diff options
author | pukkandan <pukkandan.ytdlp@gmail.com> | 2022-01-30 01:07:28 +0530 |
---|---|---|
committer | pukkandan <pukkandan.ytdlp@gmail.com> | 2022-01-30 01:07:28 +0530 |
commit | a3373da70c97d356bd4927eff403abd261dd8f9f (patch) | |
tree | fb77c8f297d7ca1bd6a6f3e37a4c0a4b12399db5 /youtube_dl/extractor | |
parent | 2c4cb134a90b49a4d44965b57ff43cfd45ec2d69 (diff) | |
parent | 5014bd67c22b421207b2650d4dc874b95b36dda1 (diff) | |
download | youtube-dl-a3373da70c97d356bd4927eff403abd261dd8f9f.tar.xz |
Merge branch 'UP/youtube-dl' into dl/YoutubeSearchURLIE
Diffstat (limited to 'youtube_dl/extractor')
139 files changed, 7612 insertions, 4716 deletions
diff --git a/youtube_dl/extractor/abcnews.py b/youtube_dl/extractor/abcnews.py index 8b407bf9c..908c83377 100644 --- a/youtube_dl/extractor/abcnews.py +++ b/youtube_dl/extractor/abcnews.py @@ -1,14 +1,15 @@ # coding: utf-8 from __future__ import unicode_literals -import calendar import re -import time from .amp import AMPIE from .common import InfoExtractor -from .youtube import YoutubeIE -from ..compat import compat_urlparse +from ..utils import ( + parse_duration, + parse_iso8601, + try_get, +) class AbcNewsVideoIE(AMPIE): @@ -18,8 +19,8 @@ class AbcNewsVideoIE(AMPIE): (?: abcnews\.go\.com/ (?: - [^/]+/video/(?P<display_id>[0-9a-z-]+)-| - video/embed\?.*?\bid= + (?:[^/]+/)*video/(?P<display_id>[0-9a-z-]+)-| + video/(?:embed|itemfeed)\?.*?\bid= )| fivethirtyeight\.abcnews\.go\.com/video/embed/\d+/ ) @@ -36,6 +37,8 @@ class AbcNewsVideoIE(AMPIE): 'description': 'George Stephanopoulos goes one-on-one with Iranian Foreign Minister Dr. Javad Zarif.', 'duration': 180, 'thumbnail': r're:^https?://.*\.jpg$', + 'timestamp': 1380454200, + 'upload_date': '20130929', }, 'params': { # m3u8 download @@ -47,6 +50,12 @@ class AbcNewsVideoIE(AMPIE): }, { 'url': 'http://abcnews.go.com/2020/video/2020-husband-stands-teacher-jail-student-affairs-26119478', 'only_matching': True, + }, { + 'url': 'http://abcnews.go.com/video/itemfeed?id=46979033', + 'only_matching': True, + }, { + 'url': 'https://abcnews.go.com/GMA/News/video/history-christmas-story-67894761', + 'only_matching': True, }] def _real_extract(self, url): @@ -67,28 +76,23 @@ class AbcNewsIE(InfoExtractor): _VALID_URL = r'https?://abcnews\.go\.com/(?:[^/]+/)+(?P<display_id>[0-9a-z-]+)/story\?id=(?P<id>\d+)' _TESTS = [{ - 'url': 'http://abcnews.go.com/Blotter/News/dramatic-video-rare-death-job-america/story?id=10498713#.UIhwosWHLjY', + # Youtube Embeds + 'url': 'https://abcnews.go.com/Entertainment/peter-billingsley-child-actor-christmas-story-hollywood-power/story?id=51286501', 'info_dict': { - 'id': '10505354', - 'ext': 'flv', - 'display_id': 'dramatic-video-rare-death-job-america', - 'title': 'Occupational Hazards', - 'description': 'Nightline investigates the dangers that lurk at various jobs.', - 'thumbnail': r're:^https?://.*\.jpg$', - 'upload_date': '20100428', - 'timestamp': 1272412800, + 'id': '51286501', + 'title': "Peter Billingsley: From child actor in 'A Christmas Story' to Hollywood power player", + 'description': 'Billingsley went from a child actor to Hollywood power player.', }, - 'add_ie': ['AbcNewsVideo'], + 'playlist_count': 5, }, { 'url': 'http://abcnews.go.com/Entertainment/justin-timberlake-performs-stop-feeling-eurovision-2016/story?id=39125818', 'info_dict': { 'id': '38897857', 'ext': 'mp4', - 'display_id': 'justin-timberlake-performs-stop-feeling-eurovision-2016', 'title': 'Justin Timberlake Drops Hints For Secret Single', 'description': 'Lara Spencer reports the buzziest stories of the day in "GMA" Pop News.', - 'upload_date': '20160515', - 'timestamp': 1463329500, + 'upload_date': '20160505', + 'timestamp': 1462442280, }, 'params': { # m3u8 download @@ -100,49 +104,55 @@ class AbcNewsIE(InfoExtractor): }, { 'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343', 'only_matching': True, + }, { + # inline.type == 'video' + 'url': 'http://abcnews.go.com/Technology/exclusive-apple-ceo-tim-cook-iphone-cracking-software/story?id=37173343', + 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('display_id') - video_id = mobj.group('id') - - webpage = self._download_webpage(url, video_id) - video_url = self._search_regex( - r'window\.abcnvideo\.url\s*=\s*"([^"]+)"', webpage, 'video URL') - full_video_url = compat_urlparse.urljoin(url, video_url) - - youtube_url = YoutubeIE._extract_url(webpage) - - timestamp = None - date_str = self._html_search_regex( - r'<span[^>]+class="timestamp">([^<]+)</span>', - webpage, 'timestamp', fatal=False) - if date_str: - tz_offset = 0 - if date_str.endswith(' ET'): # Eastern Time - tz_offset = -5 - date_str = date_str[:-3] - date_formats = ['%b. %d, %Y', '%b %d, %Y, %I:%M %p'] - for date_format in date_formats: - try: - timestamp = calendar.timegm(time.strptime(date_str.strip(), date_format)) - except ValueError: - continue - if timestamp is not None: - timestamp -= tz_offset * 3600 - - entry = { - '_type': 'url_transparent', - 'ie_key': AbcNewsVideoIE.ie_key(), - 'url': full_video_url, - 'id': video_id, - 'display_id': display_id, - 'timestamp': timestamp, - } - - if youtube_url: - entries = [entry, self.url_result(youtube_url, ie=YoutubeIE.ie_key())] - return self.playlist_result(entries) - - return entry + story_id = self._match_id(url) + webpage = self._download_webpage(url, story_id) + story = self._parse_json(self._search_regex( + r"window\['__abcnews__'\]\s*=\s*({.+?});", + webpage, 'data'), story_id)['page']['content']['story']['everscroll'][0] + article_contents = story.get('articleContents') or {} + + def entries(): + featured_video = story.get('featuredVideo') or {} + feed = try_get(featured_video, lambda x: x['video']['feed']) + if feed: + yield { + '_type': 'url', + 'id': featured_video.get('id'), + 'title': featured_video.get('name'), + 'url': feed, + 'thumbnail': featured_video.get('images'), + 'description': featured_video.get('description'), + 'timestamp': parse_iso8601(featured_video.get('uploadDate')), + 'duration': parse_duration(featured_video.get('duration')), + 'ie_key': AbcNewsVideoIE.ie_key(), + } + + for inline in (article_contents.get('inlines') or []): + inline_type = inline.get('type') + if inline_type == 'iframe': + iframe_url = try_get(inline, lambda x: x['attrs']['src']) + if iframe_url: + yield self.url_result(iframe_url) + elif inline_type == 'video': + video_id = inline.get('id') + if video_id: + yield { + '_type': 'url', + 'id': video_id, + 'url': 'http://abcnews.go.com/video/embed?id=' + video_id, + 'thumbnail': inline.get('imgSrc') or inline.get('imgDefault'), + 'description': inline.get('description'), + 'duration': parse_duration(inline.get('duration')), + 'ie_key': AbcNewsVideoIE.ie_key(), + } + + return self.playlist_result( + entries(), story_id, article_contents.get('headline'), + article_contents.get('subHead')) diff --git a/youtube_dl/extractor/adn.py b/youtube_dl/extractor/adn.py index c95ad2173..a55ebbcbd 100644 --- a/youtube_dl/extractor/adn.py +++ b/youtube_dl/extractor/adn.py @@ -10,6 +10,7 @@ import random from .common import InfoExtractor from ..aes import aes_cbc_decrypt from ..compat import ( + compat_HTTPError, compat_b64decode, compat_ord, ) @@ -18,11 +19,14 @@ from ..utils import ( bytes_to_long, ExtractorError, float_or_none, + int_or_none, intlist_to_bytes, long_to_bytes, pkcs1pad, strip_or_none, - urljoin, + try_get, + unified_strdate, + urlencode_postdata, ) @@ -31,16 +35,30 @@ class ADNIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?animedigitalnetwork\.fr/video/[^/]+/(?P<id>\d+)' _TEST = { 'url': 'http://animedigitalnetwork.fr/video/blue-exorcist-kyoto-saga/7778-episode-1-debut-des-hostilites', - 'md5': 'e497370d847fd79d9d4c74be55575c7a', + 'md5': '0319c99885ff5547565cacb4f3f9348d', 'info_dict': { 'id': '7778', 'ext': 'mp4', - 'title': 'Blue Exorcist - Kyôto Saga - Épisode 1', + 'title': 'Blue Exorcist - Kyôto Saga - Episode 1', 'description': 'md5:2f7b5aa76edbc1a7a92cedcda8a528d5', + 'series': 'Blue Exorcist - Kyôto Saga', + 'duration': 1467, + 'release_date': '20170106', + 'comment_count': int, + 'average_rating': float, + 'season_number': 2, + 'episode': 'Début des hostilités', + 'episode_number': 1, } } + + _NETRC_MACHINE = 'animedigitalnetwork' _BASE_URL = 'http://animedigitalnetwork.fr' - _RSA_KEY = (0xc35ae1e4356b65a73b551493da94b8cb443491c0aa092a357a5aee57ffc14dda85326f42d716e539a34542a0d3f363adf16c5ec222d713d5997194030ee2e4f0d1fb328c01a81cf6868c090d50de8e169c6b13d1675b9eeed1cbc51e1fffca9b38af07f37abd790924cd3bee59d0257cfda4fe5f3f0534877e21ce5821447d1b, 65537) + _API_BASE_URL = 'https://gw.api.animedigitalnetwork.fr/' + _PLAYER_BASE_URL = _API_BASE_URL + 'player/' + _HEADERS = {} + _LOGIN_ERR_MESSAGE = 'Unable to log in' + _RSA_KEY = (0x9B42B08905199A5CCE2026274399CA560ECB209EE9878A708B1C0812E1BB8CB5D1FB7441861147C1A1F2F3A0476DD63A9CAC20D3E983613346850AA6CB38F16DC7D720FD7D86FC6E5B3D5BBC72E14CD0BF9E869F2CEA2CCAD648F1DCE38F1FF916CEFB2D339B64AA0264372344BC775E265E8A852F88144AB0BD9AA06C1A4ABB, 65537) _POS_ALIGN_MAP = { 'start': 1, 'end': 3, @@ -54,26 +72,24 @@ class ADNIE(InfoExtractor): def _ass_subtitles_timecode(seconds): return '%01d:%02d:%02d.%02d' % (seconds / 3600, (seconds % 3600) / 60, seconds % 60, (seconds % 1) * 100) - def _get_subtitles(self, sub_path, video_id): - if not sub_path: + def _get_subtitles(self, sub_url, video_id): + if not sub_url: return None enc_subtitles = self._download_webpage( - urljoin(self._BASE_URL, sub_path), - video_id, 'Downloading subtitles location', fatal=False) or '{}' + sub_url, video_id, 'Downloading subtitles location', fatal=False) or '{}' subtitle_location = (self._parse_json(enc_subtitles, video_id, fatal=False) or {}).get('location') if subtitle_location: enc_subtitles = self._download_webpage( - urljoin(self._BASE_URL, subtitle_location), - video_id, 'Downloading subtitles data', fatal=False, - headers={'Origin': 'https://animedigitalnetwork.fr'}) + subtitle_location, video_id, 'Downloading subtitles data', + fatal=False, headers={'Origin': 'https://animedigitalnetwork.fr'}) if not enc_subtitles: return None # http://animedigitalnetwork.fr/components/com_vodvideo/videojs/adn-vjs.min.js dec_subtitles = intlist_to_bytes(aes_cbc_decrypt( bytes_to_intlist(compat_b64decode(enc_subtitles[24:])), - bytes_to_intlist(binascii.unhexlify(self._K + '4b8ef13ec1872730')), + bytes_to_intlist(binascii.unhexlify(self._K + 'ab9f52f5baae7c72')), bytes_to_intlist(compat_b64decode(enc_subtitles[:24])) )) subtitles_json = self._parse_json( @@ -117,61 +133,100 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text''' }]) return subtitles + def _real_initialize(self): + username, password = self._get_login_info() + if not username: + return + try: + access_token = (self._download_json( + self._API_BASE_URL + 'authentication/login', None, + 'Logging in', self._LOGIN_ERR_MESSAGE, fatal=False, + data=urlencode_postdata({ + 'password': password, + 'rememberMe': False, + 'source': 'Web', + 'username': username, + })) or {}).get('accessToken') + if access_token: + self._HEADERS = {'authorization': 'Bearer ' + access_token} + except ExtractorError as e: + message = None + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401: + resp = self._parse_json( + e.cause.read().decode(), None, fatal=False) or {} + message = resp.get('message') or resp.get('code') + self.report_warning(message or self._LOGIN_ERR_MESSAGE) + def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - player_config = self._parse_json(self._search_regex( - r'playerConfig\s*=\s*({.+});', webpage, - 'player config', default='{}'), video_id, fatal=False) - if not player_config: - config_url = urljoin(self._BASE_URL, self._search_regex( - r'(?:id="player"|class="[^"]*adn-player-container[^"]*")[^>]+data-url="([^"]+)"', - webpage, 'config url')) - player_config = self._download_json( - config_url, video_id, - 'Downloading player config JSON metadata')['player'] - - video_info = {} - video_info_str = self._search_regex( - r'videoInfo\s*=\s*({.+});', webpage, - 'video info', fatal=False) - if video_info_str: - video_info = self._parse_json( - video_info_str, video_id, fatal=False) or {} - - options = player_config.get('options') or {} - metas = options.get('metas') or {} - links = player_config.get('links') or {} - sub_path = player_config.get('subtitles') - error = None - if not links: - links_url = player_config.get('linksurl') or options['videoUrl'] - token = options['token'] - self._K = ''.join([random.choice('0123456789abcdef') for _ in range(16)]) - message = bytes_to_intlist(json.dumps({ - 'k': self._K, - 'e': 60, - 't': token, - })) + video_base_url = self._PLAYER_BASE_URL + 'video/%s/' % video_id + player = self._download_json( + video_base_url + 'configuration', video_id, + 'Downloading player config JSON metadata', + headers=self._HEADERS)['player'] + options = player['options'] + + user = options['user'] + if not user.get('hasAccess'): + self.raise_login_required() + + token = self._download_json( + user.get('refreshTokenUrl') or (self._PLAYER_BASE_URL + 'refresh/token'), + video_id, 'Downloading access token', headers={ + 'x-player-refresh-token': user['refreshToken'] + }, data=b'')['token'] + + links_url = try_get(options, lambda x: x['video']['url']) or (video_base_url + 'link') + self._K = ''.join([random.choice('0123456789abcdef') for _ in range(16)]) + message = bytes_to_intlist(json.dumps({ + 'k': self._K, + 't': token, + })) + + # Sometimes authentication fails for no good reason, retry with + # a different random padding + links_data = None + for _ in range(3): padded_message = intlist_to_bytes(pkcs1pad(message, 128)) n, e = self._RSA_KEY encrypted_message = long_to_bytes(pow(bytes_to_long(padded_message), e, n)) authorization = base64.b64encode(encrypted_message).decode() - links_data = self._download_json( - urljoin(self._BASE_URL, links_url), video_id, - 'Downloading links JSON metadata', headers={ - 'Authorization': 'Bearer ' + authorization, - }) - links = links_data.get('links') or {} - metas = metas or links_data.get('meta') or {} - sub_path = sub_path or links_data.get('subtitles') or \ - 'index.php?option=com_vodapi&task=subtitles.getJSON&format=json&id=' + video_id - sub_path += '&token=' + token - error = links_data.get('error') - title = metas.get('title') or video_info['title'] + + try: + links_data = self._download_json( + links_url, video_id, 'Downloading links JSON metadata', headers={ + 'X-Player-Token': authorization + }, query={ + 'freeWithAds': 'true', + 'adaptive': 'false', + 'withMetadata': 'true', + 'source': 'Web' + }) + break + except ExtractorError as e: + if not isinstance(e.cause, compat_HTTPError): + raise e + + if e.cause.code == 401: + # This usually goes away with a different random pkcs1pad, so retry + continue + + error = self._parse_json(e.cause.read(), video_id) + message = error.get('message') + if e.cause.code == 403 and error.get('code') == 'player-bad-geolocation-country': + self.raise_geo_restricted(msg=message) + raise ExtractorError(message) + else: + raise ExtractorError('Giving up retrying') + + links = links_data.get('links') or {} + metas = links_data.get('metadata') or {} + sub_url = (links.get('subtitles') or {}).get('all') + video_info = links_data.get('video') or {} + title = metas['title'] formats = [] - for format_id, qualities in links.items(): + for format_id, qualities in (links.get('streaming') or {}).items(): if not isinstance(qualities, dict): continue for quality, load_balancer_url in qualities.items(): @@ -189,19 +244,26 @@ Format: Marked,Start,End,Style,Name,MarginL,MarginR,MarginV,Effect,Text''' for f in m3u8_formats: f['language'] = 'fr' formats.extend(m3u8_formats) - if not error: - error = options.get('error') - if not formats and error: - raise ExtractorError('%s said: %s' % (self.IE_NAME, error), expected=True) self._sort_formats(formats) + video = (self._download_json( + self._API_BASE_URL + 'video/%s' % video_id, video_id, + 'Downloading additional video metadata', fatal=False) or {}).get('video') or {} + show = video.get('show') or {} + return { 'id': video_id, 'title': title, - 'description': strip_or_none(metas.get('summary') or video_info.get('resume')), - 'thumbnail': video_info.get('image'), + 'description': strip_or_none(metas.get('summary') or video.get('summary')), + 'thumbnail': video_info.get('image') or player.get('image'), 'formats': formats, - 'subtitles': self.extract_subtitles(sub_path, video_id), - 'episode': metas.get('subtitle') or video_info.get('videoTitle'), - 'series': video_info.get('playlistTitle'), + 'subtitles': self.extract_subtitles(sub_url, video_id), + 'episode': metas.get('subtitle') or video.get('name'), + 'episode_number': int_or_none(video.get('shortNumber')), + 'series': show.get('title'), + 'season_number': int_or_none(video.get('season')), + 'duration': int_or_none(video_info.get('duration') or video.get('duration')), + 'release_date': unified_strdate(video.get('releaseDate')), + 'average_rating': float_or_none(video.get('rating') or metas.get('rating')), + 'comment_count': int_or_none(video.get('commentsCount')), } diff --git a/youtube_dl/extractor/aenetworks.py b/youtube_dl/extractor/aenetworks.py index 8e4963131..e55c03fd7 100644 --- a/youtube_dl/extractor/aenetworks.py +++ b/youtube_dl/extractor/aenetworks.py @@ -252,11 +252,11 @@ class AENetworksShowIE(AENetworksListBaseIE): _TESTS = [{ 'url': 'http://www.history.com/shows/ancient-aliens', 'info_dict': { - 'id': 'SH012427480000', + 'id': 'SERIES1574', 'title': 'Ancient Aliens', 'description': 'md5:3f6d74daf2672ff3ae29ed732e37ea7f', }, - 'playlist_mincount': 168, + 'playlist_mincount': 150, }] _RESOURCE = 'series' _ITEMS_KEY = 'episodes' diff --git a/youtube_dl/extractor/aljazeera.py b/youtube_dl/extractor/aljazeera.py index c68be3134..c4f915a3c 100644 --- a/youtube_dl/extractor/aljazeera.py +++ b/youtube_dl/extractor/aljazeera.py @@ -1,13 +1,16 @@ from __future__ import unicode_literals +import json +import re + from .common import InfoExtractor class AlJazeeraIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?aljazeera\.com/(?:programmes|video)/.*?/(?P<id>[^/]+)\.html' + _VALID_URL = r'https?://(?:www\.)?aljazeera\.com/(?P<type>program/[^/]+|(?:feature|video)s)/\d{4}/\d{1,2}/\d{1,2}/(?P<id>[^/?&#]+)' _TESTS = [{ - 'url': 'http://www.aljazeera.com/programmes/the-slum/2014/08/deliverance-201482883754237240.html', + 'url': 'https://www.aljazeera.com/program/episode/2014/9/19/deliverance', 'info_dict': { 'id': '3792260579001', 'ext': 'mp4', @@ -20,14 +23,34 @@ class AlJazeeraIE(InfoExtractor): 'add_ie': ['BrightcoveNew'], 'skip': 'Not accessible from Travis CI server', }, { - 'url': 'http://www.aljazeera.com/video/news/2017/05/sierra-leone-709-carat-diamond-auctioned-170511100111930.html', + 'url': 'https://www.aljazeera.com/videos/2017/5/11/sierra-leone-709-carat-diamond-to-be-auctioned-off', + 'only_matching': True, + }, { + 'url': 'https://www.aljazeera.com/features/2017/8/21/transforming-pakistans-buses-into-art', 'only_matching': True, }] - BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/665003303001/default_default/index.html?videoId=%s' + BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s' def _real_extract(self, url): - program_name = self._match_id(url) - webpage = self._download_webpage(url, program_name) - brightcove_id = self._search_regex( - r'RenderPagesVideo\(\'(.