issues/343 for more information.', expected=True)
+ raise ExtractorError('rtmpe downloads are not supported, see https://github.com/ytdl-org/youtube-dl/issues/343 for more information.', expected=True)
formats_spec = {}
fmt_list = video_info.get('fmt_list', [''])[0]
if fmt_list:
@@ -1829,7 +1935,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
formats = []
for url_data_str in encoded_url_map.split(','):
url_data = compat_parse_qs(url_data_str)
- if 'itag' not in url_data or 'url' not in url_data:
+ if 'itag' not in url_data or 'url' not in url_data or url_data.get('drm_families'):
continue
stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0]))
# Unsupported FORMAT_STREAM_TYPE_OTF
@@ -1889,7 +1995,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
signature = self._decrypt_signature(
encrypted_sig, video_id, player_url, age_gate)
- url += '&signature=' + signature
+ sp = try_get(url_data, lambda x: x['sp'][0], compat_str) or 'signature'
+ url += '&%s=%s' % (sp, signature)
if 'ratebypass' not in url:
url += '&ratebypass=yes'
@@ -1904,7 +2011,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
dct.update(formats_spec[format_id])
# Some itags are not included in DASH manifest thus corresponding formats will
- # lack metadata (see https://github.com/rg3/youtube-dl/pull/5993).
+ # lack metadata (see https://github.com/ytdl-org/youtube-dl/pull/5993).
# Trying to extract metadata from url_encoded_fmt_stream_map entry.
mobj = re.search(r'^(?P<width>\d+)[xX](?P<height>\d+)$', url_data.get('size', [''])[0])
width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None)
@@ -1953,8 +2060,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
url_or_none(try_get(
player_response,
lambda x: x['streamingData']['hlsManifestUrl'],
- compat_str)) or
- url_or_none(try_get(
+ compat_str))
+ or url_or_none(try_get(
video_info, lambda x: x['hlsvp'][0], compat_str)))
if manifest_url:
formats = []
@@ -2002,8 +2109,13 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
else:
self._downloader.report_warning('unable to extract uploader nickname')
- channel_id = self._html_search_meta(
- 'channelId', video_webpage, 'channel id')
+ channel_id = (
+ str_or_none(video_details.get('channelId'))
+ or self._html_search_meta(
+ 'channelId', video_webpage, 'channel id', default=None)
+ or self._search_regex(
+ r'data-channel-external-id=(["\'])(?P<id>(?:(?!\1).)+)\1',
+ video_webpage, 'channel id', default=None, group='id'))
channel_url = 'http://www.youtube.com/channel/%s' % channel_id if channel_id else None
# thumbnail image
@@ -2062,6 +2174,27 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
track = extract_meta('Song')
artist = extract_meta('Artist')
+ album = extract_meta('Album')
+
+ # Youtube Music Auto-generated description
+ release_date = release_year = None
+ if video_description:
+ mobj = re.search(r'(?s)Provided to YouTube by [^\n]+\n+(?P<track>[^·]+)·(?P<artist>[^\n]+)\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]+))?', video_description)
+ if mobj:
+ if not track:
+ track = mobj.group('track').strip()
+ if not artist:
+ artist = mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·'))
+ if not album:
+ album = mobj.group('album'.strip())
+ release_year = mobj.group('release_year')
+ release_date = mobj.group('release_date')
+ if release_date:
+ release_date = release_date.replace('-', '')
+ if not release_year:
+ release_year = int(release_date[:4])
+ if release_year:
+ release_year = int(release_year)
m_episode = re.search(
r'<div[^>]+id="watch7-headline"[^>]*>\s*<span[^>]*>.*?>(?P<series>[^<]+)</a></b>\s*S(?P<season>\d+)\s*•\s*E(?P<episode>\d+)</span>',
@@ -2102,6 +2235,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
r'<[^>]+class=["\']watch-view-count[^>]+>\s*([\d,\s]+)', video_webpage,
'view count', default=None))
+ average_rating = (
+ float_or_none(video_details.get('averageRating'))
+ or try_get(video_info, lambda x: float_or_none(x['avg_rating'][0])))
+
# subtitles
video_subtitles = self.extract_subtitles(video_id, video_webpage)
automatic_captions = self.extract_automatic_captions(video_id, video_webpage)