+?)\'', webpage, 'brightcove id') - return self.url_result(self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, 'BrightcoveNew', brightcove_id) + post_type, name = re.match(self._VALID_URL, url).groups() + post_type = { + 'features': 'post', + 'program': 'episode', + 'videos': 'video', + }[post_type.split('/')[0]] + video = self._download_json( + 'https://www.aljazeera.com/graphql', name, query={ + 'operationName': 'SingleArticleQuery', + 'variables': json.dumps({ + 'name': name, + 'postType': post_type, + }), + }, headers={ + 'wp-site': 'aje', + })['data']['article']['video'] + video_id = video['id'] + account_id = video.get('accountId') or '665003303001' + player_id = video.get('playerId') or 'BkeSH5BDb' + return self.url_result( + self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, video_id), + 'BrightcoveNew', video_id) diff --git a/youtube_dl/extractor/americastestkitchen.py b/youtube_dl/extractor/americastestkitchen.py index e20f00fc3..be960c0f9 100644 --- a/youtube_dl/extractor/americastestkitchen.py +++ b/youtube_dl/extractor/americastestkitchen.py @@ -1,13 +1,16 @@ # coding: utf-8 from __future__ import unicode_literals +import json import re from .common import InfoExtractor from ..utils import ( clean_html, + int_or_none, try_get, unified_strdate, + unified_timestamp, ) @@ -22,8 +25,8 @@ class AmericasTestKitchenIE(InfoExtractor): 'ext': 'mp4', 'description': 'md5:64e606bfee910627efc4b5f050de92b3', 'thumbnail': r're:^https?://', - 'timestamp': 1523664000, - 'upload_date': '20180414', + 'timestamp': 1523318400, + 'upload_date': '20180410', 'release_date': '20180410', 'series': "America's Test Kitchen", 'season_number': 18, @@ -34,6 +37,27 @@ class AmericasTestKitchenIE(InfoExtractor): 'skip_download': True, }, }, { + # Metadata parsing behaves differently for newer episodes (705) as opposed to older episodes (582 above) + 'url': 'https://www.americastestkitchen.com/episode/705-simple-chicken-dinner', + 'md5': '06451608c57651e985a498e69cec17e5', + 'info_dict': { + 'id': '5fbe8c61bda2010001c6763b', + 'title': 'Simple Chicken Dinner', + 'ext': 'mp4', + 'description': 'md5:eb68737cc2fd4c26ca7db30139d109e7', + 'thumbnail': r're:^https?://', + 'timestamp': 1610755200, + 'upload_date': '20210116', + 'release_date': '20210116', + 'series': "America's Test Kitchen", + 'season_number': 21, + 'episode': 'Simple Chicken Dinner', + 'episode_number': 3, + }, + 'params': { + 'skip_download': True, + }, + }, { 'url': 'https://www.americastestkitchen.com/videos/3420-pan-seared-salmon', 'only_matching': True, }, { @@ -60,7 +84,76 @@ class AmericasTestKitchenIE(InfoExtractor): 'url': 'https://player.zype.com/embed/%s.js?api_key=jZ9GUhRmxcPvX7M3SlfejB6Hle9jyHTdk2jVxG7wOHPLODgncEKVdPYBhuz9iWXQ' % video['zypeId'], 'ie_key': 'Zype', 'description': clean_html(video.get('description')), + 'timestamp': unified_timestamp(video.get('publishDate')), 'release_date': unified_strdate(video.get('publishDate')), + 'episode_number': int_or_none(episode.get('number')), + 'season_number': int_or_none(episode.get('season')), 'series': try_get(episode, lambda x: x['show']['title']), 'episode': episode.get('title'), } + + +class AmericasTestKitchenSeasonIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?(?P<show>americastestkitchen|cookscountry)\.com/episodes/browse/season_(?P<id>\d+)' + _TESTS = [{ + # ATK Season + 'url': 'https://www.americastestkitchen.com/episodes/browse/season_1', + 'info_dict': { + 'id': 'season_1', + 'title': 'Season 1', + }, + 'playlist_count': 13, + }, { + # Cooks Country Season + 'url': 'https://www.cookscountry.com/episodes/browse/season_12', + 'info_dict': { + 'id': 'season_12', + 'title': 'Season 12', + }, + 'playlist_count': 13, + }] + + def _real_extract(self, url): + show_name, season_number = re.match(self._VALID_URL, url).groups() + season_number = int(season_number) + + slug = 'atk' if show_name == 'americastestkitchen' else 'cco' + + season = 'Season %d' % season_number + + season_search = self._download_json( + 'https://y1fnzxui30-dsn.algolia.net/1/indexes/everest_search_%s_season_desc_production' % slug, + season, headers={ + 'Origin': 'https://www.%s.com' % show_name, + 'X-Algolia-API-Key': '8d504d0099ed27c1b73708d22871d805', + 'X-Algolia-Application-Id': 'Y1FNZXUI30', + }, query={ + 'facetFilters': json.dumps([ + 'search_season_list:' + season, + 'search_document_klass:episode', + 'search_show_slug:' + slug, + ]), + 'attributesToRetrieve': 'description,search_%s_episode_number,search_document_date,search_url,title' % slug, + 'attributesToHighlight': '', + 'hitsPerPage': 1000, + }) + + def entries(): + for episode in (season_search.get('hits') or []): + search_url = episode.get('search_url') + if not search_url: + continue + yield { + '_type': 'url', + 'url': 'https://www.%s.com%s' % (show_name, search_url), + 'id': try_get(episode, lambda e: e['objectID'].split('_')[-1]), + 'title': episode.get('title'), + 'description': episode.get('description'), + 'timestamp': unified_timestamp(episode.get('search_document_date')), + 'season_number': season_number, + 'episode_number': int_or_none(episode.get('search_%s_episode_number' % slug)), + 'ie_key': AmericasTestKitchenIE.ie_key(), + } + + return self.playlist_result( + entries(), 'season_%d' % season_number, season) diff --git a/youtube_dl/extractor/amp.py b/youtube_dl/extractor/amp.py index 7ff098cfa..24c684cad 100644 --- a/youtube_dl/extractor/amp.py +++ b/youtube_dl/extractor/amp.py @@ -8,6 +8,7 @@ from ..utils import ( int_or_none, mimetype2ext, parse_iso8601, + unified_timestamp, url_or_none, ) @@ -88,7 +89,7 @@ class AMPIE(InfoExtractor): self._sort_formats(formats) - timestamp = parse_iso8601(item.get('pubDate'), ' ') or parse_iso8601(item.get('dc-date')) + timestamp = unified_timestamp(item.get('pubDate'), ' ') or parse_iso8601(item.get('dc-date')) return { 'id': video_id, diff --git a/youtube_dl/extractor/animeondemand.py b/youtube_dl/extractor/animeondemand.py index 00ce684d1..54e097d2f 100644 --- a/youtube_dl/extractor/animeondemand.py +++ b/youtube_dl/extractor/animeondemand.py @@ -116,8 +116,6 @@ class AnimeOnDemandIE(InfoExtractor): r'(?s)<div[^>]+itemprop="description"[^>]*>(.+?)</div>', webpage, 'anime description', default=None) - entries = [] - def extract_info(html, video_id, num=None): title, description = [None] * 2 formats = [] @@ -233,7 +231,7 @@ class AnimeOnDemandIE(InfoExtractor): self._sort_formats(info['formats']) f = common_info.copy() f.update(info) - entries.append(f) + yield f # Extract teaser/trailer only when full episode is not available if not info['formats']: @@ -247,7 +245,7 @@ class AnimeOnDemandIE(InfoExtractor): 'title': m.group('title'), 'url': urljoin(url, m.group('href')), }) - entries.append(f) + yield f def extract_episodes(html): for num, episode_html in enumerate(re.findall( @@ -275,7 +273,8 @@ class AnimeOnDemandIE(InfoExtractor): 'episode_number': episode_number, } - extract_entries(episode_html, video_id, common_info) + for e in extract_entries(episode_html, video_id, common_info): + yield e def extract_film(html, video_id): common_info = { @@ -283,11 +282,18 @@ class AnimeOnDemandIE(InfoExtractor): 'title': anime_title, 'description': anime_description, } - extract_entries(html, video_id, common_info) + for e in extract_entries(html, video_id, common_info): + yield e - extract_episodes(webpage) + def entries(): + has_episodes = False + for e in extract_episodes(webpage): + has_episodes = True + yield e - if not entries: - extract_film(webpage, anime_id) + if not has_episodes: + for e in extract_film(webpage, anime_id): + yield e - return self.playlist_result(entries, anime_id, anime_title, anime_description) + return self.playlist_result( + entries(), anime_id, anime_title, anime_description) diff --git a/youtube_dl/extractor/aol.py b/youtube_dl/extractor/aol.py index e87994a6a..f6ecb8438 100644 --- a/youtube_dl/extractor/aol.py +++ b/youtube_dl/extractor/aol.py @@ -3,7 +3,7 @@ from __future__ import unicode_literals import re -from .common import InfoExtractor +from .yahoo import YahooIE from ..compat import ( compat_parse_qs, compat_urllib_parse_urlparse, @@ -15,9 +15,9 @@ from ..utils import ( ) -class AolIE(InfoExtractor): +class AolIE(YahooIE): IE_NAME = 'aol.com' - _VALID_URL = r'(?:aol-video:|https?://(?:www\.)?aol\.(?:com|ca|co\.uk|de|jp)/video/(?:[^/]+/)*)(?P<id>[0-9a-f]+)' + _VALID_URL = r'(?:aol-video:|https?://(?:www\.)?aol\.(?:com|ca|co\.uk|de|jp)/video/(?:[^/]+/)*)(?P<id>\d{9}|[0-9a-f]{24}|[0-9a-f]{8}-(?:[0-9a-f]{4}-){3}[0-9a-f]{12})' _TESTS = [{ # video with 5min ID @@ -76,10 +76,16 @@ class AolIE(InfoExtractor): }, { 'url': 'https://www.aol.jp/video/playlist/5a28e936a1334d000137da0c/5a28f3151e642219fde19831/', 'only_matching': True, + }, { + # Yahoo video + 'url': 'https://www.aol.com/video/play/991e6700-ac02-11ea-99ff-357400036f61/24bbc846-3e30-3c46-915e-fe8ccd7fcc46/', + 'only_matching': True, }] def _real_extract(self, url): video_id = self._match_id(url) + if '-' in video_id: + return self._extract_yahoo_video(video_id, 'us') response = self._download_json( 'https://feedapi.b2c.on.aol.com/v1.0/app/videos/aolon/%s/details' % video_id, diff --git a/youtube_dl/extractor/apa.py b/youtube_dl/extractor/apa.py index 98ccdaa4a..cbc1c0ecb 100644 --- a/youtube_dl/extractor/apa.py +++ b/youtube_dl/extractor/apa.py @@ -6,25 +6,21 @@ import re from .common import InfoExtractor from ..utils import ( determine_ext, - js_to_json, + int_or_none, url_or_none, ) class APAIE(InfoExtractor): - _VALID_URL = r'https?://[^/]+\.apa\.at/embed/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' + _VALID_URL = r'(?P<base_url>https?://[^/]+\.apa\.at)/embed/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' _TESTS = [{ 'url': 'http://uvp.apa.at/embed/293f6d17-692a-44e3-9fd5-7b178f3a1029', 'md5': '2b12292faeb0a7d930c778c7a5b4759b', 'info_dict': { - 'id': 'jjv85FdZ', + 'id': '293f6d17-692a-44e3-9fd5-7b178f3a1029', 'ext': 'mp4', - 'title': '"Blau ist mysteriös": Die Blue Man Group im Interview', - 'description': 'md5:d41d8cd98f00b204e9800998ecf8427e', + 'title': '293f6d17-692a-44e3-9fd5-7b178f3a1029', 'thumbnail': r're:^https?://.*\.jpg$', - 'duration': 254, - 'timestamp': 1519211149, - 'upload_date': '20180221', }, }, { 'url': 'https://uvp-apapublisher.sf.apa.at/embed/2f94e9e6-d945-4db2-9548-f9a41ebf7b78', @@ -46,9 +42,11 @@ class APAIE(InfoExtractor): webpage)] def _real_extract(self, url): - video_id = self._match_id(url) + mobj = re.match(self._VALID_URL, url) + video_id, base_url = mobj.group('id', 'base_url') - webpage = self._download_webpage(url, video_id) + webpage = self._download_webpage( + '%s/player/%s' % (base_url, video_id), video_id) jwplatform_id = self._search_regex( r'media[iI]d\s*:\s*["\'](?P<id>[a-zA-Z0-9]{8})', webpage, @@ -59,16 +57,18 @@ class APAIE(InfoExtractor): 'jwplatform:' + jwplatform_id, ie='JWPlatform', video_id=video_id) - sources = self._parse_json( - self._search_regex( - r'sources\s*=\s*(\[.+?\])\s*;', webpage, 'sources'), - video_id, transform_source=js_to_json) + def extract(field, name=None): + return self._search_regex( + r'\b%s["\']\s*:\s*(["\'])(?P<value>(?:(?!\1).)+)\1' % field, + webpage, name or field, default=None, group='value') + + title = extract('title') or video_id + description = extract('description') + thumbnail = extract('poster', 'thumbnail') formats = [] - for source in sources: - if not isinstance(source, dict): - continue - source_url = url_or_none(source.get('file')) + for format_id in ('hls', 'progressive'): + source_url = url_or_none(extract(format_id)) if not source_url: continue ext = determine_ext(source_url) @@ -77,18 +77,19 @@ class APAIE(InfoExtractor): source_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)) else: + height = int_or_none(self._search_regex( + r'(\d+)\.mp4', source_url, 'height', default=None)) formats.append({ 'url': source_url, + 'format_id': format_id, + 'height': height, }) self._sort_formats(formats) - thumbnail = self._search_regex( - r'image\s*:\s*(["\'])(?P<url>(?:(?!\1).)+)\1', webpage, - 'thumbnail', fatal=False, group='url') - return { 'id': video_id, - 'title': video_id, + 'title': title, + 'description': description, 'thumbnail': thumbnail, 'formats': formats, } diff --git a/youtube_dl/extractor/appleconnect.py b/youtube_dl/extractor/appleconnect.py index a84b8b1eb..494f8330c 100644 --- a/youtube_dl/extractor/appleconnect.py +++ b/youtube_dl/extractor/appleconnect.py @@ -9,10 +9,10 @@ from ..utils import ( class AppleConnectIE(InfoExtractor): - _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/idsa\.(?P<id>[\w-]+)' - _TEST = { + _VALID_URL = r'https?://itunes\.apple\.com/\w{0,2}/?post/(?:id)?sa\.(?P<id>[\w-]+)' + _TESTS = [{ 'url': 'https://itunes.apple.com/us/post/idsa.4ab17a39-2720-11e5-96c5-a5b38f6c42d3', - 'md5': 'e7c38568a01ea45402570e6029206723', + 'md5': 'c1d41f72c8bcaf222e089434619316e4', 'info_dict': { 'id': '4ab17a39-2720-11e5-96c5-a5b38f6c42d3', 'ext': 'm4v', @@ -22,7 +22,10 @@ class AppleConnectIE(InfoExtractor): 'upload_date': '20150710', 'timestamp': 1436545535, }, - } + }, { + 'url': 'https://itunes.apple.com/us/post/sa.0fe0229f-2457-11e5-9f40-1bb645f2d5d9', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) @@ -36,7 +39,7 @@ class AppleConnectIE(InfoExtractor): video_data = self._parse_json(video_json, video_id) timestamp = str_to_int(self._html_search_regex(r'data-timestamp="(\d+)"', webpage, 'timestamp')) - like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count')) + like_count = str_to_int(self._html_search_regex(r'(\d+) Loves', webpage, 'like count', default=None)) return { 'id': video_id, diff --git a/youtube_dl/extractor/applepodcasts.py b/youtube_dl/extractor/applepodcasts.py index 95758fece..6a74de758 100644 --- a/youtube_dl/extractor/applepodcasts.py +++ b/youtube_dl/extractor/applepodcasts.py @@ -42,6 +42,7 @@ class ApplePodcastsIE(InfoExtractor): ember_data = self._parse_json(self._search_regex( r'id="shoebox-ember-data-store"[^>]*>\s*({.+?})\s*<', webpage, 'ember data'), episode_id) + ember_data = ember_data.get(episode_id) or ember_data episode = ember_data['data']['attributes'] description = episode.get('description') or {} diff --git a/youtube_dl/extractor/archiveorg.py b/youtube_dl/extractor/archiveorg.py index c79c58e82..e42ed5e79 100644 --- a/youtube_dl/extractor/archiveorg.py +++ b/youtube_dl/extractor/archiveorg.py @@ -2,15 +2,17 @@ from __future__ import unicode_literals from .common import InfoExtractor from ..utils import ( - unified_strdate, clean_html, + extract_attributes, + unified_strdate, + unified_timestamp, ) class ArchiveOrgIE(InfoExtractor): IE_NAME = 'archive.org' IE_DESC = 'archive.org videos' - _VALID_URL = r'https?://(?:www\.)?archive\.org/(?:details|embed)/(?P<id>[^/?#]+)(?:[?].*)?$' + _VALID_URL = r'https?://(?:www\.)?archive\.org/(?:details|embed)/(?P<id>[^/?#&]+)' _TESTS = [{ 'url': 'http://archive.org/details/XD300-23_68HighlightsAResearchCntAugHumanIntellect', 'md5': '8af1d4cf447933ed3c7f4871162602db', @@ -19,8 +21,11 @@ class ArchiveOrgIE(InfoExtractor): 'ext': 'ogg', 'title': '1968 Demo - FJCC Conference Presentation Reel #1', 'description': 'md5:da45c349df039f1cc8075268eb1b5c25', - 'upload_date': '19681210', - 'uploader': 'SRI International' + 'creator': 'SRI International', + 'release_date': '19681210', + 'uploader': 'SRI International', + 'timestamp': 1268695290, + 'upload_date': '20100315', } }, { 'url': 'https://archive.org/details/Cops1922', @@ -29,22 +34,43 @@ class ArchiveOrgIE(InfoExtractor): 'id': 'Cops1922', 'ext': 'mp4', 'title': 'Buster Keaton\'s "Cops" (1922)', - 'description': 'md5:89e7c77bf5d965dd5c0372cfb49470f6', + 'description': 'md5:43a603fd6c5b4b90d12a96b921212b9c', + 'timestamp': 1387699629, + 'upload_date': '20131222', } }, { 'url': 'http://archive.org/embed/XD300-23_68HighlightsAResearchCntAugHumanIntellect', 'only_matching': True, + }, { + 'url': 'https://archive.org/details/MSNBCW_20131125_040000_To_Catch_a_Predator/', + 'only_matching': True, }] def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage( 'http://archive.org/embed/' + video_id, video_id) - jwplayer_playlist = self._parse_json(self._search_regex( - r"(?s)Play\('[^']+'\s*,\s*(\[.+\])\s*,\s*{.*?}\)", - webpage, 'jwplayer playlist'), video_id) - info = self._parse_jwplayer_data( - {'playlist': jwplayer_playlist}, video_id, base_url=url) + + playlist = None + play8 = self._search_regex( + r'(<[^>]+\bclass=["\']js-play8-playlist[^>]+>)', webpage, + 'playlist', default=None) + if play8: + attrs = extract_attributes(play8) + playlist = attrs.get('value') + if not playlist: + # Old jwplayer fallback + playlist = self._search_regex( + r"(?s)Play\('[^']+'\s*,\s*(\[.+\])\s*,\s*{.*?}\)", + webpage, 'jwplayer playlist', default='[]') + jwplayer_playlist = self._parse_json(playlist, video_id, fatal=False) + if jwplayer_playlist: + info = self._parse_jwplayer_data( + {'playlist': jwplayer_playlist}, video_id, base_url=url) + else: + # HTML5 media fallback + info = self._parse_html5_media_entries(url, webpage, video_id)[0] + info['id'] = video_id def get_optional(metadata, field): return metadata.get(field, [None])[0] @@ -58,8 +84,12 @@ class ArchiveOrgIE(InfoExtractor): 'description': clean_html(get_optional(metadata, 'description')), }) if info.get('_type') != 'playlist': + creator = get_optional(metadata, 'creator') info.update({ - 'uploader': get_optional(metadata, 'creator'), - 'upload_date': unified_strdate(get_optional(metadata, 'date')), + 'creator': creator, + 'release_date': unified_strdate(get_optional(metadata, 'date')), + 'uploader': get_optional(metadata, 'publisher') or creator, + 'timestamp': unified_timestamp(get_optional(metadata, 'publicdate')), + 'language': get_optional(metadata, 'language'), }) return info diff --git a/youtube_dl/extractor/ard.py b/youtube_dl/extractor/ard.py index 5b7b2dd6d..d45a9fe52 100644 --- a/youtube_dl/extractor/ard.py +++ b/youtube_dl/extractor/ard.py @@ -187,13 +187,13 @@ class ARDMediathekIE(ARDMediathekBaseIE): if doc.tag == 'rss': return GenericIE()._extract_rss(url, video_id, doc) - title = self._html_search_regex( + title = self._og_search_title(webpage, default=None) or self._html_search_regex( [r'<h1(?:\s+class="boxTopHeadline")?>(.*?)</h1>', r'<meta name="dcterms\.title" content="(.*?)"/>', r'<h4 class="headline">(.*?)</h4>', r'<title[^>]*>(.*?)</title>'], webpage, 'title') - description = self._html_search_meta( + description = self._og_search_description(webpage, default=None) or self._html_search_meta( 'dcterms.abstract', webpage, 'description', default=None) if description is None: description = self._html_search_meta( @@ -249,31 +249,40 @@ class ARDMediathekIE(ARDMediathekBaseIE): class ARDIE(InfoExtractor): - _VALID_URL = r'(?P<mainurl>https?://(www\.)?daserste\.de/[^?#]+/videos(?:extern)?/(?P<display_id>[^/?#]+)-(?P<id>[0-9]+))\.html' + _VALID_URL = r'(?P<mainurl>https?://(?:www\.)?daserste\.de/(?:[^/?#&]+/)+(?P<id>[^/?#&]+))\.html' _TESTS = [{ - # available till 14.02.2019 - 'url': 'http://www.daserste.de/information/talk/maischberger/videos/das-groko-drama-zerlegen-sich-die-volksparteien-video-102.html', - 'md5': '8e4ec85f31be7c7fc08a26cdbc5a1f49', + # available till 7.01.2022 + 'url': 'https://www.daserste.de/information/talk/maischberger/videos/maischberger-die-woche-video100.html', + 'md5': '867d8aa39eeaf6d76407c5ad1bb0d4c1', 'info_dict': { - 'display_id': 'das-groko-drama-zerlegen-sich-die-volksparteien-video', - 'id': '102', + 'id': 'maischberger-die-woche-video100', + 'display_id': 'maischberger-die-woche-video100', 'ext': 'mp4', - 'duration': 4435.0, - 'title': 'Das GroKo-Drama: Zerlegen sich die Volksparteien?', - 'upload_date': '20180214', + 'duration': 3687.0, + 'title': 'maischberger. die woche vom 7. Januar 2021', + 'upload_date': '20210107', 'thumbnail': r're:^https?://.*\.jpg$', }, }, { - 'url': 'https://www.daserste.de/information/reportage-dokumentation/erlebnis-erde/videosextern/woelfe-und-herdenschutzhunde-ungleiche-brueder-102.html', + 'url': 'https://www.daserste.de/information/politik-weltgeschehen/morgenmagazin/videosextern/dominik-kahun-aus-der-nhl-direkt-zur-weltmeisterschaft-100.html', + 'only_matching': True, + }, { + 'url': 'https://www.daserste.de/information/nachrichten-wetter/tagesthemen/videosextern/tagesthemen-17736.html', 'only_matching': True, }, { 'url': 'http://www.daserste.de/information/reportage-dokumentation/dokus/videos/die-story-im-ersten-mission-unter-falscher-flagge-100.html', 'only_matching': True, + }, { + 'url': 'https://www.daserste.de/unterhaltung/serie/in-aller-freundschaft-die-jungen-aerzte/Drehpause-100.html', + 'only_matching': True, + }, { + 'url': 'https://www.daserste.de/unterhaltung/film/filmmittwoch-im-ersten/videos/making-ofwendezeit-video-100.html', + 'only_matching': True, }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - display_id = mobj.group('display_id') + display_id = mobj.group('id') player_url = mobj.group('mainurl') + '~playerXml.xml' doc = self._download_xml(player_url, display_id) @@ -284,25 +293,47 @@ class ARDIE(InfoExtractor): formats = [] for a in video_node.findall('.//asset'): + file_name = xpath_text(a, './fileName', default=None) + if not file_name: + continue + format_type = a.attrib.get('type') + format_url = url_or_none(file_name) + if format_url: + ext = determine_ext(file_name) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + format_url, display_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id=format_type or 'hls', fatal=False)) + continue + elif ext == 'f4m': + formats.extend(self._extract_f4m_formats( + update_url_query(format_url, {'hdcore': '3.7.0'}), + display_id, f4m_id=format_type or 'hds', fatal=False)) + continue f = { - 'format_id': a.attrib['type'], - 'width': int_or_none(a.find('./frameWidth').text), - 'height': int_or_none(a.find('./frameHeight').text), - 'vbr': int_or_none(a.find('./bitrateVideo').text), - 'abr': int_or_none(a.find('./bitrateAudio').text), - 'vcodec': a.find('./codecVideo').text, - 'tbr': int_or_none(a.find('./totalBitrate').text), + 'format_id': format_type, + 'width': int_or_none(xpath_text(a, './frameWidth')), + 'height': int_or_none(xpath_text(a, './frameHeight')), + 'vbr': int_or_none(xpath_text(a, './bitrateVideo')), + 'abr': int_or_none(xpath_text(a, './bitrateAudio')), + 'vcodec': xpath_text(a, './codecVideo'), + 'tbr': int_or_none(xpath_text(a, './totalBitrate')), } - if a.find('./serverPrefix').text: - f['url'] = a.find('./serverPrefix').text - f['playpath'] = a.find('./fileName').text + server_prefix = xpath_text(a, './serverPrefix', default=None) + if server_prefix: + f.update({ + 'url': server_prefix, + 'playpath': file_name, + }) else: - f['url'] = a.find('./fileName').text + if not format_url: + continue + f['url'] = format_url formats.append(f) self._sort_formats(formats) return { - 'id': mobj.group('id'), + 'id': xpath_text(video_node, './videoId', default=display_id), 'formats': formats, 'display_id': display_id, 'title': video_node.find('./title').text, @@ -313,19 +344,19 @@ class ARDIE(InfoExtractor): class ARDBetaMediathekIE(ARDMediathekBaseIE): - _VALID_URL = r'https://(?:(?:beta|www)\.)?ardmediathek\.de/(?P<client>[^/]+)/(?:player|live|video)/(?P<display_id>(?:[^/]+/)*)(?P<video_id>[a-zA-Z0-9]+)' + _VALID_URL = r'https://(?:(?:beta|www)\.)?ardmediathek\.de/(?:[^/]+/)?(?:player|live|video)/(?:[^/]+/)*(?P<id>Y3JpZDovL[a-zA-Z0-9]+)' _TESTS = [{ - 'url': 'https://ardmediathek.de/ard/video/die-robuste-roswita/Y3JpZDovL2Rhc2Vyc3RlLmRlL3RhdG9ydC9mYmM4NGM1NC0xNzU4LTRmZGYtYWFhZS0wYzcyZTIxNGEyMDE', - 'md5': 'dfdc87d2e7e09d073d5a80770a9ce88f', + 'url': 'https://www.ardmediathek.de/mdr/video/die-robuste-roswita/Y3JpZDovL21kci5kZS9iZWl0cmFnL2Ntcy84MWMxN2MzZC0wMjkxLTRmMzUtODk4ZS0wYzhlOWQxODE2NGI/', + 'md5': 'a1dc75a39c61601b980648f7c9f9f71d', 'info_dict': { 'display_id': 'die-robuste-roswita', - 'id': '70153354', + 'id': '78566716', 'title': 'Die robuste Roswita', - 'description': r're:^Der Mord.*trüber ist als die Ilm.', + 'description': r're:^Der Mord.*totgeglaubte Ehefrau Roswita', 'duration': 5316, - 'thumbnail': 'https://img.ardmediathek.de/standard/00/70/15/33/90/-1852531467/16x9/960?mandant=ard', - 'timestamp': 1577047500, - 'upload_date': '20191222', + 'thumbnail': 'https://img.ardmediathek.de/standard/00/78/56/67/84/575672121/16x9/960?mandant=ard', + 'timestamp': 1596658200, + 'upload_date': '20200805', 'ext': 'mp4', }, }, { @@ -343,22 +374,22 @@ class ARDBetaMediathekIE(ARDMediathekBaseIE): }, { 'url': 'https://www.ardmediathek.de/swr/live/Y3JpZDovL3N3ci5kZS8xMzQ4MTA0Mg', 'only_matching': True, + }, { + 'url': 'https://www.ardmediathek.de/video/coronavirus-update-ndr-info/astrazeneca-kurz-lockdown-und-pims-syndrom-81/ndr/Y3JpZDovL25kci5kZS84NzE0M2FjNi0wMWEwLTQ5ODEtOTE5NS1mOGZhNzdhOTFmOTI/', + 'only_matching': True, + }, { + 'url': 'https://www.ardmediathek.de/ard/player/Y3JpZDovL3dkci5kZS9CZWl0cmFnLWQ2NDJjYWEzLTMwZWYtNGI4NS1iMTI2LTU1N2UxYTcxOGIzOQ/tatort-duo-koeln-leipzig-ihr-kinderlein-kommet', + 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('video_id') - display_id = mobj.group('display_id') - if display_id: - display_id = display_id.rstrip('/') - if not display_id: - display_id = video_id + video_id = self._match_id(url) player_page = self._download_json( 'https://api.ardmediathek.de/public-gateway', - display_id, data=json.dumps({ + video_id, data=json.dumps({ 'query': '''{ - playerPage(client:"%s", clipId: "%s") { + playerPage(client: "ard", clipId: "%s") { blockedByFsk broadcastedOn maturityContentRating @@ -388,7 +419,7 @@ class ARDBetaMediathekIE(ARDMediathekBaseIE): } } } -}''' % (mobj.group('client'), video_id), +}''' % video_id, }).encode(), headers={ 'Content-Type': 'application/json' })['data']['playerPage'] @@ -413,7 +444,6 @@ class ARDBetaMediathekIE(ARDMediathekBaseIE): r'\(FSK\s*(\d+)\)\s*$', description, 'age limit', default=None)) info.update({ 'age_limit': age_limit, - 'display_id': display_id, 'title': title, 'description': description, 'timestamp': unified_timestamp(player_page.get('broadcastedOn')), diff --git a/youtube_dl/extractor/arnes.py b/youtube_dl/extractor/arnes.py new file mode 100644 index 000000000..c0032fcab --- /dev/null +++ b/youtube_dl/extractor/arnes.py @@ -0,0 +1,101 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import ( + compat_parse_qs, + compat_urllib_parse_urlparse, +) +from ..utils import ( + float_or_none, + int_or_none, + parse_iso8601, + remove_start, +) + + +class ArnesIE(InfoExtractor): + IE_NAME = 'video.arnes.si' + IE_DESC = 'Arnes Video' + _VALID_URL = r'https?://video\.arnes\.si/(?:[a-z]{2}/)?(?:watch|embed|api/(?:asset|public/video))/(?P<id>[0-9a-zA-Z]{12})' + _TESTS = [{ + 'url': 'https://video.arnes.si/watch/a1qrWTOQfVoU?t=10', + 'md5': '4d0f4d0a03571b33e1efac25fd4a065d', + 'info_dict': { + 'id': 'a1qrWTOQfVoU', + 'ext': 'mp4', + 'title': 'Linearna neodvisnost, definicija', + 'description': 'Linearna neodvisnost, definicija', + 'license': 'PRIVATE', + 'creator': 'Polona Oblak', + 'timestamp': 1585063725, + 'upload_date': '20200324', + 'channel': 'Polona Oblak', + 'channel_id': 'q6pc04hw24cj', + 'channel_url': 'https://video.arnes.si/?channel=q6pc04hw24cj', + 'duration': 596.75, + 'view_count': int, + 'tags': ['linearna_algebra'], + 'start_time': 10, + } + }, { + 'url': 'https://video.arnes.si/api/asset/s1YjnV7hadlC/play.mp4', + 'only_matching': True, + }, { + 'url': 'https://video.arnes.si/embed/s1YjnV7hadlC', + 'only_matching': True, + }, { + 'url': 'https://video.arnes.si/en/watch/s1YjnV7hadlC', + 'only_matching': True, + }, { + 'url': 'https://video.arnes.si/embed/s1YjnV7hadlC?t=123&hideRelated=1', + 'only_matching': True, + }, { + 'url': 'https://video.arnes.si/api/public/video/s1YjnV7hadlC', + 'only_matching': True, + }] + _BASE_URL = 'https://video.arnes.si' + + def _real_extract(self, url): + video_id = self._match_id(url) + + video = self._download_json( + self._BASE_URL + '/api/public/video/' + video_id, video_id)['data'] + title = video['title'] + + formats = [] + for media in (video.get('media') or []): + media_url = media.get('url') + if not media_url: + continue + formats.append({ + 'url': self._BASE_URL + media_url, + 'format_id': remove_start(media.get('format'), 'FORMAT_'), + 'format_note': media.get('formatTranslation'), + 'width': int_or_none(media.get('width')), + 'height': int_or_none(media.get('height')), + }) + self._sort_formats(formats) + + channel = video.get('channel') or {} + channel_id = channel.get('url') + thumbnail = video.get('thumbnailUrl') + + return { + 'id': video_id, + 'title': title, + 'formats': formats, + 'thumbnail': self._BASE_URL + thumbnail, + 'description': video.get('description'), + 'license': video.get('license'), + 'creator': video.get('author'), + 'timestamp': parse_iso8601(video.get('creationTime')), + 'channel': channel.get('name'), + 'channel_id': channel_id, + 'channel_url': self._BASE_URL + '/?channel=' + channel_id if channel_id else None, + 'duration': float_or_none(video.get('duration'), 1000), + 'view_count': int_or_none(video.get('views')), + 'tags': video.get('hashtags'), + 'start_time': int_or_none(compat_parse_qs( + compat_urllib_parse_urlparse(url).query).get('t', [None])[0]), + } diff --git a/youtube_dl/extractor/awaan.py b/youtube_dl/extractor/awaan.py index a2603bbff..3a7700cd4 100644 --- a/youtube_dl/extractor/awaan.py +++ b/youtube_dl/extractor/awaan.py @@ -48,6 +48,7 @@ class AWAANBaseIE(InfoExtractor): 'duration': int_or_none(video_data.get('duration')), 'timestamp': parse_iso8601(video_data.get('create_time'), ' '), 'is_live': is_live, + 'uploader_id': video_data.get('user_id'), } @@ -107,6 +108,7 @@ class AWAANLiveIE(AWAANBaseIE): 'title': 're:Dubai Al Oula [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', 'upload_date': '20150107', 'timestamp': 1420588800, + 'uploader_id': '71', }, 'params': { # m3u8 download diff --git a/youtube_dl/extractor/azmedien.py b/youtube_dl/extractor/azmedien.py index b1e20def5..930266990 100644 --- a/youtube_dl/extractor/azmedien.py +++ b/youtube_dl/extractor/azmedien.py @@ -47,7 +47,7 @@ class AZMedienIE(InfoExtractor): 'url': 'https://www.telebaern.tv/telebaern-news/montag-1-oktober-2018-ganze-sendung-133531189#video=0_7xjo9lf1', 'only_matching': True }] - _API_TEMPL = 'https://www.%s/api/pub/gql/%s/NewsArticleTeaser/cb9f2f81ed22e9b47f4ca64ea3cc5a5d13e88d1d' + _API_TEMPL = 'https://www.%s/api/pub/gql/%s/NewsArticleTeaser/a4016f65fe62b81dc6664dd9f4910e4ab40383be' _PARTNER_ID = '1719221' def _real_extract(self, url): diff --git a/youtube_dl/extractor/bandaichannel.py b/youtube_dl/extractor/bandaichannel.py new file mode 100644 index 000000000..d67285913 --- /dev/null +++ b/youtube_dl/extractor/bandaichannel.py @@ -0,0 +1,37 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .brightcove import BrightcoveNewIE +from ..utils import extract_attributes + + +class BandaiChannelIE(BrightcoveNewIE): + IE_NAME = 'bandaichannel' + _VALID_URL = r'https?://(?:www\.)?b-ch\.com/titles/(?P<id>\d+/\d+)' + _TESTS = [{ + 'url': 'https://www.b-ch.com/titles/514/001', + 'md5': 'a0f2d787baa5729bed71108257f613a4', + 'info_dict': { + 'id': '6128044564001', + 'ext': 'mp4', + 'title': 'メタルファイターMIKU 第1話', + 'timestamp': 1580354056, + 'uploader_id': '5797077852001', + 'upload_date': '20200130', + 'duration': 1387.733, + }, + 'params': { + 'format': 'bestvideo', + 'skip_download': True, + }, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + webpage = self._download_webpage(url, video_id) + attrs = extract_attributes(self._search_regex( + r'(<video-js[^>]+\bid="bcplayer"[^>]*>)', webpage, 'player')) + bc = self._download_json( + 'https://pbifcd.b-ch.com/v1/playbackinfo/ST/70/' + attrs['data-info'], + video_id, headers={'X-API-KEY': attrs['data-auth'].strip()})['bc'] + return self._parse_brightcove_metadata(bc, bc['id']) diff --git a/youtube_dl/extractor/bandcamp.py b/youtube_dl/extractor/bandcamp.py index 69e673a26..006aab3b4 100644 --- a/youtube_dl/extractor/bandcamp.py +++ b/youtube_dl/extractor/bandcamp.py @@ -49,6 +49,7 @@ class BandcampIE(InfoExtractor): 'uploader': 'Ben Prunty', 'timestamp': 1396508491, 'upload_date': '20140403', + 'release_timestamp': 1396483200, 'release_date': '20140403', 'duration': 260.877, 'track': 'Lanius (Battle)', @@ -69,6 +70,7 @@ class BandcampIE(InfoExtractor): 'uploader': 'Mastodon', 'timestamp': 1322005399, 'upload_date': '20111122', + 'release_timestamp': 1076112000, 'release_date': '20040207', 'duration': 120.79, 'track': 'Hail to Fire', @@ -197,7 +199,7 @@ class BandcampIE(InfoExtractor): 'thumbnail': thumbnail, 'uploader': artist, 'timestamp': timestamp, - 'release_date': unified_strdate(tralbum.get('album_release_date')), + 'release_timestamp': unified_timestamp(tralbum.get('album_release_date')), 'duration': duration, 'track': track, 'track_number': track_number, diff --git a/youtube_dl/extractor/bbc.py b/youtube_dl/extractor/bbc.py index b4daee54e..247d982ce 100644 --- a/youtube_dl/extractor/bbc.py +++ b/youtube_dl/extractor/bbc.py @@ -1,31 +1,39 @@ # coding: utf-8 from __future__ import unicode_literals +import functools import itertools +import json import re from .common import InfoExtractor +from ..compat import ( + compat_etree_Element, + compat_HTTPError, + compat_parse_qs, + compat_str, + compat_urllib_parse_urlparse, + compat_urlparse, +) from ..utils import ( + ExtractorError, + OnDemandPagedList, clean_html, dict_get, - ExtractorError, float_or_none, get_element_by_class, int_or_none, js_to_json, parse_duration, parse_iso8601, + strip_or_none, try_get, unescapeHTML, + unified_timestamp, url_or_none, urlencode_postdata, urljoin, ) -from ..compat import ( - compat_etree_Element, - compat_HTTPError, - compat_urlparse, -) class BBCCoUkIE(InfoExtractor): @@ -756,8 +764,17 @@ class BBCIE(BBCCoUkIE): 'only_matching': True, }, { # custom redirection to www.bbc.com + # also, video with window.__INITIAL_DATA__ 'url': 'http://www.bbc.co.uk/news/science-environment-33661876', - 'only_matching': True, + 'info_dict': { + 'id': 'p02xzws1', + 'ext': 'mp4', + 'title': "Pluto may have 'nitrogen glaciers'", + 'description': 'md5:6a95b593f528d7a5f2605221bc56912f', + 'thumbnail': r're:https?://.+/.+\.jpg', + 'timestamp': 1437785037, + 'upload_date': '20150725', + }, }, { # single video article embedded with data-media-vpid 'url': 'http://www.bbc.co.uk/sport/rowing/35908187', @@ -793,11 +810,25 @@ class BBCIE(BBCCoUkIE): 'description': 'Learn English words and phrases from this story', }, 'add_ie': [BBCCoUkIE.ie_key()], + }, { + # BBC Reel + 'url': 'https://www.bbc.com/reel/video/p07c6sb6/how-positive-thinking-is-harming-your-happiness', + 'info_dict': { + 'id': 'p07c6sb9', + 'ext': 'mp4', + 'title': 'How positive thinking is harming your happiness', + 'alt_title': 'The downsides of positive thinking', + 'description': 'md5:fad74b31da60d83b8265954ee42d85b4', + 'duration': 235, + 'thumbnail': r're:https?://.+/p07c9dsr.jpg', + 'upload_date': '20190604', + 'categories': ['Psychology'], + }, }] @classmethod def suitable(cls, url): - EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerPlaylistIE, BBCCoUkPlaylistIE) + EXCLUDE_IE = (BBCCoUkIE, BBCCoUkArticleIE, BBCCoUkIPlayerEpisodesIE, BBCCoUkIPlayerGroupIE, BBCCoUkPlaylistIE) return (False if any(ie.suitable(url) for ie in EXCLUDE_IE) else super(BBCIE, cls).suitable(url)) @@ -929,7 +960,7 @@ class BBCIE(BBCCoUkIE): else: entry['title'] = info['title'] entry['formats'].extend(info['formats']) - except Exception as e: + except ExtractorError as e: # Some playlist URL may fail with 500, at the same time # the other one may work fine (e.g. # http://www.bbc.com/turkce/haberler/2015/06/150615_telabyad_kentin_cogu) @@ -980,6 +1011,37 @@ class BBCIE(BBCCoUkIE): 'subtitles': subtitles, } + # bbc reel (e.g. https://www.bbc.com/reel/video/p07c6sb6/how-positive-thinking-is-harming-your-happiness) + initial_data = self._parse_json(self._html_search_regex( + r'<script[^>]+id=(["\'])initial-data\1[^>]+data-json=(["\'])(?P<json>(?:(?!\2).)+)', + webpage, 'initial data', default='{}', group='json'), playlist_id, fatal=False) + if initial_data: + init_data = try_get( + initial_data, lambda x: x['initData']['items'][0], dict) or {} + smp_data = init_data.get('smpData') or {} + clip_data = try_get(smp_data, lambda x: x['items'][0], dict) or {} + version_id = clip_data.get('versionID') + if version_id: + title = smp_data['title'] + formats, subtitles = self._download_media_selector(version_id) + self._sort_formats(formats) + image_url = smp_data.get('holdingImageURL') + display_date = init_data.get('displayDate') + topic_title = init_data.get('topicTitle') + + return { + 'id': version_id, + 'title': title, + 'formats': formats, + 'alt_title': init_data.get('shortTitle'), + 'thumbnail': image_url.replace('$recipe', 'raw') if image_url else None, + 'description': smp_data.get('summary') or init_data.get('shortSummary'), + 'upload_date': display_date.replace('-', '') if display_date else None, + 'subtitles': subtitles, + 'duration': int_or_none(clip_data.get('duration')), + 'categories': [topic_title] if topic_title else None, + } + # Morph based embed (e.g. http://www.bbc.co.uk/sport/live/olympics/36895975) # There are several setPayload calls may be present but the video # seems to be always related to the first one @@ -1041,7 +1103,7 @@ class BBCIE(BBCCoUkIE): thumbnail = None image_url = current_programme.get('image_url') if image_url: - thumbnail = image_url.replace('{recipe}', '1920x1920') + thumbnail = image_url.replace('{recipe}', 'raw') return { 'id': programme_id, 'title': title, @@ -1114,12 +1176,29 @@ class BBCIE(BBCCoUkIE): continue formats, subtitles = self._download_media_selector(item_id) self._sort_formats(formats) + item_desc = None + blocks = try_get(media, lambda x: x['summary']['blocks'], list) + if blocks: + summary = [] + for block in blocks: + text = try_get(block, lambda x: x['model']['text'], compat_str) + if text: + summary.append(text) + if summary: + item_desc = '\n\n'.join(summary) + item_time = None + for meta in try_get(media, lambda x: x['metadata']['items'], list) or []: + if try_get(meta, lambda x: x['label']) == 'Published': + item_time = unified_timestamp(meta.get('timestamp')) + break entries.append({ 'id': item_id, 'title': item_title, 'thumbnail': item.get('holdingImageUrl'), 'formats': formats, 'subtitles': subtitles, + 'timestamp': item_time, + 'description': strip_or_none(item_desc), }) for resp in (initial_data.get('data') or {}).values(): name = resp.get('name') @@ -1293,21 +1372,149 @@ class BBCCoUkPlaylistBaseIE(InfoExtractor): playlist_id, title, description) -class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE): - IE_NAME = 'bbc.co.uk:iplayer:playlist' - _VALID_URL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/(?:episodes|group)/(?P<id>%s)' % BBCCoUkIE._ID_REGEX - _URL_TEMPLATE = 'http://www.bbc.co.uk/iplayer/episode/%s' - _VIDEO_ID_TEMPLATE = r'data-ip-id=["\'](%s)' +class BBCCoUkIPlayerPlaylistBaseIE(InfoExtractor): + _VALID_URL_TMPL = r'https?://(?:www\.)