@@ -2155,7 +2292,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
# Remove the formats we found through non-DASH, they
# contain less info and it can be wrong, because we use
# fixed values (for example the resolution). See
- # https://github.com/rg3/youtube-dl/issues/5774 for an
+ # https://github.com/ytdl-org/youtube-dl/issues/5774 for an
# example.
formats = [f for f in formats if f['format_id'] not in dash_formats.keys()]
formats.extend(dash_formats.values())
@@ -2175,6 +2312,32 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
if f.get('vcodec') != 'none':
f['stretched_ratio'] = ratio
+ if not formats:
+ token = extract_token(video_info)
+ if not token:
+ if 'reason' in video_info:
+ if 'The uploader has not made this video available in your country.' in video_info['reason']:
+ regions_allowed = self._html_search_meta(
+ 'regionsAllowed', video_webpage, default=None)
+ countries = regions_allowed.split(',') if regions_allowed else None
+ self.raise_geo_restricted(
+ msg=video_info['reason'][0], countries=countries)
+ reason = video_info['reason'][0]
+ if 'Invalid parameters' in reason:
+ unavailable_message = extract_unavailable_message()
+ if unavailable_message:
+ reason = unavailable_message
+ raise ExtractorError(
+ 'YouTube said: %s' % reason,
+ expected=True, video_id=video_id)
+ else:
+ raise ExtractorError(
+ '"token" parameter not in video info for unknown reason',
+ video_id=video_id)
+
+ if not formats and (video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos'])):
+ raise ExtractorError('This video is DRM protected.', expected=True)
+
self._sort_formats(formats)
self.mark_watched(video_id, video_info, player_response)
@@ -2205,7 +2368,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'view_count': view_count,
'like_count': like_count,
'dislike_count': dislike_count,
- 'average_rating': float_or_none(video_info.get('avg_rating', [None])[0]),
+ 'average_rating': average_rating,
'formats': formats,
'is_live': is_live,
'start_time': start_time,
@@ -2215,6 +2378,9 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
'episode_number': episode_number,
'track': track,
'artist': artist,
+ 'album': album,
+ 'release_date': release_date,
+ 'release_year': release_year,
}
@@ -2413,9 +2579,9 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
search_title = lambda class_name: get_element_by_attribute('class', class_name, webpage)
title_span = (
- search_title('playlist-title') or
- search_title('title long-title') or
- search_title('title'))
+ search_title('playlist-title')
+ or search_title('title long-title')
+ or search_title('title'))
title = clean_html(title_span)
return self.playlist_result(url_results, playlist_id, title)
@@ -2424,7 +2590,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
url = self._TEMPLATE_URL % playlist_id
page = self._download_webpage(url, playlist_id)
- # the yt-alert-message now has tabindex attribute (see https://github.com/rg3/youtube-dl/issues/11604)
+ # the yt-alert-message now has tabindex attribute (see https://github.com/ytdl-org/youtube-dl/issues/11604)
for match in re.findall(r'<div class="yt-alert-message"[^>]*>([^<]+)</div>', page):
match = match.strip()
# Check if the playlist exists or is private
@@ -2517,7 +2683,7 @@ class YoutubePlaylistIE(YoutubePlaylistBaseInfoExtractor):
return playlist
# Some playlist URLs don't actually serve a playlist (see
- # https://github.com/rg3/youtube-dl/issues/10537).
+ # https://github.com/ytdl-org/youtube-dl/issues/10537).
# Fallback to plain video extraction if there is a video id
# along with playlist id.
return self.url_result(video_id, 'Youtube', video_id=video_id)