?bbc\.co\.uk/iplayer/%%s/(?P<id>%s)' % BBCCoUkIE._ID_REGEX + + @staticmethod + def _get_default(episode, key, default_key='default'): + return try_get(episode, lambda x: x[key][default_key]) + + def _get_description(self, data): + synopsis = data.get(self._DESCRIPTION_KEY) or {} + return dict_get(synopsis, ('large', 'medium', 'small')) + + def _fetch_page(self, programme_id, per_page, series_id, page): + elements = self._get_elements(self._call_api( + programme_id, per_page, page + 1, series_id)) + for element in elements: + episode = self._get_episode(element) + episode_id = episode.get('id') + if not episode_id: + continue + thumbnail = None + image = self._get_episode_image(episode) + if image: + thumbnail = image.replace('{recipe}', 'raw') + category = self._get_default(episode, 'labels', 'category') + yield { + '_type': 'url', + 'id': episode_id, + 'title': self._get_episode_field(episode, 'subtitle'), + 'url': 'https://www.bbc.co.uk/iplayer/episode/' + episode_id, + 'thumbnail': thumbnail, + 'description': self._get_description(episode), + 'categories': [category] if category else None, + 'series': self._get_episode_field(episode, 'title'), + 'ie_key': BBCCoUkIE.ie_key(), + } + + def _real_extract(self, url): + pid = self._match_id(url) + qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query) + series_id = qs.get('seriesId', [None])[0] + page = qs.get('page', [None])[0] + per_page = 36 if page else self._PAGE_SIZE + fetch_page = functools.partial(self._fetch_page, pid, per_page, series_id) + entries = fetch_page(int(page) - 1) if page else OnDemandPagedList(fetch_page, self._PAGE_SIZE) + playlist_data = self._get_playlist_data(self._call_api(pid, 1)) + return self.playlist_result( + entries, pid, self._get_playlist_title(playlist_data), + self._get_description(playlist_data)) + + +class BBCCoUkIPlayerEpisodesIE(BBCCoUkIPlayerPlaylistBaseIE): + IE_NAME = 'bbc.co.uk:iplayer:episodes' + _VALID_URL = BBCCoUkIPlayerPlaylistBaseIE._VALID_URL_TMPL % 'episodes' _TESTS = [{ 'url': 'http://www.bbc.co.uk/iplayer/episodes/b05rcz9v', 'info_dict': { 'id': 'b05rcz9v', 'title': 'The Disappearance', - 'description': 'French thriller serial about a missing teenager.', + 'description': 'md5:58eb101aee3116bad4da05f91179c0cb', + }, + 'playlist_mincount': 8, + }, { + # all seasons + 'url': 'https://www.bbc.co.uk/iplayer/episodes/b094m5t9/doctor-foster', + 'info_dict': { + 'id': 'b094m5t9', + 'title': 'Doctor Foster', + 'description': 'md5:5aa9195fad900e8e14b52acd765a9fd6', + }, + 'playlist_mincount': 10, + }, { + # explicit season + 'url': 'https://www.bbc.co.uk/iplayer/episodes/b094m5t9/doctor-foster?seriesId=b094m6nv', + 'info_dict': { + 'id': 'b094m5t9', + 'title': 'Doctor Foster', + 'description': 'md5:5aa9195fad900e8e14b52acd765a9fd6', }, - 'playlist_mincount': 6, - 'skip': 'This programme is not currently available on BBC iPlayer', + 'playlist_mincount': 5, }, { + # all pages + 'url': 'https://www.bbc.co.uk/iplayer/episodes/m0004c4v/beechgrove', + 'info_dict': { + 'id': 'm0004c4v', + 'title': 'Beechgrove', + 'description': 'Gardening show that celebrates Scottish horticulture and growing conditions.', + }, + 'playlist_mincount': 37, + }, { + # explicit page + 'url': 'https://www.bbc.co.uk/iplayer/episodes/m0004c4v/beechgrove?page=2', + 'info_dict': { + 'id': 'm0004c4v', + 'title': 'Beechgrove', + 'description': 'Gardening show that celebrates Scottish horticulture and growing conditions.', + }, + 'playlist_mincount': 1, + }] + _PAGE_SIZE = 100 + _DESCRIPTION_KEY = 'synopsis' + + def _get_episode_image(self, episode): + return self._get_default(episode, 'image') + + def _get_episode_field(self, episode, field): + return self._get_default(episode, field) + + @staticmethod + def _get_elements(data): + return data['entities']['results'] + + @staticmethod + def _get_episode(element): + return element.get('episode') or {} + + def _call_api(self, pid, per_page, page=1, series_id=None): + variables = { + 'id': pid, + 'page': page, + 'perPage': per_page, + } + if series_id: + variables['sliceId'] = series_id + return self._download_json( + 'https://graph.ibl.api.bbc.co.uk/', pid, headers={ + 'Content-Type': 'application/json' + }, data=json.dumps({ + 'id': '5692d93d5aac8d796a0305e895e61551', + 'variables': variables, + }).encode('utf-8'))['data']['programme'] + + @staticmethod + def _get_playlist_data(data): + return data + + def _get_playlist_title(self, data): + return self._get_default(data, 'title') + + +class BBCCoUkIPlayerGroupIE(BBCCoUkIPlayerPlaylistBaseIE): + IE_NAME = 'bbc.co.uk:iplayer:group' + _VALID_URL = BBCCoUkIPlayerPlaylistBaseIE._VALID_URL_TMPL % 'group' + _TESTS = [{ # Available for over a year unlike 30 days for most other programmes 'url': 'http://www.bbc.co.uk/iplayer/group/p02tcc32', 'info_dict': { @@ -1316,14 +1523,56 @@ class BBCCoUkIPlayerPlaylistIE(BBCCoUkPlaylistBaseIE): 'description': 'md5:683e901041b2fe9ba596f2ab04c4dbe7', }, 'playlist_mincount': 10, + }, { + # all pages + 'url': 'https://www.bbc.co.uk/iplayer/group/p081d7j7', + 'info_dict': { + 'id': 'p081d7j7', + 'title': 'Music in Scotland', + 'description': 'Perfomances in Scotland and programmes featuring Scottish acts.', + }, + 'playlist_mincount': 47, + }, { + # explicit page + 'url': 'https://www.bbc.co.uk/iplayer/group/p081d7j7?page=2', + 'info_dict': { + 'id': 'p081d7j7', + 'title': 'Music in Scotland', + 'description': 'Perfomances in Scotland and programmes featuring Scottish acts.', + }, + 'playlist_mincount': 11, }] + _PAGE_SIZE = 200 + _DESCRIPTION_KEY = 'synopses' - def _extract_title_and_description(self, webpage): - title = self._search_regex(r'<h1>([^<]+)</h1>', webpage, 'title', fatal=False) - description = self._search_regex( - r'<p[^>]+class=(["\'])subtitle\1[^>]*>(?P<value>[^<]+)</p>', - webpage, 'description', fatal=False, group='value') - return title, description + def _get_episode_image(self, episode): + return self._get_default(episode, 'images', 'standard') + + def _get_episode_field(self, episode, field): + return episode.get(field) + + @staticmethod + def _get_elements(data): + return data['elements'] + + @staticmethod + def _get_episode(element): + return element + + def _call_api(self, pid, per_page, page=1, series_id=None): + return self._download_json( + 'http://ibl.api.bbc.co.uk/ibl/v1/groups/%s/episodes' % pid, + pid, query={ + 'page': page, + 'per_page': per_page, + })['group_episodes'] + + @staticmethod + def _get_playlist_data(data): + return data['group'] + + def _get_playlist_title(self, data): + return data.get('title') class BBCCoUkPlaylistIE(BBCCoUkPlaylistBaseIE): diff --git a/youtube_dl/extractor/bilibili.py b/youtube_dl/extractor/bilibili.py index 4dc597e16..bff6ea194 100644 --- a/youtube_dl/extractor/bilibili.py +++ b/youtube_dl/extractor/bilibili.py @@ -156,6 +156,7 @@ class BiliBiliIE(InfoExtractor): cid = js['result']['cid'] headers = { + 'Accept': 'application/json', 'Referer': url } headers.update(self.geo_verification_headers()) @@ -232,7 +233,7 @@ class BiliBiliIE(InfoExtractor): webpage) if uploader_mobj: info.update({ - 'uploader': uploader_mobj.group('name'), + 'uploader': uploader_mobj.group('name').strip(), 'uploader_id': uploader_mobj.group('id'), }) if not info.get('uploader'): diff --git a/youtube_dl/extractor/bleacherreport.py b/youtube_dl/extractor/bleacherreport.py index dc60224d0..d1bf8e829 100644 --- a/youtube_dl/extractor/bleacherreport.py +++ b/youtube_dl/extractor/bleacherreport.py @@ -90,13 +90,19 @@ class BleacherReportCMSIE(AMPIE): _VALID_URL = r'https?://(?:www\.)?bleacherreport\.com/video_embed\?id=(?P<id>[0-9a-f-]{36}|\d{5})' _TESTS = [{ 'url': 'http://bleacherreport.com/video_embed?id=8fd44c2f-3dc5-4821-9118-2c825a98c0e1&library=video-cms', - 'md5': '2e4b0a997f9228ffa31fada5c53d1ed1', + 'md5': '670b2d73f48549da032861130488c681', 'info_dict': { 'id': '8fd44c2f-3dc5-4821-9118-2c825a98c0e1', - 'ext': 'flv', + 'ext': 'mp4', 'title': 'Cena vs. Rollins Would Expose the Heavyweight Division', 'description': 'md5:984afb4ade2f9c0db35f3267ed88b36e', + 'upload_date': '20150723', + 'timestamp': 1437679032, + }, + 'expected_warnings': [ + 'Unable to download f4m manifest' + ] }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/blinkx.py b/youtube_dl/extractor/blinkx.py deleted file mode 100644 index db5e12b21..000000000 --- a/youtube_dl/extractor/blinkx.py +++ /dev/null @@ -1,86 +0,0 @@ -from __future__ import unicode_literals - -import json - -from .common import InfoExtractor -from ..utils import ( - remove_start, - int_or_none, -) - - -class BlinkxIE(InfoExtractor): - _VALID_URL = r'(?:https?://(?:www\.)blinkx\.com/#?ce/|blinkx:)(?P<id>[^?]+)' - IE_NAME = 'blinkx' - - _TEST = { - 'url': 'http://www.blinkx.com/ce/Da0Gw3xc5ucpNduzLuDDlv4WC9PuI4fDi1-t6Y3LyfdY2SZS5Urbvn-UPJvrvbo8LTKTc67Wu2rPKSQDJyZeeORCR8bYkhs8lI7eqddznH2ofh5WEEdjYXnoRtj7ByQwt7atMErmXIeYKPsSDuMAAqJDlQZ-3Ff4HJVeH_s3Gh8oQ', - 'md5': '337cf7a344663ec79bf93a526a2e06c7', - 'info_dict': { - 'id': 'Da0Gw3xc', - 'ext': 'mp4', - 'title': 'No Daily Show for John Oliver; HBO Show Renewed - IGN News', - 'uploader': 'IGN News', - 'upload_date': '20150217', - 'timestamp': 1424215740, - 'description': 'HBO has renewed Last Week Tonight With John Oliver for two more seasons.', - 'duration': 47.743333, - }, - } - - def _real_extract(self, url): - video_id = self._match_id(url) - display_id = video_id[:8] - - api_url = ('https://apib4.blinkx.com/api.php?action=play_video&' - + 'video=%s' % video_id) - data_json = self._download_webpage(api_url, display_id) - data = json.loads(data_json)['api']['results'][0] - duration = None - thumbnails = [] - formats = [] - for m in data['media']: - if m['type'] == 'jpg': - thumbnails.append({ - 'url': m['link'], - 'width': int(m['w']), - 'height': int(m['h']), - }) - elif m['type'] == 'original': - duration = float(m['d']) - elif m['type'] == 'youtube': - yt_id = m['link'] - self.to_screen('Youtube video detected: %s' % yt_id) - return self.url_result(yt_id, 'Youtube', video_id=yt_id) - elif m['type'] in ('flv', 'mp4'): - vcodec = remove_start(m['vcodec'], 'ff') - acodec = remove_start(m['acodec'], 'ff') - vbr = int_or_none(m.get('vbr') or m.get('vbitrate'), 1000) - abr = int_or_none(m.get('abr') or m.get('abitrate'), 1000) - tbr = vbr + abr if vbr and abr else None - format_id = '%s-%sk-%s' % (vcodec, tbr, m['w']) - formats.append({ - 'format_id': format_id, - 'url': m['link'], - 'vcodec': vcodec, - 'acodec': acodec, - 'abr': abr, - 'vbr': vbr, - 'tbr': tbr, - 'width': int_or_none(m.get('w')), - 'height': int_or_none(m.get('h')), - }) - - self._sort_formats(formats) - - return { - 'id': display_id, - 'fullid': video_id, - 'title': data['title'], - 'formats': formats, - 'uploader': data['channel_name'], - 'timestamp': data['pubdate_epoch'], - 'description': data.get('description'), - 'thumbnails': thumbnails, - 'duration': duration, - } diff --git a/youtube_dl/extractor/bravotv.py b/youtube_dl/extractor/bravotv.py index b9715df00..bae2aedce 100644 --- a/youtube_dl/extractor/bravotv.py +++ b/youtube_dl/extractor/bravotv.py @@ -12,7 +12,7 @@ from ..utils import ( class BravoTVIE(AdobePassIE): - _VALID_URL = r'https?://(?:www\.)?bravotv\.com/(?:[^/]+/)+(?P<id>[^/?#]+)' + _VALID_URL = r'https?://(?:www\.)?(?P<req_id>bravotv|oxygen)\.com/(?:[^/]+/)+(?P<id>[^/?#]+)' _TESTS = [{ 'url': 'https://www.bravotv.com/top-chef/season-16/episode-15/videos/the-top-chef-season-16-winner-is', 'md5': 'e34684cfea2a96cd2ee1ef3a60909de9', @@ -28,10 +28,13 @@ class BravoTVIE(AdobePassIE): }, { 'url': 'http://www.bravotv.com/below-deck/season-3/ep-14-reunion-part-1', 'only_matching': True, + }, { + 'url': 'https://www.oxygen.com/in-ice-cold-blood/season-2/episode-16/videos/handling-the-horwitz-house-after-the-murder-season-2', + 'only_matching': True, }] def _real_extract(self, url): - display_id = self._match_id(url) + site, display_id = re.match(self._VALID_URL, url).groups() webpage = self._download_webpage(url, display_id) settings = self._parse_json(self._search_regex( r'<script[^>]+data-drupal-selector="drupal-settings-json"[^>]*>({.+?})</script>', webpage, 'drupal settings'), @@ -53,11 +56,14 @@ class BravoTVIE(AdobePassIE): tp_path = release_pid = tve['release_pid'] if tve.get('entitlement') == 'auth': adobe_pass = settings.get('tve_adobe_auth', {}) + if site == 'bravotv': + site = 'bravo' resource = self._get_mvpd_resource( - adobe_pass.get('adobePassResourceId', 'bravo'), + adobe_pass.get('adobePassResourceId') or site, tve['title'], release_pid, tve.get('rating')) query['auth'] = self._extract_mvpd_auth( - url, release_pid, adobe_pass.get('adobePassRequestorId', 'bravo'), resource) + url, release_pid, + adobe_pass.get('adobePassRequestorId') or site, resource) else: shared_playlist = settings['ls_playlist'] account_pid = shared_playlist['account_pid'] diff --git a/youtube_dl/extractor/canvas.py b/youtube_dl/extractor/canvas.py index 8b76a0200..eefbab241 100644 --- a/youtube_dl/extractor/canvas.py +++ b/youtube_dl/extractor/canvas.py @@ -7,19 +7,21 @@ from .common import InfoExtractor from .gigya import GigyaBaseIE from ..compat import compat_HTTPError from ..utils import ( - extract_attributes, ExtractorError, - strip_or_none, + clean_html, + extract_attributes, float_or_none, + get_element_by_class, int_or_none, merge_dicts, str_or_none, + strip_or_none, url_or_none, ) class CanvasIE(InfoExtractor): - _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet|vrt(?:video|nieuws)|sporza)/assets/(?P<id>[^/?#&]+)' + _VALID_URL = r'https?://mediazone\.vrt\.be/api/v1/(?P<site_id>canvas|een|ketnet|vrt(?:video|nieuws)|sporza|dako)/assets/(?P<id>[^/?#&]+)' _TESTS = [{ 'url': 'https://mediazone.vrt.be/api/v1/ketnet/assets/md-ast-4ac54990-ce66-4d00-a8ca-9eac86f4c475', 'md5': '68993eda72ef62386a15ea2cf3c93107', @@ -332,3 +334,51 @@ class VrtNUIE(GigyaBaseIE): 'display_id': display_id, 'season_number': int_or_none(page.get('episode_season')), }) + + +class DagelijkseKostIE(InfoExtractor): + IE_DESC = 'dagelijksekost.een.be' + _VALID_URL = r'https?://dagelijksekost\.een\.be/gerechten/(?P<id>[^/?#&]+)' + _TEST = { + 'url': 'https://dagelijksekost.een.be/gerechten/hachis-parmentier-met-witloof', + 'md5': '30bfffc323009a3e5f689bef6efa2365', + 'info_dict': { + 'id': 'md-ast-27a4d1ff-7d7b-425e-b84f-a4d227f592fa', + 'display_id': 'hachis-parmentier-met-witloof', + 'ext': 'mp4', + 'title': 'Hachis parmentier met witloof', + 'description': 'md5:9960478392d87f63567b5b117688cdc5', + 'thumbnail': r're:^https?://.*\.jpg$', + 'duration': 283.02, + }, + 'expected_warnings': ['is not a supported codec'], + } + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + + title = strip_or_none(get_element_by_class( + 'dish-metadata__title', webpage + ) or self._html_search_meta( + 'twitter:title', webpage)) + + description = clean_html(get_element_by_class( + 'dish-description', webpage) + ) or self._html_search_meta( + ('description', 'twitter:description', 'og:description'), + webpage) + + video_id = self._html_search_regex( + r'data-url=(["\'])(?P<id>(?:(?!\1).)+)\1', webpage, 'video id', + group='id') + + return { + '_type': 'url_transparent', + 'url': 'https://mediazone.vrt.be/api/v1/dako/assets/%s' % video_id, + 'ie_key': CanvasIE.ie_key(), + 'id': video_id, + 'display_id': display_id, + 'title': title, + 'description': description, + } diff --git a/youtube_dl/extractor/cbs.py b/youtube_dl/extractor/cbs.py index 4a19a73d2..c79e55a75 100644 --- a/youtube_dl/extractor/cbs.py +++ b/youtube_dl/extractor/cbs.py @@ -27,7 +27,7 @@ class CBSBaseIE(ThePlatformFeedIE): class CBSIE(CBSBaseIE): - _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:cbs\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)' + _VALID_URL = r'(?:cbs:|https?://(?:www\.)?(?:(?:cbs|paramountplus)\.com/shows/[^/]+/video|colbertlateshow\.com/(?:video|podcasts))/)(?P<id>[\w-]+)' _TESTS = [{ 'url': 'http://www.cbs.com/shows/garth-brooks/video/_u7W953k6la293J7EPTd9oHkSPs6Xn6_/connect-chat-feat-garth-brooks/', @@ -52,6 +52,9 @@ class CBSIE(CBSBaseIE): }, { 'url': 'http://www.colbertlateshow.com/podcasts/dYSwjqPs_X1tvbV_P2FcPWRa_qT6akTC/in-the-bad-room-with-stephen/', 'only_matching': True, + }, { + 'url': 'https://www.paramountplus.com/shows/all-rise/video/QmR1WhNkh1a_IrdHZrbcRklm176X_rVc/all-rise-space/', + 'only_matching': True, }] def _extract_video_info(self, content_id, site='cbs', mpx_acc=2198311517): diff --git a/youtube_dl/extractor/cbsnews.py b/youtube_dl/extractor/cbsnews.py index 345debcf0..1285ed65e 100644 --- a/youtube_dl/extractor/cbsnews.py +++ b/youtube_dl/extractor/cbsnews.py @@ -26,7 +26,7 @@ class CBSNewsEmbedIE(CBSIE): def _real_extract(self, url): item = self._parse_json(zlib.decompress(compat_b64decode( compat_urllib_parse_unquote(self._match_id(url))), - -zlib.MAX_WBITS), None)['video']['items'][0] + -zlib.MAX_WBITS).decode('utf-8'), None)['video']['items'][0] return self._extract_video_info(item['mpxRefId'], 'cbsnews') diff --git a/youtube_dl/extractor/cbssports.py b/youtube_dl/extractor/cbssports.py index 83b764762..a891c9a55 100644 --- a/youtube_dl/extractor/cbssports.py +++ b/youtube_dl/extractor/cbssports.py @@ -1,38 +1,113 @@ from __future__ import unicode_literals -from .cbs import CBSBaseIE +import re +# from .cbs import CBSBaseIE +from .common import InfoExtractor +from ..utils import ( + int_or_none, + try_get, +) -class CBSSportsIE(CBSBaseIE): - _VALID_URL = r'https?://(?:www\.)?cbssports\.com/[^/]+/(?:video|news)/(?P<id>[^/?#&]+)' +# class CBSSportsEmbedIE(CBSBaseIE): +class CBSSportsEmbedIE(InfoExtractor): + IE_NAME = 'cbssports:embed' + _VALID_URL = r'''(?ix)https?://(?:(?:www\.)?cbs|embed\.247)sports\.com/player/embed.+? + (?: + ids%3D(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})| + pcid%3D(?P<pcid>\d+) + )''' _TESTS = [{ - 'url': 'https://www.cbssports.com/nba/video/donovan-mitchell-flashes-star-potential-in-game-2-victory-over-thunder/', - 'info_dict': { - 'id': '1214315075735', - 'ext': 'mp4', - 'title': 'Donovan Mitchell flashes star potential in Game 2 victory over Thunder', - 'description': 'md5:df6f48622612c2d6bd2e295ddef58def', - 'timestamp': 1524111457, - 'upload_date': '20180419', - 'uploader': 'CBSI-NEW', - }, - 'params': { - # m3u8 download - 'skip_download': True, - } + 'url': 'https://www.cbssports.com/player/embed/?args=player_id%3Db56c03a6-231a-4bbe-9c55-af3c8a8e9636%26ids%3Db56c03a6-231a-4bbe-9c55-af3c8a8e9636%26resizable%3D1%26autoplay%3Dtrue%26domain%3Dcbssports.com%26comp_ads_enabled%3Dfalse%26watchAndRead%3D0%26startTime%3D0%26env%3Dprod', + 'only_matching': True, }, { - 'url': 'https://www.cbssports.com/nba/news/nba-playoffs-2018-watch-76ers-vs-heat-game-3-series-schedule-tv-channel-online-stream/', + 'url': 'https://embed.247sports.com/player/embed/?args=%3fplayer_id%3d1827823171591%26channel%3dcollege-football-recruiting%26pcid%3d1827823171591%26width%3d640%26height%3d360%26autoplay%3dTrue%26comp_ads_enabled%3dFalse%26uvpc%3dhttps%253a%252f%252fwww.cbssports.com%252fapi%252fcontent%252fvideo%252fconfig%252f%253fcfg%253duvp_247sports_v4%2526partner%253d247%26uvpc_m%3dhttps%253a%252f%252fwww.cbssports.com%252fapi%252fcontent%252fvideo%252fconfig%252f%253fcfg%253duvp_247sports_m_v4%2526partner_m%253d247_mobile%26utag%3d247sportssite%26resizable%3dTrue', 'only_matching': True, }] - def _extract_video_info(self, filter_query, video_id): - return self._extract_feed_info('dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id) + # def _extract_video_info(self, filter_query, video_id): + # return self._extract_feed_info('dJ5BDC', 'VxxJg8Ymh8sE', filter_query, video_id) def _real_extract(self, url): + uuid, pcid = re.match(self._VALID_URL, url).groups() + query = {'id': uuid} if uuid else {'pcid': pcid} + video = self._download_json( + 'https://www.cbssports.com/api/content/video/', + uuid or pcid, query=query)[0] + video_id = video['id'] + title = video['title'] + metadata = video.get('metaData') or {} + # return self._extract_video_info('byId=%d' % metadata['mpxOutletId'], video_id) + # return self._extract_video_info('byGuid=' + metadata['mpxRefId'], video_id) + + formats = self._extract_m3u8_formats( + metadata['files'][0]['url'], video_id, 'mp4', + 'm3u8_native', m3u8_id='hls', fatal=False) + self._sort_formats(formats) + + image = video.get('image') + thumbnails = None + if image: + image_path = image.get('path') + if image_path: + thumbnails = [{ + 'url': image_path, + 'width': int_or_none(image.get('width')), + 'height': int_or_none(image.get('height')), + 'filesize': int_or_none(image.get('size')), + }] + + return { + 'id': video_id, + 'title': title, + 'formats': formats, + 'thumbnails': thumbnails, + 'description': video.get('description'), + 'timestamp': int_or_none(try_get(video, lambda x: x['dateCreated']['epoch'])), + 'duration': int_or_none(metadata.get('duration')), + } + + +class CBSSportsBaseIE(InfoExtractor): + def _real_extract(self, url): display_id = self._match_id(url) webpage = self._download_webpage(url, display_id) - video_id = self._search_regex( - [r'(?:=|%26)pcid%3D(\d+)', r'embedVideo(?:Container)?_(\d+)'], - webpage, 'video id') - return self._extract_video_info('byId=%s' % video_id, video_id) + iframe_url = self._search_regex( + r'<iframe[^>]+(?:data-)?src="(https?://[^/]+/player/embed[^"]+)"', + webpage, 'embed url') + return self.url_result(iframe_url, CBSSportsEmbedIE.ie_key()) + + +class CBSSportsIE(CBSSportsBaseIE): + IE_NAME = 'cbssports' + _VALID_URL = r'https?://(?:www\.)?cbssports\.com/[^/]+/video/(?P<id>[^/?#&]+)' + _TESTS = [{ + 'url': 'https://www.cbssports.com/college-football/video/cover-3-stanford-spring-gleaning/', + 'info_dict': { + 'id': 'b56c03a6-231a-4bbe-9c55-af3c8a8e9636', + 'ext': 'mp4', + 'title': 'Cover 3: Stanford Spring Gleaning', + 'description': 'The Cover 3 crew break down everything you need to know about the Stanford Cardinal this spring.', + 'timestamp': 1617218398, + 'upload_date': '20210331', + 'duration': 502, + }, + }] + + +class TwentyFourSevenSportsIE(CBSSportsBaseIE): + IE_NAME = '247sports' + _VALID_URL = r'https?://(?:www\.)?247sports\.com/Video/(?:[^/?#&]+-)?(?P<id>\d+)' + _TESTS = [{ + 'url': 'https://247sports.com/Video/2021-QB-Jake-Garcia-senior-highlights-through-five-games-10084854/', + 'info_dict': { + 'id': '4f1265cb-c3b5-44a8-bb1d-1914119a0ccc', + 'ext': 'mp4', + 'title': '2021 QB Jake Garcia senior highlights through five games', + 'description': 'md5:8cb67ebed48e2e6adac1701e0ff6e45b', + 'timestamp': 1607114223, + 'upload_date': '20201204', + 'duration': 208, + }, + }] diff --git a/youtube_dl/extractor/ccma.py b/youtube_dl/extractor/ccma.py index 544647f92..e6ae49352 100644 --- a/youtube_dl/extractor/ccma.py +++ b/youtube_dl/extractor/ccma.py @@ -1,15 +1,18 @@ # coding: utf-8 from __future__ import unicode_literals +import calendar +import datetime import re from .common import InfoExtractor from ..utils import ( clean_html, + extract_timezone, int_or_none, parse_duration, - parse_iso8601, parse_resolution, + try_get, url_or_none, ) @@ -24,8 +27,9 @@ class CCMAIE(InfoExtractor): 'ext': 'mp4', 'title': 'L\'espot de La Marató de TV3', 'description': 'md5:f12987f320e2f6e988e9908e4fe97765', - 'timestamp': 1470918540, - 'upload_date': '20160811', + 'timestamp': 1478608140, + 'upload_date': '20161108', + 'age_limit': 0, } }, { 'url': 'http://www.ccma.cat/catradio/alacarta/programa/el-consell-de-savis-analitza-el-derbi/audio/943685/', @@ -35,8 +39,24 @@ class CCMAIE(InfoExtractor): 'ext': 'mp3', 'title': 'El Consell de Savis analitza el derbi', 'description': 'md5:e2a3648145f3241cb9c6b4b624033e53', - 'upload_date': '20171205', - 'timestamp': 1512507300, + 'upload_date': '20170512', + 'timestamp': 1494622500, + 'vcodec': 'none', + 'categories': ['Esports'], + } + }, { + 'url': 'http://www.ccma.cat/tv3/alacarta/crims/crims-josep-tallada-lespereu-me-capitol-1/video/6031387/', + 'md5': 'b43c3d3486f430f3032b5b160d80cbc3', + 'info_dict': { + 'id': '6031387', + 'ext': 'mp4', + 'title': 'Crims - Josep Talleda, l\'"Espereu-me" (capítol 1)', + 'description': 'md5:7cbdafb640da9d0d2c0f62bad1e74e60', + 'timestamp': 1582577700, + 'upload_date': '20200224', + 'subtitles': 'mincount:4', + 'age_limit': 16, + 'series': 'Crims', } }] @@ -72,17 +92,28 @@ class CCMAIE(InfoExtractor): informacio = media['informacio'] title = informacio['titol'] - durada = informacio.get('durada', {}) + durada = informacio.get('durada') or {} duration = int_or_none(durada.get('milisegons'), 1000) or parse_duration(durada.get('text')) - timestamp = parse_iso8601(informacio.get('data_emissio', {}).get('utc')) + tematica = try_get(informacio, lambda x: x['tematica']['text']) + + timestamp = None + data_utc = try_get(informacio, lambda x: x['data_emissio']['utc']) + try: + timezone, data_utc = extract_timezone(data_utc) + timestamp = calendar.timegm((datetime.datetime.strptime( + data_utc, '%Y-%d-%mT%H:%M:%S') - timezone).timetuple()) + except TypeError: + pass subtitles = {} - subtitols = media.get('subtitols', {}) - if subtitols: - sub_url = subtitols.get('url') + subtitols = media.get('subtitols') or [] + if isinstance(subtitols, dict): + subtitols = [subtitols] + for st in subtitols: + sub_url = st.get('url') if sub_url: subtitles.setdefault( - subtitols.get('iso') or subtitols.get('text') or 'ca', []).append({ + st.get('iso') or st.get('text') or 'ca', []).append({ 'url': sub_url, }) @@ -97,6 +128,16 @@ class CCMAIE(InfoExtractor): 'height': int_or_none(imatges.get('alcada')), }] + age_limit = None + codi_etic = try_get(informacio, lambda x: x['codi_etic']['id']) + if codi_etic: + codi_etic_s = codi_etic.split('_') + if len(codi_etic_s) == 2: + if codi_etic_s[1] == 'TP': + age_limit = 0 + else: + age_limit = int_or_none(codi_etic_s[1]) + return { 'id': media_id, 'title': title, @@ -106,4 +147,9 @@ class CCMAIE(InfoExtractor): 'thumbnails': thumbnails, 'subtitles': subtitles, 'formats': formats, + 'age_limit': age_limit, + 'alt_title': informacio.get('titol_complet'), + 'episode_number': int_or_none(informacio.get('capitol')), + 'categories': [tematica] if tematica else None, + 'series': informacio.get('programa'), } diff --git a/youtube_dl/extractor/cda.py b/youtube_dl/extractor/cda.py index d67900e62..e1b391937 100644 --- a/youtube_dl/extractor/cda.py +++ b/youtube_dl/extractor/cda.py @@ -95,8 +95,11 @@ class CDAIE(InfoExtractor): if 'Ten film jest dostępny dla użytkowników premium' in webpage: raise ExtractorError('This video is only available for premium users.', expected=True) + if re.search(r'niedostępn[ey] w(?: |\s+)Twoim kraju\s*<', webpage): + self.raise_geo_restricted() + need_confirm_age = False - if self._html_search_regex(r'(<form[^>]+action="/a/validatebirth")', + if self._html_search_regex(r'(<form[^>]+action="[^"]*/a/validatebirth[^"]*")', webpage, 'birthday validate form', default=None): webpage = self._download_age_confirm_page( url, video_id, note='Confirming age') @@ -130,6 +133,8 @@ class CDAIE(InfoExtractor): 'age_limit': 18 if need_confirm_age else 0, } + info = self._search_json_ld(webpage, video_id, default={}) + # Source: https://www.cda.pl/js/player.js?t=1606154898 def decrypt_file(a): for p in ('_XDDD', '_CDA', '_ADC', '_CXD', '_QWE', '_Q5', '_IKSDE'): @@ -194,7 +199,7 @@ class CDAIE(InfoExtractor): handler = self._download_webpage webpage = handler( - self._BASE_URL + href, video_id, + urljoin(self._BASE_URL, href), video_id, 'Downloading %s version information' % resolution, fatal=False) if not webpage: # Manually report warning because empty page is returned when @@ -206,6 +211,4 @@ class CDAIE(InfoExtractor): self._sort_formats(formats) - info = self._search_json_ld(webpage, video_id, default={}) - return merge_dicts(info_dict, info) diff --git a/youtube_dl/extractor/comedycentral.py b/youtube_dl/extractor/comedycentral.py index d08b909a6..1bfa912be 100644 --- a/youtube_dl/extractor/comedycentral.py +++ b/youtube_dl/extractor/comedycentral.py @@ -1,142 +1,51 @@ from __future__ import unicode_literals from .mtv import MTVServicesInfoExtractor -from .common import InfoExtractor class ComedyCentralIE(MTVServicesInfoExtractor): - _VALID_URL = r'''(?x)https?://(?:www\.)?cc\.com/ - (video-clips|episodes|cc-studios|video-collections|shows(?=/[^/]+/(?!full-episodes))) - /(?P<title>.*)''' + _VALID_URL = r'https?://(?:www\.)?cc\.com/(?:episodes|video(?:-clips)?)/(?P<id>[0-9a-z]{6})' _FEED_URL = 'http://comedycentral.com/feeds/mrss/' _TESTS = [{ - 'url': 'http://www.cc.com/video-clips/kllhuv/stand-up-greg-fitzsimmons--uncensored---too-good-of-a-mother', - 'md5': 'c4f48e9eda1b16dd10add0744344b6d8', + 'url': 'http://www.cc.com/video-clips/5ke9v2/the-daily-show-with-trevor-noah-doc-rivers-and-steve-ballmer---the-nba-player-strike', + 'md5': 'b8acb347177c680ff18a292aa2166f80', 'info_dict': { - 'id': 'cef0cbb3-e776-4bc9-b62e-8016deccb354', + 'id': '89ccc86e-1b02-4f83-b0c9-1d9592ecd025', 'ext': 'mp4', - 'title': 'CC:Stand-Up|August 18, 2013|1|0101|Uncensored - Too Good of a Mother', - 'description': 'After a certain point, breastfeeding becomes c**kblocking.', - 'timestamp': 1376798400, - 'upload_date': '20130818', + 'title': 'The Daily Show with Trevor Noah|August 28, 2020|25|25149|Doc Rivers and Steve Ballmer - The NBA Player Strike', + 'description': 'md5:5334307c433892b85f4f5e5ac9ef7498', + 'timestamp': 1598670000, + 'upload_date': '20200829', }, }, { - 'url': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/interviews/6yx39d/exclusive-rand-paul-extended-interview', + 'url': 'http://www.cc.com/episodes/pnzzci/drawn-together--american-idol--parody-clip-show-season-3-ep-314', 'only_matching': True, - }] - - -class ComedyCentralFullEpisodesIE(MTVServicesInfoExtractor): - _VALID_URL = r'''(?x)https?://(?:www\.)?cc\.com/ - (?:full-episodes|shows(?=/[^/]+/full-episodes)) - /(?P<id>[^?]+)''' - _FEED_URL = 'http://comedycentral.com/feeds/mrss/' - - _TESTS = [{ - 'url': 'http://www.cc.com/full-episodes/pv391a/the-daily-show-with-trevor-noah-november-28--2016---ryan-speedo-green-season-22-ep-22028', - 'info_dict': { - 'description': 'Donald Trump is accused of exploiting his president-elect status for personal gain, Cuban leader Fidel Castro dies, and Ryan Speedo Green discusses "Sing for Your Life."', - 'title': 'November 28, 2016 - Ryan Speedo Green', - }, - 'playlist_count': 4, }, { - 'url': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes', - 'only_matching': True, - }] - - def _real_extract(self, url): - playlist_id = self._match_id(url) - webpage = self._download_webpage(url, playlist_id) - mgid = self._extract_triforce_mgid(webpage, data_zone='t2_lc_promo1') - videos_info = self._get_videos_info(mgid) - return videos_info - - -class ToshIE(MTVServicesInfoExtractor): - IE_DESC = 'Tosh.0' - _VALID_URL = r'^https?://tosh\.cc\.com/video-(?:clips|collections)/[^/]+/(?P<videotitle>[^/?#]+)' - _FEED_URL = 'http://tosh.cc.com/feeds/mrss' - - _TESTS = [{ - 'url': 'http://tosh.cc.com/video-clips/68g93d/twitter-users-share-summer-plans', - 'info_dict': { - 'description': 'Tosh asked fans to share their summer plans.', - 'title': 'Twitter Users Share Summer Plans', - }, - 'playlist': [{ - 'md5': 'f269e88114c1805bb6d7653fecea9e06', - 'info_dict': { - 'id': '90498ec2-ed00-11e0-aca6-0026b9414f30', - 'ext': 'mp4', - 'title': 'Tosh.0|June 9, 2077|2|211|Twitter Users Share Summer Plans', - 'description': 'Tosh asked fans to share their summer plans.', - 'thumbnail': r're:^https?://.*\.jpg', - # It's really reported to be published on year 2077 - 'upload_date': '20770610', - 'timestamp': 3390510600, - 'subtitles': { - 'en': 'mincount:3', - }, - }, - }] - }, { - 'url': 'http://tosh.cc.com/video-collections/x2iz7k/just-plain-foul/m5q4fp', + 'url': 'https://www.cc.com/video/k3sdvm/the-daily-show-with-jon-stewart-exclusive-the-fourth-estate', 'only_matching': True, }] class ComedyCentralTVIE(MTVServicesInfoExtractor): - _VALID_URL = r'https?://(?:www\.)?comedycentral\.tv/(?:staffeln|shows)/(?P<id>[^/?#&]+)' + _VALID_URL = r'https?://(?:www\.)?comedycentral\.tv/folgen/(?P<id>[0-9a-z]{6})' _TESTS = [{ - 'url': 'http://www.comedycentral.tv/staffeln/7436-the-mindy-project-staffel-4', + 'url': 'https://www.comedycentral.tv/folgen/pxdpec/josh-investigates-klimawandel-staffel-1-ep-1', 'info_dict': { - 'id': 'local_playlist-f99b626bdfe13568579a', - 'ext': 'flv', - 'title': 'Episode_the-mindy-project_shows_season-4_episode-3_full-episode_part1', - }, - 'params': { - # rtmp download - 'skip_download': True, + 'id': '15907dc3-ec3c-11e8-a442-0e40cf2fc285', + 'ext': 'mp4', + 'title': 'Josh Investigates', + 'description': 'Steht uns das Ende der Welt bevor?', }, - }, { - 'url': 'http://www.comedycentral.tv/shows/1074-workaholics', - 'only_matching': True, - }, { - 'url': 'http://www.comedycentral.tv/shows/1727-the-mindy-project/bonus', - 'only_matching': True, }] - - def _real_extract(self, url): - video_id = self._match_id(url) - - webpage = self._download_webpage(url, video_id) - - mrss_url = self._search_regex( - r'data-mrss=(["\'])(?P<url>(?:(?!\1).)+)\1', - webpage, 'mrss url', group='url') - - return self._get_videos_info_from_url(mrss_url, video_id) - - -class ComedyCentralShortnameIE(InfoExtractor): - _VALID_URL = r'^:(?P<id>tds|thedailyshow|theopposition)$' - _TESTS = [{ - 'url': ':tds', - 'only_matching': True, - }, { - 'url': ':thedailyshow', - 'only_matching': True, - }, { - 'url': ':theopposition', - 'only_matching': True, - }] - - def _real_extract(self, url): - video_id = self._match_id(url) - shortcut_map = { - 'tds': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes', - 'thedailyshow': 'http://www.cc.com/shows/the-daily-show-with-trevor-noah/full-episodes', - 'theopposition': 'http://www.cc.com/shows/the-opposition-with-jordan-klepper/full-episodes', + _FEED_URL = 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed' + _GEO_COUNTRIES = ['DE'] + + def _get_feed_query(self, uri): + return { + 'accountOverride': 'intl.mtvi.com', + 'arcEp': 'web.cc.tv', + 'ep': 'b9032c3a', + 'imageEp': 'web.cc.tv', + 'mgid': uri, } - return self.url_result(shortcut_map[video_id]) diff --git a/youtube_dl/extractor/common.py b/youtube_dl/extractor/common.py index d5faa0eb7..797c35fd5 100644 --- a/youtube_dl/extractor/common.py +++ b/youtube_dl/extractor/common.py @@ -17,7 +17,7 @@ import math from ..compat import ( compat_cookiejar_Cookie, - compat_cookies, + compat_cookies_SimpleCookie, compat_etree_Element, compat_etree_fromstring, compat_getpass, @@ -230,8 +230,10 @@ class InfoExtractor(object): uploader: Full name of the video uploader. license: License name the video is licensed under. creator: The creator of the video. + release_timestamp: UNIX timestamp of the moment the video was released. release_date: The date (YYYYMMDD) when the video was released. - timestamp: UNIX timestamp of the moment the video became available. + timestamp: UNIX timestamp of the moment the video became available + (uploaded). upload_date: Video upload date (YYYYMMDD). If not explicitly set, calculated from timestamp. uploader_id: Nickname or id of the video uploader. @@ -1273,6 +1275,7 @@ class InfoExtractor(object): def extract_video_object(e): assert e['@type'] == 'VideoObject' + author = e.get('author') info.update({ 'url': url_or_none(e.get('contentUrl')), 'title': unescapeHTML(e.get('name')), @@ -1280,7 +1283,11 @@ class InfoExtractor(object): 'thumbnail': url_or_none(e.get('thumbnailUrl') or e.get('thumbnailURL')), 'duration': parse_duration(e.get('duration')), 'timestamp': unified_timestamp(e.get('uploadDate')), - 'uploader': str_or_none(e.get('author')), + # author can be an instance of 'Organization' or 'Person' types. + # both types can have 'name' property(inherited from 'Thing' type). [1] + # however some websites are using 'Text' type instead. + # 1. https://schema.org/VideoObject + 'uploader': author.get('name') if isinstance(author, dict) else author if isinstance(author, compat_str) else None, 'filesize': float_or_none(e.get('contentSize')), 'tbr': int_or_none(e.get('bitrate')), 'width': int_or_none(e.get('width')), @@ -2064,7 +2071,7 @@ class InfoExtractor(object): }) return entries - def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, formats_dict={}, data=None, headers={}, query={}): + def _extract_mpd_formats(self, mpd_url, video_id, mpd_id=None, note=None, errnote=None, fatal=True, data=None, headers={}, query={}): res = self._download_xml_handle( mpd_url, video_id, note=note or 'Downloading MPD manifest', @@ -2078,10 +2085,9 @@ class InfoExtractor(object): mpd_base_url = base_url(urlh.geturl()) return self._parse_mpd_formats( - mpd_doc, mpd_id=mpd_id, mpd_base_url=mpd_base_url, - formats_dict=formats_dict, mpd_url=mpd_url) + mpd_doc, mpd_id, mpd_base_url, mpd_url) - def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', formats_dict={}, mpd_url=None): + def _parse_mpd_formats(self, mpd_doc, mpd_id=None, mpd_base_url='', mpd_url=None): """ Parse formats from MPD manifest. References: @@ -2359,15 +2365,7 @@ class InfoExtractor(object): else: # Assuming direct URL to unfragmented media. f['url'] = base_url - - # According to [1, 5.3.5.2, Table 7, page 35] @id of Representation - # is not necessarily unique within a Period thus formats with - # the same `format_id` are quite possible. There are numerous examples - # of such manifests (see https://github.com/ytdl-org/youtube-dl/issues/15111, - # https://github.com/ytdl-org/youtube-dl/issues/13919) - full_info = formats_dict.get(representation_id, {}).copy() - full_info.update(f) - formats.append(full_info) + formats.append(f) else: self.report_warning('Unknown MIME type %s in DASH manifest' % mime_type) return formats @@ -2903,10 +2901,10 @@ class InfoExtractor(object): self._downloader.cookiejar.set_cookie(cookie) def _get_cookies(self, url): - """ Return a compat_cookies.SimpleCookie with the cookies for the url """ + """ Return a compat_cookies_SimpleCookie with the cookies for the url """ req = sanitized_Request(url) self._downloader.cookiejar.add_cookie_header(req) - return compat_cookies.SimpleCookie(req.get_header('Cookie')) + return compat_cookies_SimpleCookie(req.get_header('Cookie')) def _apply_first_set_cookie_header(self, url_handle, cookie): """ diff --git a/youtube_dl/extractor/cspan.py b/youtube_dl/extractor/cspan.py index 766942146..2e01aff48 100644 --- a/youtube_dl/extractor/cspan.py +++ b/youtube_dl/extractor/cspan.py @@ -8,11 +8,14 @@ from ..utils import ( ExtractorError, extract_attributes, find_xpath_attr, + get_element_by_attribute, get_element_by_class, int_or_none, js_to_json, merge_dicts, + parse_iso8601, smuggle_url, + str_to_int, unescapeHTML, ) from .senateisvp import SenateISVPIE @@ -116,8 +119,30 @@ class CSpanIE(InfoExtractor): jwsetup, video_id, require_title=False, m3u8_id='hls', base_url=url) add_referer(info['formats']) + for subtitles in info['subtitles'].values(): + for subtitle in subtitles: + ext = determine_ext(subtitle['url']) + if ext == 'php': + ext = 'vtt' + subtitle['ext'] = ext ld_info = self._search_json_ld(webpage, video_id, default={}) - return merge_dicts(info, ld_info) + title = get_element_by_class('video-page-title', webpage) or \ + self._og_search_title(webpage) + description = get_element_by_attribute('itemprop', 'description', webpage) or \ + self._html_search_meta(['og:description', 'description'], webpage) + return merge_dicts(info, ld_info, { + 'title': title, + 'thumbnail': get_element_by_attribute('itemprop', 'thumbnailUrl', webpage), + 'description': description, + 'timestamp': parse_iso8601(get_element_by_attribute('itemprop', 'uploadDate', webpage)), + 'location': get_element_by_attribute('itemprop', 'contentLocation', webpage), + 'duration': int_or_none(self._search_regex( + r'jwsetup\.seclength\s*=\s*(\d+);', + webpage, 'duration', fatal=False)), + 'view_count': str_to_int(self._search_regex( + r"<span[^>]+class='views'[^>]*>([\d,]+)\s+Views</span>", + webpage, 'views', fatal=False)), + }) # Obsolete # We first look for clipid, because clipprog always appears before diff --git a/youtube_dl/extractor/curiositystream.py b/youtube_dl/extractor/curiositystream.py index e4a7fca6c..48ff30432 100644 --- a/youtube_dl/extractor/curiositystream.py +++ b/youtube_dl/extractor/curiositystream.py @@ -25,12 +25,12 @@ class CuriosityStreamBaseIE(InfoExtractor): raise ExtractorError( '%s said: %s' % (self.IE_NAME, error), expected=True) - def _call_api(self, path, video_id): + def _call_api(self, path, video_id, query=None): headers = {} if self._auth_token: headers['X-Auth-Token'] = self._auth_token result = self._download_json( - self._API_BASE_URL + path, video_id, headers=headers) + self._API_BASE_URL + path, video_id, headers=headers, query=query) self._handle_errors(result) return result['data'] @@ -52,62 +52,75 @@ class CuriosityStreamIE(CuriosityStreamBaseIE): _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/video/(?P<id>\d+)' _TEST = { 'url': 'https://app.curiositystream.com/video/2', - 'md5': '262bb2f257ff301115f1973540de8983', 'info_dict': { 'id': '2', 'ext': 'mp4', 'title': 'How Did You Develop The Internet?', 'description': 'Vint Cerf, Google\'s Chief Internet Evangelist, describes how he and Bob Kahn created the internet.', - } + }, + 'params': { + 'format': 'bestvideo', + # m3u8 download + 'skip_download': True, + }, } def _real_extract(self, url): video_id = self._match_id(url) - media = self._call_api('media/' + video_id, video_id) - title = media['title'] formats = [] - for encoding in media.get('encodings', []): - m3u8_url = encoding.get('master_playlist_url') - if m3u8_url: - formats.extend(self._extract_m3u8_formats( - m3u8_url, video_id, 'mp4', 'm3u8_native', - m3u8_id='hls', fatal=False)) - encoding_url = encoding.get('url') - file_url = encoding.get('file_url') - if not encoding_url and not file_url: - continue - f = { - 'width': int_or_none(encoding.get('width')), - 'height': int_or_none(encoding.get('height')), - 'vbr': int_or_none(encoding.get('video_bitrate')), - 'abr': int_or_none(encoding.get('audio_bitrate')), - 'filesize': int_or_none(encoding.get('size_in_bytes')), - 'vcodec': encoding.get('video_codec'), - 'acodec': encoding.get('audio_codec'), - 'container': encoding.get('container_type'), - } - for f_url in (encoding_url, file_url): - if not f_url: + for encoding_format in ('m3u8', 'mpd'): + media = self._call_api('media/' + video_id, video_id, query={ + 'encodingsNew': 'true', + 'encodingsFormat': encoding_format, + }) + for encoding in media.get('encodings', []): + playlist_url = encoding.get('master_playlist_url') + if encoding_format == 'm3u8': + # use `m3u8` entry_protocol until EXT-X-MAP is properly supported by `m3u8_native` entry_protocol + formats.extend(self._extract_m3u8_formats( + playlist_url, video_id, 'mp4', + m3u8_id='hls', fatal=False)) + elif encoding_format == 'mpd': + formats.extend(self._extract_mpd_formats( + playlist_url, video_id, mpd_id='dash', fatal=False)) + encoding_url = encoding.get('url') + file_url = encoding.get('file_url') + if not encoding_url and not file_url: continue - fmt = f.copy() - rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', f_url) - if rtmp: - fmt.update({ - 'url': rtmp.group('url'), - 'play_path': rtmp.group('playpath'), - 'app': rtmp.group('app'), - 'ext': 'flv', - 'format_id': 'rtmp', - }) - else: - fmt.update({ - 'url': f_url, - 'format_id': 'http', - }) - formats.append(fmt) + f = { + 'width': int_or_none(encoding.get('width')), + 'height': int_or_none(encoding.get('height')), + 'vbr': int_or_none(encoding.get('video_bitrate')), + 'abr': int_or_none(encoding.get('audio_bitrate')), + 'filesize': int_or_none(encoding.get('size_in_bytes')), + 'vcodec': encoding.get('video_codec'), + 'acodec': encoding.get('audio_codec'), + 'container': encoding.get('container_type'), + } + for f_url in (encoding_url, file_url): + if not f_url: + continue + fmt = f.copy() + rtmp = re.search(r'^(?P<url>rtmpe?://(?P<host>[^/]+)/(?P<app>.+))/(?P<playpath>mp[34]:.+)$', f_url) + if rtmp: + fmt.update({ + 'url': rtmp.group('url'), + 'play_path': rtmp.group('playpath'), + 'app': rtmp.group('app'), + 'ext': 'flv', + 'format_id': 'rtmp', + }) + else: + fmt.update({ + 'url': f_url, + 'format_id': 'http', + }) + formats.append(fmt) self._sort_formats(formats) + title = media['title'] + subtitles = {} for closed_caption in media.get('closed_captions', []): sub_url = closed_caption.get('file') @@ -132,7 +145,7 @@ class CuriosityStreamIE(CuriosityStreamBaseIE): class CuriosityStreamCollectionIE(CuriosityStreamBaseIE): IE_NAME = 'curiositystream:collection' - _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/(?:collection|series)/(?P<id>\d+)' + _VALID_URL = r'https?://(?:app\.)?curiositystream\.com/(?:collections?|series)/(?P<id>\d+)' _TESTS = [{ 'url': 'https://app.curiositystream.com/collection/2', 'info_dict': { @@ -140,10 +153,13 @@ class CuriosityStreamCollectionIE(CuriosityStreamBaseIE): 'title': 'Curious Minds: The Internet', 'description': 'How is the internet shaping our lives in the 21st Century?', }, - 'playlist_mincount': 17, + 'playlist_mincount': 16, }, { 'url': 'https://curiositystream.com/series/2', 'only_matching': True, + }, { + 'url': 'https://curiositystream.com/collections/36', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/dispeak.py b/youtube_dl/extractor/dispeak.py index c345e0274..276fd4b09 100644 --- a/youtube_dl/extractor/dispeak.py +++ b/youtube_dl/extractor/dispeak.py @@ -32,6 +32,18 @@ class DigitallySpeakingIE(InfoExtractor): # From http://www.gdcvault.com/play/1013700/Advanced-Material 'url': 'http://sevt.dispeak.com/ubm/gdc/eur10/xml/11256_1282118587281VNIT.xml', 'only_matching': True, + }, { + # From https://gdcvault.com/play/1016624, empty speakerVideo + 'url': 'https://sevt.dispeak.com/ubm/gdc/online12/xml/201210-822101_1349794556671DDDD.xml', + 'info_dict': { + 'id': '201210-822101_1349794556671DDDD', + 'ext': 'flv', + 'title': 'Pre-launch - Preparing to Take the Plunge', + }, + }, { + # From http://www.gdcvault.com/play/1014846/Conference-Keynote-Shigeru, empty slideVideo + 'url': 'http://events.digitallyspeaking.com/gdc/project25/xml/p25-miyamoto1999_1282467389849HSVB.xml', + 'only_matching': True, }] def _parse_mp4(self, metadata): @@ -84,26 +96,20 @@ class DigitallySpeakingIE(InfoExtractor): 'vcodec': 'none', 'format_id': audio.get('code'), }) - slide_video_path = xpath_text(metadata, './slideVideo', fatal=True) - formats.append({ - 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url, - 'play_path': remove_end(slide_video_path, '.flv'), - 'ext': 'flv', - 'format_note': 'slide deck video', - 'quality': -2, - 'preference': -2, - 'format_id': 'slides', - }) - speaker_video_path = xpath_text(metadata, './speakerVideo', fatal=True) - formats.append({ - 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url, - 'play_path': remove_end(speaker_video_path, '.flv'), - 'ext': 'flv', - 'format_note': 'speaker video', - 'quality': -1, - 'preference': -1, - 'format_id': 'speaker', - }) + for video_key, format_id, preference in ( + ('slide', 'slides', -2), ('speaker', 'speaker', -1)): + video_path = xpath_text(metadata, './%sVideo' % video_key) + if not video_path: + continue + formats.append({ + 'url': 'rtmp://%s/ondemand?ovpfv=1.1' % akamai_url, + 'play_path': remove_end(video_path, '.flv'), + 'ext': 'flv', + 'format_note': '%s video' % video_key, + 'quality': preference, + 'preference': preference, + 'format_id': format_id, + }) return formats def _real_extract(self, url): diff --git a/youtube_dl/extractor/dplay.py b/youtube_dl/extractor/dplay.py index 47501dbe6..bbb199094 100644 --- a/youtube_dl/extractor/dplay.py +++ b/youtube_dl/extractor/dplay.py @@ -1,6 +1,7 @@ # coding: utf-8 from __future__ import unicode_literals +import json import re from .common import InfoExtractor @@ -10,11 +11,13 @@ from ..utils import ( ExtractorError, float_or_none, int_or_none, + strip_or_none, unified_timestamp, ) class DPlayIE(InfoExtractor): + _PATH_REGEX = r'/(?P<id>[^/]+/[^/?#]+)' _VALID_URL = r'''(?x)https?:// (?P<domain> (?:www\.)?(?P<host>d @@ -24,7 +27,7 @@ class DPlayIE(InfoExtractor): ) )| (?P<subdomain_country>es|it)\.dplay\.com - )/[^/]+/(?P<id>[^/]+/[^/?#]+)''' + )/[^/]+''' + _PATH_REGEX _TESTS = [{ # non geo restricted, via secure api, unsigned download hls URL @@ -151,56 +154,79 @@ class DPlayIE(InfoExtractor): 'only_matching': True, }] + def _process_errors(self, e, geo_countries): + info = self._parse_json(e.cause.read().decode('utf-8'), None) + error = info['errors'][0] + error_code = error.get('code') + if error_code == 'access.denied.geoblocked': + self.raise_geo_restricted(countries=geo_countries) + elif error_code in ('access.denied.missingpackage', 'invalid.token'): + raise ExtractorError( + 'This video is only available for registered users. You may want to use --cookies.', expected=True) + raise ExtractorError(info['errors'][0]['detail'], expected=True) + + def _update_disco_api_headers(self, headers, disco_base, display_id, realm): + headers['Authorization'] = 'Bearer ' + self._download_json( + disco_base + 'token', display_id, 'Downloading token', + query={ + 'realm': realm, + })['data']['attributes']['token'] + + def _download_video_playback_info(self, disco_base, video_id, headers): + streaming = self._download_json( + disco_base + 'playback/videoPlaybackInfo/' + video_id, + video_id, headers=headers)['data']['attributes']['streaming'] + streaming_list = [] + for format_id, format_dict in streaming.items(): + streaming_list.append({ + 'type': format_id, + 'url': format_dict.get('url'), + }) + return streaming_list + def _get_disco_api_info(self, url, display_id, disco_host, realm, country): geo_countries = [country.upper()] self._initialize_geo_bypass({ 'countries': geo_countries, }) disco_base = 'https://%s/' % disco_host - token = self._download_json( - disco_base + 'token', display_id, 'Downloading token', - query={ - 'realm': realm, - })['data']['attributes']['token'] headers = { 'Referer': url, - 'Authorization': 'Bearer ' + token, } - video = self._download_json( - disco_base + 'content/videos/' + display_id, display_id, - headers=headers, query={ - 'fields[channel]': 'name', - 'fields[image]': 'height,src,width', - 'fields[show]': 'name', - 'fields[tag]': 'name', - 'fields[video]': 'description,episodeNumber,name,publishStart,seasonNumber,videoDuration', - 'include': 'images,primaryChannel,show,tags' - }) + self._update_disco_api_headers(headers, disco_base, display_id, realm) + try: + video = self._download_json( + disco_base + 'content/videos/' + display_id, display_id, + headers=headers, query={ + 'fields[channel]': 'name', + 'fields[image]': 'height,src,width', + 'fields[show]': 'name', + 'fields[tag]': 'name', + 'fields[video]': 'description,episodeNumber,name,publishStart,seasonNumber,videoDuration', + 'include': 'images,primaryChannel,show,tags' + }) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400: + self._process_errors(e, geo_countries) + raise video_id = video['data']['id'] info = video['data']['attributes'] title = info['name'].strip() formats = [] try: - streaming = self._download_json( - disco_base + 'playback/videoPlaybackInfo/' + video_id, - display_id, headers=headers)['data']['attributes']['streaming'] + streaming = self._download_video_playback_info( + disco_base, video_id, headers) except ExtractorError as e: if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: - info = self._parse_json(e.cause.read().decode('utf-8'), display_id) - error = info['errors'][0] - error_code = error.get('code') - if error_code == 'access.denied.geoblocked': - self.raise_geo_restricted(countries=geo_countries) - elif error_code == 'access.denied.missingpackage': - self.raise_login_required() - raise ExtractorError(info['errors'][0]['detail'], expected=True) + self._process_errors(e, geo_countries) raise - for format_id, format_dict in streaming.items(): + for format_dict in streaming: if not isinstance(format_dict, dict): continue format_url = format_dict.get('url') if not format_url: continue + format_id = format_dict.get('type') ext = determine_ext(format_url) if format_id == 'dash' or ext == 'mpd': formats.extend(self._extract_mpd_formats( @@ -248,7 +274,7 @@ class DPlayIE(InfoExtractor): 'id': video_id, 'display_id': display_id, 'title': title, - 'description': info.get('description'), + 'description': strip_or_none(info.get('description')), 'duration': float_or_none(info.get('videoDuration'), 1000), 'timestamp': unified_timestamp(info.get('publishStart')), 'series': series, @@ -268,3 +294,76 @@ class DPlayIE(InfoExtractor): host = 'disco-api.' + domain if domain[0] == 'd' else 'eu2-prod.disco-api.com' return self._get_disco_api_info( url, display_id, host, 'dplay' + country, country) + + +class DiscoveryPlusIE(DPlayIE): + _VALID_URL = r'https?://(?:www\.)?discoveryplus\.com/video' + DPlayIE._PATH_REGEX + _TESTS = [{ + 'url': 'https://www.discoveryplus.com/video/property-brothers-forever-home/food-and-family', + 'info_dict': { + 'id': '1140794', + 'display_id': 'property-brothers-forever-home/food-and-family', + 'ext': 'mp4', + 'title': 'Food and Family', + 'description': 'The brothers help a Richmond family expand their single-level home.', + 'duration': 2583.113, + 'timestamp': 1609304400, + 'upload_date': '20201230', + 'creator': 'HGTV', + 'series': 'Property Brothers: Forever Home', + 'season_number': 1, + 'episode_number': 1, + }, + 'skip': 'Available for Premium users', + }] + + def _update_disco_api_headers(self, headers, disco_base, display_id, realm): + headers['x-disco-client'] = 'WEB:UNKNOWN:dplus_us:15.0.0' + + def _download_video_playback_info(self, disco_base, video_id, headers): + return self._download_json( + disco_base + 'playback/v3/videoPlaybackInfo', + video_id, headers=headers, data=json.dumps({ + 'deviceInfo': { + 'adBlocker': False, + }, + 'videoId': video_id, + 'wisteriaProperties': { + 'platform': 'desktop', + 'product': 'dplus_us', + }, + }).encode('utf-8'))['data']['attributes']['streaming'] + + def _real_extract(self, url): + display_id = self._match_id(url) + return self._get_disco_api_info( + url, display_id, 'us1-prod-direct.discoveryplus.com', 'go', 'us') + + +class HGTVDeIE(DPlayIE): + _VALID_URL = r'https?://de\.hgtv\.com/sendungen' + DPlayIE._PATH_REGEX + _TESTS = [{ + 'url': 'https://de.hgtv.com/sendungen/tiny-house-klein-aber-oho/wer-braucht-schon-eine-toilette/', + 'info_dict': { + 'id': '151205', + 'display_id': 'tiny-house-klein-aber-oho/wer-braucht-schon-eine-toilette', + 'ext': 'mp4', + 'title': 'Wer braucht schon eine Toilette', + 'description': 'md5:05b40a27e7aed2c9172de34d459134e2', + 'duration': 1177.024, + 'timestamp': 1595705400, + 'upload_date': '20200725', + 'creator': 'HGTV', + 'series': 'Tiny House - klein, aber oho', + 'season_number': 3, + 'episode_number': 3, + }, + 'params': { + 'format': 'bestvideo', + }, + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + return self._get_disco_api_info( + url, display_id, 'eu1-prod.disco-api.com', 'hgtv', 'de') diff --git a/youtube_dl/extractor/dreisat.py b/youtube_dl/extractor/dreisat.py index 848d387d1..5a07c18f4 100644 --- a/youtube_dl/extractor/dreisat.py +++ b/youtube_dl/extractor/dreisat.py @@ -1,193 +1,43 @@ from __future__ import unicode_literals -import re +from .zdf import ZDFIE -from .common import InfoExtractor -from ..utils import ( - int_or_none, - unified_strdate, - xpath_text, - determine_ext, - float_or_none, - ExtractorError, -) - -class DreiSatIE(InfoExtractor): +class DreiSatIE(ZDFIE): IE_NAME = '3sat' - _GEO_COUNTRIES = ['DE'] - _VALID_URL = r'https?://(?:www\.)?3sat\.de/mediathek/(?:(?:index|mediathek)\.php)?\?(?:(?:mode|display)=[^&]+&)*obj=(?P<id>[0-9]+)' - _TESTS = [ - { - 'url': 'http://www.3sat.de/mediathek/index.php?mode=play&obj=45918', - 'md5': 'be37228896d30a88f315b638900a026e', - 'info_dict': { - 'id': '45918', - 'ext': 'mp4', - 'title': 'Waidmannsheil', - 'description': 'md5:cce00ca1d70e21425e72c86a98a56817', - 'uploader': 'SCHWEIZWEIT', - 'uploader_id': '100000210', - 'upload_date': '20140913' - }, - 'params': { - 'skip_download': True, # m3u8 downloads - } + _VALID_URL = r'https?://(?:www\.)?3sat\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html' + _TESTS = [{ + # Same as https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html + 'url': 'https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html', + 'md5': '0aff3e7bc72c8813f5e0fae333316a1d', + 'info_dict': { + 'id': '141007_ab18_10wochensommer_film', + 'ext': 'mp4', + 'title': 'Ab 18! - 10 Wochen Sommer', + 'description': 'md5:8253f41dc99ce2c3ff892dac2d65fe26', + 'duration': 2660, + 'timestamp': 1608604200, + 'upload_date': '20201222', }, - { - 'url': 'http://www.3sat.de/mediathek/mediathek.php?mode=play&obj=51066', - 'only_matching': True, + }, { + 'url': 'https://www.3sat.de/gesellschaft/schweizweit/waidmannsheil-100.html', + 'info_dict': { + 'id': '140913_sendung_schweizweit', + 'ext': 'mp4', + 'title': 'Waidmannsheil', + 'description': 'md5:cce00ca1d70e21425e72c86a98a56817', + 'timestamp': 1410623100, + 'upload_date': '20140913' }, - ] - - def _parse_smil_formats(self, smil, smil_url, video_id, namespace=None, f4m_params=None, transform_rtmp_url=None): - param_groups = {} - for param_group in smil.findall(self._xpath_ns('./head/paramGroup', namespace)): - group_id = param_group.get(self._xpath_ns( - 'id', 'http://www.w3.org/XML/1998/namespace')) - params = {} - for param in param_group: - params[param.get('name')] = param.get('value') - param_groups[group_id] = params - - formats = [] - for video in smil.findall(self._xpath_ns('.//video', namespace)): - src = video.get('src') - if not src: - continue - bitrate = int_or_none(self._search_regex(r'_(\d+)k', src, 'bitrate', None)) or float_or_none(video.get('system-bitrate') or video.get('systemBitrate'), 1000) - group_id = video.get('paramGroup') - param_group = param_groups[group_id] - for proto in param_group['protocols'].split(','): - formats.append({ - 'url': '%s://%s' % (proto, param_group['host']), - 'app': param_group['app'], - 'play_path': src, - 'ext': 'flv', - 'format_id': '%s-%d' % (proto, bitrate), - 'tbr': bitrate, - }) - self._sort_formats(formats) - return formats - - def extract_from_xml_url(self, video_id, xml_url): - doc = self._download_xml( - xml_url, video_id, - note='Downloading video info', - errnote='Failed to download video info') - - status_code = xpath_text(doc, './status/statuscode') - if status_code and status_code != 'ok': - if status_code == 'notVisibleAnymore': - message = 'Video %s is not available' % video_id - else: - message = '%s returned error: %s' % (self.IE_NAME, status_code) - raise ExtractorError(message, expected=True) - - title = xpath_text(doc, './/information/title', 'title', True) - - urls = [] - formats = [] - for fnode in doc.findall('.//formitaeten/formitaet'): - video_url = xpath_text(fnode, 'url') - if not video_url or video_url in urls: - continue - urls.append(video_url) - - is_available = 'http://www.metafilegenerator' not in video_url - geoloced = 'static_geoloced_online' in video_url - if not is_available or geoloced: - continue - - format_id = fnode.attrib['basetype'] - format_m = re.match(r'''(?x) - (?P<vcodec>[^_]+)_(?P<acodec>[^_]+)_(?P<container>[^_]+)_ - (?P<proto>[^_]+)_(?P<index>[^_]+)_(?P<indexproto>[^_]+) - ''', format_id) - - ext = determine_ext(video_url, None) or format_m.group('container') - - if ext == 'meta': - continue - elif ext == 'smil': - formats.extend(self._extract_smil_formats( - video_url, video_id, fatal=False)) - elif ext == 'm3u8': - # the certificates are misconfigured (see - # https://github.com/ytdl-org/youtube-dl/issues/8665) - if video_url.startswith('https://'): - continue - formats.extend(self._extract_m3u8_formats( - video_url, video_id, 'mp4', 'm3u8_native', - m3u8_id=format_id, fatal=False)) - elif ext == 'f4m': - formats.extend(self._extract_f4m_formats( - video_url, video_id, f4m_id=format_id, fatal=False)) - else: - quality = xpath_text(fnode, './quality') - if quality: - format_id += '-' + quality - - abr = int_or_none(xpath_text(fnode, './audioBitrate'), 1000) - vbr = int_or_none(xpath_text(fnode, './videoBitrate'), 1000) - - tbr = int_or_none(self._search_regex( - r'_(\d+)k', video_url, 'bitrate', None)) - if tbr and vbr and not abr: - abr = tbr - vbr - - formats.append({ - 'format_id': format_id, - 'url': video_url, - 'ext': ext, - 'acodec': format_m.group('acodec'), - 'vcodec': format_m.group('vcodec'), - 'abr': abr, - 'vbr': vbr, - 'tbr': tbr, - 'width': int_or_none(xpath_text(fnode, './width')), - 'height': int_or_none(xpath_text(fnode, './height')), - 'filesize': int_or_none(xpath_text(fnode, './filesize')), - 'protocol': format_m.group('proto').lower(), - }) - - geolocation = xpath_text(doc, './/details/geolocation') - if not formats and geolocation and geolocation != 'none': - self.raise_geo_restricted(countries=self._GEO_COUNTRIES) - - self._sort_formats(formats) - - thumbnails = [] - for node in doc.findall('.//teaserimages/teaserimage'): - thumbnail_url = node.text - if not thumbnail_url: - continue - thumbnail = { - 'url': thumbnail_url, - } - thumbnail_key = node.get('key') - if thumbnail_key: - m = re.match('^([0-9]+)x([0-9]+)$', thumbnail_key) - if m: - thumbnail['width'] = int(m.group(1)) - thumbnail['height'] = int(m.group(2)) - thumbnails.append(thumbnail) - - upload_date = unified_strdate(xpath_text(doc, './/details/airtime')) - - return { - 'id': video_id, - 'title': title, - 'description': xpath_text(doc, './/information/detail'), - 'duration': int_or_none(xpath_text(doc, './/details/lengthSec')), - 'thumbnails': thumbnails, - 'uploader': xpath_text(doc, './/details/originChannelTitle'), - 'uploader_id': xpath_text(doc, './/details/originChannelId'), - 'upload_date': upload_date, - 'formats': formats, + 'params': { + 'skip_download': True, } - - def _real_extract(self, url): - video_id = self._match_id(url) - details_url = 'http://www.3sat.de/mediathek/xmlservice/web/beitragsDetails?id=%s' % video_id - return self.extract_from_xml_url(video_id, details_url) + }, { + # Same as https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html + 'url': 'https://www.3sat.de/film/spielfilm/der-hauptmann-100.html', + 'only_matching': True, + }, { + # Same as https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids + 'url': 'https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html', + 'only_matching': True, + }] diff --git a/youtube_dl/extractor/egghead.py b/youtube_dl/extractor/egghead.py index df11dc206..9bbd703e0 100644 --- a/youtube_dl/extractor/egghead.py +++ b/youtube_dl/extractor/egghead.py @@ -12,26 +12,35 @@ from ..utils import ( ) -class EggheadCourseIE(InfoExtractor): +class EggheadBaseIE(InfoExtractor): + def _call_api(self, path, video_id, resource, fatal=True): + return self._download_json( + 'https://app.egghead.io/api/v1/' + path, + video_id, 'Downloading %s JSON' % resource, fatal=fatal) + + +class EggheadCourseIE(EggheadBaseIE): IE_DESC = 'egghead.io course' IE_NAME = 'egghead:course' - _VALID_URL = r'https://egghead\.io/courses/(?P<id>[^/?#&]+)' - _TEST = { + _VALID_URL = r'https://(?:app\.)?egghead\.io/(?:course|playlist)s/(?P<id>[^/?#&]+)' + _TESTS = [{ 'url': 'https://egghead.io/courses/professor-frisby-introduces-composable-functional-javascript', 'playlist_count': 29, 'info_dict': { - 'id': '72', + 'id': '432655', 'title': 'Professor Frisby Introduces Composable Functional JavaScript', 'description': 're:(?s)^This course teaches the ubiquitous.*You\'ll start composing functionality before you know it.$', }, - } + }, { + 'url': 'https://app.egghead.io/playlists/professor-frisby-introduces-composable-functional-javascript', + 'only_matching': True, + }] def _real_extract(self, url): playlist_id = self._match_id(url) - - lessons = self._download_json( - 'https://egghead.io/api/v1/series/%s/lessons' % playlist_id, - playlist_id, 'Downloading course lessons JSON') + series_path = 'series/' + playlist_id + lessons = self._call_api( + series_path + '/lessons', playlist_id, 'course lessons') entries = [] for lesson in lessons: @@ -44,9 +53,8 @@ class EggheadCourseIE(InfoExtractor): entries.append(self.url_result( lesson_url, ie=EggheadLessonIE.ie_key(), video_id=lesson_id)) - course = self._download_json( - 'https://egghead.io/api/v1/series/%s' % playlist_id, - playlist_id, 'Downloading course JSON', fatal=False) or {} + course = self._call_api( + series_path, playlist_id, 'course', False) or {} playlist_id = course.get('id') if playlist_id: @@ -57,10 +65,10 @@ class EggheadCourseIE(InfoExtractor): course.get('description')) -class EggheadLessonIE(InfoExtractor): +class EggheadLessonIE(EggheadBaseIE): IE_DESC = 'egghead.io lesson' IE_NAME = 'egghead:lesson' - _VALID_URL = r'https://egghead\.io/(?:api/v1/)?lessons/(?P<id>[^/?#&]+)' + _VALID_URL = r'https://(?:app\.)?egghead\.io/(?:api/v1/)?lessons/(?P<id>[^/?#&]+)' _TESTS = [{ 'url': 'https://egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box', 'info_dict': { @@ -74,7 +82,7 @@ class EggheadLessonIE(InfoExtractor): 'upload_date': '20161209', 'duration': 304, 'view_count': 0, - 'tags': ['javascript', 'free'], + 'tags': 'count:2', }, 'params': { 'skip_download': True, @@ -83,13 +91,16 @@ class EggheadLessonIE(InfoExtractor): }, { 'url': 'https://egghead.io/api/v1/lessons/react-add-redux-to-a-react-application', 'only_matching': True, + }, { + 'url': 'https://app.egghead.io/lessons/javascript-linear-data-flow-with-container-style-types-box', + 'only_matching': True, }] def _real_extract(self, url): display_id = self._match_id(url) - lesson = self._download_json( - 'https://egghead.io/api/v1/lessons/%s' % display_id, display_id) + lesson = self._call_api( + 'lessons/' + display_id, display_id, 'lesson') lesson_id = compat_str(lesson['id']) title = lesson['title'] diff --git a/youtube_dl/extractor/eroprofile.py b/youtube_dl/extractor/eroprofile.py index c08643a17..c460dc7f9 100644 --- a/youtube_dl/extractor/eroprofile.py +++ b/youtube_dl/extractor/eroprofile.py @@ -6,7 +6,7 @@ from .common import InfoExtractor from ..compat import compat_urllib_parse_urlencode from ..utils import ( ExtractorError, - unescapeHTML + merge_dicts, ) @@ -24,7 +24,8 @@ class EroProfileIE(InfoExtractor): 'title': 'sexy babe softcore', 'thumbnail': r're:https?://.*\.jpg', 'age_limit': 18, - } + }, + 'skip': 'Video not found', }, { 'url': 'http://www.eroprofile.com/m/videos/view/Try-It-On-Pee_cut_2-wmv-4shared-com-file-sharing-download-movie-file', 'md5': '1baa9602ede46ce904c431f5418d8916', @@ -77,19 +78,15 @@ class EroProfileIE(InfoExtractor): [r"glbUpdViews\s*\('\d*','(\d+)'", r'p/report/video/(\d+)'], webpage, 'video id', default=None) - video_url = unescapeHTML(self._search_regex( - r'<source src="([^"]+)', webpage, 'video url')) title = self._html_search_regex( - r'Title:</th><td>([^<]+)</td>', webpage, 'title') - thumbnail = self._search_regex( - r'onclick="showVideoPlayer\(\)"><img src="([^"]+)', - webpage, 'thumbnail', fatal=False) + (r'Title:</th><td>([^<]+)</td>', r'<h1[^>]*>(.+?)</h1>'), + webpage, 'title') + + info = self._parse_html5_media_entries(url, webpage, video_id)[0] - return { + return merge_dicts(info, { 'id': video_id, 'display_id': display_id, - 'url': video_url, 'title': title, - 'thumbnail': thumbnail, 'age_limit': 18, - } + }) diff --git a/youtube_dl/extractor/extractors.py b/youtube_dl/extractor/extractors.py index d403a2dbe..4e9954c6a 100644 --- a/youtube_dl/extractor/extractors.py +++ b/youtube_dl/extractor/extractors.py @@ -42,7 +42,10 @@ from .aljazeera import AlJazeeraIE from .alphaporno import AlphaPornoIE from .amara import AmaraIE from .amcnetworks import AMCNetworksIE -from .americastestkitchen import AmericasTestKitchenIE +from .americastestkitchen import ( + AmericasTestKitchenIE, + AmericasTestKitchenSeasonIE, +) from .animeondemand import AnimeOnDemandIE from .anvato import AnvatoIE from .aol import AolIE @@ -69,6 +72,7 @@ from .arte import ( ArteTVEmbedIE, ArteTVPlaylistIE, ) +from .arnes import ArnesIE from .asiancrush import ( AsianCrushIE, AsianCrushPlaylistIE, @@ -87,11 +91,13 @@ from .awaan import ( ) from .azmedien import AZMedienIE from .baidu import BaiduVideoIE +from .bandaichannel import BandaiChannelIE from .bandcamp import BandcampIE, BandcampAlbumIE, BandcampWeeklyIE from .bbc import ( BBCCoUkIE, BBCCoUkArticleIE, - BBCCoUkIPlayerPlaylistIE, + BBCCoUkIPlayerEpisodesIE, + BBCCoUkIPlayerGroupIE, BBCCoUkPlaylistIE, BBCIE, ) @@ -126,7 +132,6 @@ from .bleacherreport import ( BleacherReportIE, BleacherReportCMSIE, ) -from .blinkx import BlinkxIE from .bloomberg import BloombergIE from .bokecc import BokeCCIE from .bongacams import BongaCamsIE @@ -160,6 +165,7 @@ from .canvas import ( CanvasIE, CanvasEenIE, VrtNUIE, + DagelijkseKostIE, ) from .carambatv import ( CarambaTVIE, @@ -184,7 +190,11 @@ from .cbsnews import ( CBSNewsIE, CBSNewsLiveVideoIE, ) -from .cbssports import CBSSportsIE +from .cbssports import ( + CBSSportsEmbedIE, + CBSSportsIE, + TwentyFourSevenSportsIE, +) from .ccc import ( CCCIE, CCCPlaylistIE, @@ -232,11 +242,8 @@ from .cnn import ( ) from .coub import CoubIE from .comedycentral import ( - ComedyCentralFullEpisodesIE, ComedyCentralIE, - ComedyCentralShortnameIE, ComedyCentralTVIE, - ToshIE, ) from .commonmistakes import CommonMistakesIE, UnicodeBOMIE from .commonprotocols import ( @@ -287,7 +294,11 @@ from .douyutv import ( DouyuShowIE, DouyuTVIE, ) -from .dplay import DPlayIE +from .dplay import ( + DPlayIE, + DiscoveryPlusIE, + HGTVDeIE, +) from .dreisat import DreiSatIE from .drbonanza import DRBonanzaIE from .drtuber import DrTuberIE @@ -416,6 +427,7 @@ from .gamestar import GameStarIE from .gaskrank import GaskrankIE from .gazeta import GazetaIE from .gdcvault import GDCVaultIE +from .gedidigital import GediDigitalIE from .generic import GenericIE from .gfycat import GfycatIE from .giantbomb import GiantBombIE @@ -470,8 +482,8 @@ from .hungama import ( from .hypem import HypemIE from .ign import ( IGNIE, - OneUPIE, - PCMagIE, + IGNVideoIE, + IGNArticleIE, ) from .iheart import ( IHeartRadioIE, @@ -526,7 +538,10 @@ from .karaoketv import KaraoketvIE from .karrierevideos import KarriereVideosIE from .keezmovies import KeezMoviesIE from .ketnet import KetnetIE -from .khanacademy import KhanAcademyIE +from .khanacademy import ( + KhanAcademyIE, + KhanAcademyUnitIE, +) from .kickstarter import KickStarterIE from .kinja import KinjaEmbedIE from .kinopoisk import KinoPoiskIE @@ -583,7 +598,11 @@ from .limelight import ( LimelightChannelIE, LimelightChannelListIE, ) -from .line import LineTVIE +from .line import ( + LineTVIE, + LineLiveIE, + LineLiveChannelIE, +) from .linkedin import ( LinkedInLearningIE, LinkedInLearningCourseIE, @@ -591,10 +610,6 @@ from .linkedin import ( from .linuxacademy import LinuxAcademyIE from .litv import LiTVIE from .livejournal import LiveJournalIE -from .liveleak import ( - LiveLeakIE, - LiveLeakEmbedIE, -) from .livestream import ( LivestreamIE, LivestreamOriginalIE, @@ -620,6 +635,7 @@ from .mangomolo import ( MangomoloLiveIE, ) from .manyvids import ManyVidsIE +from .maoritv import MaoriTVIE from .markiza import ( MarkizaIE, MarkizaPageIE, @@ -648,6 +664,11 @@ from .microsoftvirtualacademy import ( MicrosoftVirtualAcademyIE, MicrosoftVirtualAcademyCourseIE, ) +from .minds import ( + MindsIE, + MindsChannelIE, + MindsGroupIE, +) from .ministrygrid import MinistryGridIE from .minoto import MinotoIE from .miomio import MioMioIE @@ -658,7 +679,10 @@ from .mixcloud import ( MixcloudUserIE, MixcloudPlaylistIE, ) -from .mlb import MLBIE +from .mlb import ( + MLBIE, + MLBVideoIE, +) from .mnet import MnetIE from .moevideo import MoeVideoIE from .mofosex import ( @@ -859,6 +883,11 @@ from .packtpub import ( PacktPubIE, PacktPubCourseIE, ) +from .palcomp3 import ( + PalcoMP3IE, + PalcoMP3ArtistIE, + PalcoMP3VideoIE, +) from .pandoratv import PandoraTVIE from .parliamentliveuk import ParliamentLiveUKIE from .patreon import PatreonIE @@ -892,6 +921,7 @@ from .platzi import ( from .playfm import PlayFMIE from .playplustv import PlayPlusTVIE from .plays import PlaysTVIE +from .playstuff import PlayStuffIE from .playtvak import PlaytvakIE from .playvid import PlayvidIE from .playwire import PlaywireIE @@ -1016,6 +1046,7 @@ from .safari import ( SafariApiIE, SafariCourseIE, ) +from .samplefocus import SampleFocusIE from .sapo import SapoIE from .savefrom import SaveFromIE from .sbs import SBSIE @@ -1048,6 +1079,11 @@ from .shared import ( VivoIE, ) from .showroomlive import ShowRoomLiveIE +from .simplecast import ( + SimplecastIE, + SimplecastEpisodeIE, + SimplecastPodcastIE, +) from .sina import SinaIE from .sixplay import SixPlayIE from .skyit import ( @@ -1113,6 +1149,10 @@ from .stitcher import ( from .sport5 import Sport5IE from .sportbox import SportBoxIE from .sportdeutschland import SportDeutschlandIE +from .spotify import ( + SpotifyIE, + SpotifyShowIE, +) from .spreaker import ( SpreakerIE, SpreakerPageIE, @@ -1128,6 +1168,11 @@ from .srgssr import ( from .srmediathek import SRMediathekIE from .stanfordoc import StanfordOpenClassroomIE from .steam import SteamIE +from .storyfire import ( + StoryFireIE, + StoryFireUserIE, + StoryFireSeriesIE, +) from .streamable import StreamableIE from .streamcloud import StreamcloudIE from .streamcz import StreamCZIE @@ -1226,6 +1271,10 @@ from .toutv import TouTvIE from .toypics import ToypicsUserIE, ToypicsIE from .traileraddict import TrailerAddictIE from .trilulilu import TriluliluIE +from .trovo import ( + TrovoIE, + TrovoVodIE, +) from .trunews import TruNewsIE from .trutv import TruTVIE from .tube8 import Tube8IE @@ -1244,6 +1293,7 @@ from .tv2 import ( TV2IE, TV2ArticleIE, KatsomoIE, + MTVUutisetArticleIE, ) from .tv2dk import ( TV2DKIE, @@ -1382,7 +1432,6 @@ from .vidme import ( VidmeUserIE, VidmeUserLikesIE, ) -from .vidzi import VidziIE from .vier import VierIE, VierVideosIE from .viewlift import ( ViewLiftIE, @@ -1442,6 +1491,7 @@ from .vrv import ( VRVSeriesIE, ) from .vshare import VShareIE +from .vtm import VTMIE from .medialaan import MedialaanIE from .vube import VubeIE from .vuclip import VuClipIE @@ -1585,5 +1635,10 @@ from .zattoo import ( ZattooLiveIE, ) from .zdf import ZDFIE, ZDFChannelIE -from .zingmp3 import ZingMp3IE +from .zhihu import ZhihuIE +from .zingmp3 import ( + ZingMp3IE, + ZingMp3AlbumIE, +) +from .zoom import ZoomIE from .zype import ZypeIE diff --git a/youtube_dl/extractor/facebook.py b/youtube_dl/extractor/facebook.py index cb34c59f5..04650af39 100644 --- a/youtube_dl/extractor/facebook.py +++ b/youtube_dl/extractor/facebook.py @@ -521,7 +521,10 @@ class FacebookIE(InfoExtractor): raise ExtractorError( 'The video is not available, Facebook said: "%s"' % m_msg.group(1), expected=True) - elif '>You must log in to continue' in webpage: + elif any(p in webpage for p in ( + '>You must log in to continue', + 'id="login_form"', + 'id="loginbutton"')): self.raise_login_required() if not video_data and '/watchparty/' in url: diff --git a/youtube_dl/extractor/formula1.py b/youtube_dl/extractor/formula1.py index fecfc28ae..67662e6de 100644 --- a/youtube_dl/extractor/formula1.py +++ b/youtube_dl/extractor/formula1.py @@ -5,29 +5,23 @@ from .common import InfoExtractor class Formula1IE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?formula1\.com/(?:content/fom-website/)?en/video/\d{4}/\d{1,2}/(?P<id>.+?)\.html' - _TESTS = [{ - 'url': 'http://www.formula1.com/content/fom-website/en/video/2016/5/Race_highlights_-_Spain_2016.html', - 'md5': '8c79e54be72078b26b89e0e111c0502b', + _VALID_URL = r'https?://(?:www\.)?formula1\.com/en/latest/video\.[^.]+\.(?P<id>\d+)\.html' + _TEST = { + 'url': 'https://www.formula1.com/en/latest/video.race-highlights-spain-2016.6060988138001.html', + 'md5': 'be7d3a8c2f804eb2ab2aa5d941c359f8', 'info_dict': { - 'id': 'JvYXJpMzE6pArfHWm5ARp5AiUmD-gibV', + 'id': '6060988138001', 'ext': 'mp4', 'title': 'Race highlights - Spain 2016', + 'timestamp': 1463332814, + 'upload_date': '20160515', + 'uploader_id': '6057949432001', }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - 'add_ie': ['Ooyala'], - }, { - 'url': 'http://www.formula1.com/en/video/2016/5/Race_highlights_-_Spain_2016.html', - 'only_matching': True, - }] + 'add_ie': ['BrightcoveNew'], + } + BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/6057949432001/S1WMrhjlh_default/index.html?videoId=%s' def _real_extract(self, url): - display_id = self._match_id(url) - webpage = self._download_webpage(url, display_id) - ooyala_embed_code = self._search_regex( - r'data-videoid="([^"]+)"', webpage, 'ooyala embed code') + bc_id = self._match_id(url) return self.url_result( - 'ooyala:%s' % ooyala_embed_code, 'Ooyala', ooyala_embed_code) + self.BRIGHTCOVE_URL_TEMPLATE % bc_id, 'BrightcoveNew', bc_id) diff --git a/youtube_dl/extractor/franceculture.py b/youtube_dl/extractor/franceculture.py index 306b45fc9..14f4cb489 100644 --- a/youtube_dl/extractor/franceculture.py +++ b/youtube_dl/extractor/franceculture.py @@ -11,7 +11,7 @@ from ..utils import ( class FranceCultureIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?franceculture\.fr/emissions/(?:[^/]+/)*(?P<id>[^/?#&]+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.franceculture.fr/emissions/carnet-nomade/rendez-vous-au-pays-des-geeks', 'info_dict': { 'id': 'rendez-vous-au-pays-des-geeks', @@ -20,10 +20,14 @@ class FranceCultureIE(InfoExtractor): 'title': 'Rendez-vous au pays des geeks', 'thumbnail': r're:^https?://.*\.jpg$', 'upload_date': '20140301', - 'timestamp': 1393642916, + 'timestamp': 1393700400, 'vcodec': 'none', } - } + }, { + # no thumbnail + 'url': 'https://www.franceculture.fr/emissions/la-recherche-montre-en-main/la-recherche-montre-en-main-du-mercredi-10-octobre-2018', + 'only_matching': True, + }] def _real_extract(self, url): display_id = self._match_id(url) @@ -36,19 +40,19 @@ class FranceCultureIE(InfoExtractor): </h1>| <div[^>]+class="[^"]*?(?:title-zone-diffusion|heading-zone-(?:wrapper|player-button))[^"]*?"[^>]*> ).*? - (<button[^>]+data-asset-source="[^"]+"[^>]+>) + (<button[^>]+data-(?:url|asset-source)="[^"]+"[^>]+>) ''', webpage, 'video data')) - video_url = video_data['data-asset-source'] - title = video_data.get('data-asset-title') or self._og_search_title(webpage) + video_url = video_data.get('data-url') or video_data['data-asset-source'] + title = video_data.get('data-asset-title') or video_data.get('data-diffusion-title') or self._og_search_title(webpage) description = self._html_search_regex( r'(?s)<div[^>]+class="intro"[^>]*>.*?<h2>(.+?)</h2>', webpage, 'description', default=None) thumbnail = self._search_regex( r'(?s)<figure[^>]+itemtype="https://schema.org/ImageObject"[^>]*>.*?<img[^>]+(?:data-dejavu-)?src="([^"]+)"', - webpage, 'thumbnail', fatal=False) + webpage, 'thumbnail', default=None) uploader = self._html_search_regex( r'(?s)<span class="author">(.*?)</span>', webpage, 'uploader', default=None) @@ -64,6 +68,6 @@ class FranceCultureIE(InfoExtractor): 'ext': ext, 'vcodec': 'none' if ext == 'mp3' else None, 'uploader': uploader, - 'timestamp': int_or_none(video_data.get('data-asset-created-date')), + 'timestamp': int_or_none(video_data.get('data-start-time')) or int_or_none(video_data.get('data-asset-created-date')), 'duration': int_or_none(video_data.get('data-duration')), } diff --git a/youtube_dl/extractor/francetv.py b/youtube_dl/extractor/francetv.py index 3ca415077..e4ec2e200 100644 --- a/youtube_dl/extractor/francetv.py +++ b/youtube_dl/extractor/francetv.py @@ -383,6 +383,10 @@ class FranceTVInfoIE(FranceTVBaseInfoExtractor): }, { 'url': 'http://france3-regions.francetvinfo.fr/limousin/emissions/jt-1213-limousin', 'only_matching': True, + }, { + # "<figure id=" pattern (#28792) + 'url': 'https://www.francetvinfo.fr/culture/patrimoine/incendie-de-notre-dame-de-paris/notre-dame-de-paris-de-l-incendie-de-la-cathedrale-a-sa-reconstruction_4372291.html', + 'only_matching': True, }] def _real_extract(self, url): @@ -399,7 +403,8 @@ class FranceTVInfoIE(FranceTVBaseInfoExtractor): video_id = self._search_regex( (r'player\.load[^;]+src:\s*["\']([^"\']+)', r'id-video=([^@]+@[^"]+)', - r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"'), + r'<a[^>]+href="(?:https?:)?//videos\.francetv\.fr/video/([^@]+@[^"]+)"', + r'(?:data-id|<figure[^<]+\bid)=["\']([\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})'), webpage, 'video id') return self._make_url_result(video_id) diff --git a/youtube_dl/extractor/fujitv.py b/youtube_dl/extractor/fujitv.py index 39685e075..a02a94374 100644 --- a/youtube_dl/extractor/fujitv.py +++ b/youtube_dl/extractor/fujitv.py @@ -17,7 +17,7 @@ class FujiTVFODPlus7IE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) formats = self._extract_m3u8_formats( - self._BASE_URL + 'abr/pc_html5/%s.m3u8' % video_id, video_id) + self._BASE_URL + 'abr/pc_html5/%s.m3u8' % video_id, video_id, 'mp4') for f in formats: wh = self._BITRATE_MAP.get(f.get('tbr')) if wh: diff --git a/youtube_dl/extractor/funimation.py b/youtube_dl/extractor/funimation.py index 8bbedca26..d8f1e169a 100644 --- a/youtube_dl/extractor/funimation.py +++ b/youtube_dl/extractor/funimation.py @@ -16,7 +16,7 @@ from ..utils import ( class FunimationIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?funimation(?:\.com|now\.uk)/shows/[^/]+/(?P<id>[^/?#&]+)' + _VALID_URL = r'https?://(?:www\.)?funimation(?:\.com|now\.uk)/(?:[^/]+/)?shows/[^/]+/(?P<id>[^/?#&]+)' _NETRC_MACHINE = 'funimation' _TOKEN = None @@ -51,6 +51,10 @@ class FunimationIE(InfoExtractor): }, { 'url': 'https://www.funimationnow.uk/shows/puzzle-dragons-x/drop-impact/simulcast/', 'only_matching': True, + }, { + # with lang code + 'url': 'https://www.funimation.com/en/shows/hacksign/role-play/', + 'only_matching': True, }] def _login(self): diff --git a/youtube_dl/extractor/gdcvault.py b/youtube_dl/extractor/gdcvault.py index 2f555c1d4..acc6478b8 100644 --- a/youtube_dl/extractor/gdcvault.py +++ b/youtube_dl/extractor/gdcvault.py @@ -6,6 +6,7 @@ from .common import InfoExtractor from .kaltura import KalturaIE from ..utils import ( HEADRequest, + remove_start, sanitized_Request, smuggle_url, urlencode_postdata, @@ -102,6 +103,26 @@ class GDCVaultIE(InfoExtractor): 'format': 'mp4-408', }, }, + { + # Kaltura embed, whitespace between quote and embedded URL in iframe's src + 'url': 'https://www.gdcvault.com/play/1025699', + 'info_dict': { + 'id': '0_zagynv0a', + 'ext': 'mp4', + 'title': 'Tech Toolbox', + 'upload_date': '20190408', + 'uploader_id': 'joe@blazestreaming.com', + 'timestamp': 1554764629, + }, + 'params': { + 'skip_download': True, + }, + }, + { + # HTML5 video + 'url': 'http://www.gdcvault.com/play/1014846/Conference-Keynote-Shigeru', + 'only_matching': True, + }, ] def _login(self, webpage_url, display_id): @@ -175,7 +196,18 @@ class GDCVaultIE(InfoExtractor): xml_name = self._html_search_regex( r'<iframe src=".*?\?xml(?:=|URL=xml/)(.+?\.xml).*?".*?</iframe>', - start_page, 'xml filename') + start_page, 'xml filename', default=None) + if not xml_name: + info = self._parse_html5_media_entries(url, start_page, video_id)[0] + info.update({ + 'title': remove_start(self._search_regex( + r'>Session Name:\s*<.*?>\s*<td>(.+?)</td>', start_page, + 'title', default=None) or self._og_search_title( + start_page, default=None), 'GDC Vault - '), + 'id': video_id, + 'display_id': display_id, + }) + return info embed_url = '%s/xml/%s' % (xml_root, xml_name) ie_key = 'DigitallySpeaking' diff --git a/youtube_dl/extractor/gedidigital.py b/youtube_dl/extractor/gedidigital.py new file mode 100644 index 000000000..6c4153b40 --- /dev/null +++ b/youtube_dl/extractor/gedidigital.py @@ -0,0 +1,161 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + determine_ext, + int_or_none, +) + + +class GediDigitalIE(InfoExtractor): + _VALID_URL = r'''(?x)https?://video\. + (?: + (?: + (?:espresso\.)?repubblica + |lastampa + |ilsecoloxix + )| + (?: + iltirreno + |messaggeroveneto + |ilpiccolo + |gazzettadimantova + |mattinopadova + |laprovinciapavese + |tribunatreviso + |nuovavenezia + |gazzettadimodena + |lanuovaferrara + |corrierealpi + |lasentinella + )\.gelocal + )\.it(?:/[^/]+){2,3}?/(?P<id>\d+)(?:[/?&#]|$)''' + _TESTS = [{ + 'url': 'https://video.lastampa.it/politica/il-paradosso-delle-regionali-la-lega-vince-ma-sembra-aver-perso/121559/121683', + 'md5': '84658d7fb9e55a6e57ecc77b73137494', + 'info_dict': { + 'id': '121559', + 'ext': 'mp4', + 'title': 'Il paradosso delle Regionali: ecco perché la Lega vince ma sembra aver perso', + 'description': 'md5:de7f4d6eaaaf36c153b599b10f8ce7ca', + 'thumbnail': r're:^https://www\.repstatic\.it/video/photo/.+?-thumb-full-.+?\.jpg$', + 'duration': 125, + }, + }, { + 'url': 'https://video.espresso.repubblica.it/embed/tutti-i-video/01-ted-villa/14772/14870&width=640&height=360', + 'only_matching': True, + }, { + 'url': 'https://video.repubblica.it/motori/record-della-pista-a-spa-francorchamps-la-pagani-huayra-roadster-bc-stupisce/367415/367963', + 'only_matching': True, + }, { + 'url': 'https://video.ilsecoloxix.it/sport/cassani-e-i-brividi-azzurri-ai-mondiali-di-imola-qui-mi-sono-innamorato-del-ciclismo-da-ragazzino-incredibile-tornarci-da-ct/66184/66267', + 'only_matching': True, + }, { + 'url': 'https://video.iltirreno.gelocal.it/sport/dentro-la-notizia-ferrari-cosa-succede-a-maranello/141059/142723', + 'only_matching': True, + }, { + 'url': 'https://video.messaggeroveneto.gelocal.it/locale/maria-giovanna-elmi-covid-vaccino/138155/139268', + 'only_matching': True, + }, { + 'url': 'https://video.ilpiccolo.gelocal.it/dossier/big-john/dinosauro-big-john-al-via-le-visite-guidate-a-trieste/135226/135751', + 'only_matching': True, + }, { + 'url': 'https://video.gazzettadimantova.gelocal.it/locale/dal-ponte-visconteo-di-valeggio-l-and-8217sos-dei-ristoratori-aprire-anche-a-cena/137310/137818', + 'only_matching': True, + }, { + 'url': 'https://video.mattinopadova.gelocal.it/dossier/coronavirus-in-veneto/covid-a-vo-un-anno-dopo-un-cuore-tricolore-per-non-dimenticare/138402/138964', + 'only_matching': True, + }, { + 'url': 'https://video.laprovinciapavese.gelocal.it/locale/mede-zona-rossa-via-alle-vaccinazioni-per-gli-over-80/137545/138120', + 'only_matching': True, + }, { + 'url': 'https://video.tribunatreviso.gelocal.it/dossier/coronavirus-in-veneto/ecco-le-prima-vaccinazioni-di-massa-nella-marca/134485/135024', + 'only_matching': True, + }, { + 'url': 'https://video.nuovavenezia.gelocal.it/locale/camion-troppo-alto-per-il-ponte-ferroviario-perde-il-carico/135734/136266', + 'only_matching': True, + }, { + 'url': 'https://video.gazzettadimodena.gelocal.it/locale/modena-scoperta-la-proteina-che-predice-il-livello-di-gravita-del-covid/139109/139796', + 'only_matching': True, + }, { + 'url': 'https://video.lanuovaferrara.gelocal.it/locale/due-bombole-di-gpl-aperte-e-abbandonate-i-vigili-bruciano-il-gas/134391/134957', + 'only_matching': True, + }, { + 'url': 'https://video.corrierealpi.gelocal.it/dossier/cortina-2021-i-mondiali-di-sci-alpino/mondiali-di-sci-il-timelapse-sulla-splendida-olympia/133760/134331', + 'only_matching': True, + }, { + 'url': 'https://video.lasentinella.gelocal.it/locale/vestigne-centra-un-auto-e-si-ribalta/138931/139466', + 'only_matching': True, + }, { + 'url': 'https://video.espresso.repubblica.it/tutti-i-video/01-ted-villa/14772', + 'only_matching': True, + }] + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + title = self._html_search_meta( + ['twitter:title', 'og:title'], webpage, fatal=True) + player_data = re.findall( + r"PlayerFactory\.setParam\('(?P<type>format|param)',\s*'(?P<name>[^']+)',\s*'(?P<val>[^']+)'\);", + webpage) + + formats = [] + duration = thumb = None + for t, n, v in player_data: + if t == 'format': + if n in ('video-hds-vod-ec', 'video-hls-vod-ec', 'video-viralize', 'video-youtube-pfp'): + continue + elif n.endswith('-vod-ak'): + formats.extend(self._extract_akamai_formats( + v, video_id, {'http': 'media.gedidigital.it'})) + else: + ext = determine_ext(v) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + v, video_id, 'mp4', 'm3u8_native', m3u8_id=n, fatal=False)) + continue + f = { + 'format_id': n, + 'url': v, + } + if ext == 'mp3': + abr = int_or_none(self._search_regex( + r'-mp3-audio-(\d+)', v, 'abr', default=None)) + f.update({ + 'abr': abr, + 'tbr': abr, + 'vcodec': 'none' + }) + else: + mobj = re.match(r'^video-rrtv-(\d+)(?:-(\d+))?$', n) + if mobj: + f.update({ + 'height': int(mobj.group(1)), + 'vbr': int_or_none(mobj.group(2)), + }) + if not f.get('vbr'): + f['vbr'] = int_or_none(self._search_regex( + r'-video-rrtv-(\d+)', v, 'abr', default=None)) + formats.append(f) + elif t == 'param': + if n in ['image_full', 'image']: + thumb = v + elif n == 'videoDuration': + duration = int_or_none(v) + + self._sort_formats(formats) + + return { + 'id': video_id, + 'title': title, + 'description': self._html_search_meta( + ['twitter:description', 'og:description', 'description'], webpage), + 'thumbnail': thumb or self._og_search_thumbnail(webpage), + 'formats': formats, + 'duration': duration, + } diff --git a/youtube_dl/extractor/generic.py b/youtube_dl/extractor/generic.py index 780971a92..a9c064105 100644 --- a/youtube_dl/extractor/generic.py +++ b/youtube_dl/extractor/generic.py @@ -84,7 +84,6 @@ from .jwplatform import JWPlatformIE from .digiteka import DigitekaIE from .arkena import ArkenaIE from .instagram import InstagramIE -from .liveleak import LiveLeakIE from .threeqsdn import ThreeQSDNIE from .theplatform import ThePlatformIE from .kaltura import KalturaIE @@ -126,8 +125,11 @@ from .viqeo import ViqeoIE from .expressen import ExpressenIE from .zype import ZypeIE from .odnoklassniki import OdnoklassnikiIE +from .vk import VKIE from .kinja import KinjaEmbedIE from .arcpublishing import ArcPublishingIE +from .medialaan import MedialaanIE +from .simplecast import SimplecastIE class GenericIE(InfoExtractor): @@ -1626,31 +1628,6 @@ class GenericIE(InfoExtractor): 'upload_date': '20160409', }, }, - # LiveLeak embed - { - 'url': 'http://www.wykop.pl/link/3088787/', - 'md5': '7619da8c820e835bef21a1efa2a0fc71', - 'info_dict': { - 'id': '874_1459135191', - 'ext': 'mp4', - 'title': 'Man shows poor quality of new apartment building', - 'description': 'The wall is like a sand pile.', - 'uploader': 'Lake8737', - }, - 'add_ie': [LiveLeakIE.ie_key()], - }, - # Another LiveLeak embed pattern (#13336) - { - 'url': 'https://milo.yiannopoulos.net/2017/06/concealed-carry-robbery/', - 'info_dict': { - 'id': '2eb_1496309988', - 'ext': 'mp4', - 'title': 'Thief robs place where everyone was armed', - 'description': 'md5:694d73ee79e535953cf2488562288eee', - 'uploader': 'brazilwtf', - }, - 'add_ie': [LiveLeakIE.ie_key()], - }, # Duplicated embedded video URLs { 'url': 'http://www.hudl.com/athlete/2538180/highlights/149298443', @@ -2223,6 +2200,34 @@ class GenericIE(InfoExtractor): 'duration': 1581, }, }, + { + # MyChannels SDK embed + # https://www.24kitchen.nl/populair/deskundige-dit-waarom-sommigen-gevoelig-zijn-voor-voedselallergieen + 'url': 'https://www.demorgen.be/nieuws/burgemeester-rotterdam-richt-zich-in-videoboodschap-tot-relschoppers-voelt-het-goed~b0bcfd741/', + 'md5': '90c0699c37006ef18e198c032d81739c', + 'info_dict': { + 'id': '194165', + 'ext': 'mp4', + 'title': 'Burgemeester Aboutaleb spreekt relschoppers toe', + 'timestamp': 1611740340, + 'upload_date': '20210127', + 'duration': 159, + }, + }, + { + # Simplecast player embed + 'url': 'https://www.bio.org/podcast', + 'info_dict': { + 'id': 'podcast', + 'title': 'I AM BIO Podcast | BIO', + }, + 'playlist_mincount': 52, + }, + { + # Sibnet embed (https://help.sibnet.ru/?sibnet_video_embed) + 'url': 'https://phpbb3.x-tk.ru/bbcode-video-sibnet-t24.html', + 'only_matching': True, + }, ] def report_following_redirect(self, new_url): @@ -2462,6 +2467,9 @@ class GenericIE(InfoExtractor): webpage = self._webpage_read_content( full_response, url, video_id, prefix=first_bytes) + if '<title>DPG Media Privacy Gate</title>' in webpage: + webpage = self._download_webpage(url, video_id) + self.report_extraction(video_id) # Is it an RSS feed, a SMIL file, an XSPF playlist or a MPD manifest? @@ -2593,6 +2601,11 @@ class GenericIE(InfoExtractor): if arc_urls: return self.playlist_from_matches(arc_urls, video_id, video_title, ie=ArcPublishingIE.ie_key()) + mychannels_urls = MedialaanIE._extract_urls(webpage) + if mychannels_urls: + return self.playlist_from_matches( + mychannels_urls, video_id, video_title, ie=MedialaanIE.ie_key()) + # Look for embedded rtl.nl player matches = re.findall( r'<iframe[^>]+?src="((?:https?:)?//(?:(?:www|static)\.)?rtl\.nl/(?:system/videoplayer/[^"]+(?:video_)?)?embed[^"]+)"', @@ -2744,6 +2757,11 @@ class GenericIE(InfoExtractor): if odnoklassniki_url: return self.url_result(odnoklassniki_url, OdnoklassnikiIE.ie_key()) + # Look for sibnet embedded player + sibnet_urls = VKIE._extract_sibnet_urls(webpage) + if sibnet_urls: + return self.playlist_from_matches(sibnet_urls, video_id, video_title) + # Look for embedded ivi player mobj = re.search(r'<embed[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?ivi\.ru/video/player.+?)\1', webpage) if mobj is not None: @@ -2769,6 +2787,12 @@ class GenericIE(InfoExtractor): return self.playlist_from_matches( matches, video_id, video_title, getter=unescapeHTML, ie='FunnyOrDie') + # Look for Simplecast embeds + simplecast_urls = SimplecastIE._extract_urls(webpage) + if simplecast_urls: + return self.playlist_from_matches( + simplecast_urls, video_id, video_title) + # Look for BBC iPlayer embed matches = re.findall(r'setPlaylist\("(https?://www\.bbc\.co\.uk/iplayer/[^/]+/[\da-z]{8})"\)', webpage) if matches: @@ -2914,7 +2938,7 @@ class GenericIE(InfoExtractor): webpage) if not mobj: mobj = re.search( - r'data-video-link=["\'](?P<url>http://m.mlb.com/video/[^"\']+)', + r'data-video-link=["\'](?P<url>http://m\.mlb\.com/video/[^"\']+)', webpage) if mobj is not None: return self.url_result(mobj.group('url'), 'MLB') @@ -3129,11 +3153,6 @@ class GenericIE(InfoExtractor): return self.url_result( self._proto_relative_url(instagram_embed_url), InstagramIE.ie_key()) - # Look for LiveLeak embeds - liveleak_urls = LiveLeakIE._extract_urls(webpage) - if liveleak_urls: - return self.playlist_from_matches(liveleak_urls, video_id, video_title) - # Look for 3Q SDN embeds threeqsdn_url = ThreeQSDNIE._extract_url(webpage) if threeqsdn_url: @@ -3361,6 +3380,9 @@ class GenericIE(InfoExtractor): 'url': src, 'ext': (mimetype2ext(src_type) or ext if ext in KNOWN_EXTENSIONS else 'mp4'), + 'http_headers': { + 'Referer': full_response.geturl(), + }, }) if formats: self._sort_formats(formats) @@ -3429,7 +3451,7 @@ class GenericIE(InfoExtractor): m_video_type = re.findall(r'<meta.*?property="og:video:type".*?content="video/(.*?)"', webpage) # We only look in og:video if the MIME type is a video, don't try if it's a Flash player: if m_video_type is not None: - found = filter_video(re.findall(r'<meta.*?property="og:video".*?content="(.*?)"', webpage)) + found = filter_video(re.findall(r'<meta.*?property="og:(?:video|audio)".*?content="(.*?)"', webpage)) if not found: REDIRECT_REGEX = r'[0-9]{,2};\s*(?:URL|url)=\'?([^\'"]+)' found = re.search( diff --git a/youtube_dl/extractor/go.py b/youtube_dl/extractor/go.py index 0d731e90a..878ba14e6 100644 --- a/youtube_dl/extractor/go.py +++ b/youtube_dl/extractor/go.py @@ -4,10 +4,12 @@ from __future__ import unicode_literals import re from .adobepass import AdobePassIE +from ..compat import compat_str from ..utils import ( int_or_none, determine_ext, parse_age_limit, + try_get, urlencode_postdata, ExtractorError, ) @@ -117,6 +119,18 @@ class GoIE(AdobePassIE): 'skip_download': True, }, }, { + 'url': 'https://abc.com/shows/modern-family/episode-guide/season-01/101-pilot', + 'info_dict': { + 'id': 'VDKA22600213', + 'ext': 'mp4', + 'title': 'Pilot', + 'description': 'md5:74306df917cfc199d76d061d66bebdb4', + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { 'url': 'http://abc.go.com/shows/the-catch/episode-guide/season-01/10-the-wedding', 'only_matching': True, }, { @@ -149,14 +163,30 @@ class GoIE(AdobePassIE): brand = site_info.get('brand') if not video_id or not site_info: webpage = self._download_webpage(url, display_id or video_id) - video_id = self._search_regex( - ( - # There may be inner quotes, e.g. data-video-id="'VDKA3609139'" - # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood - r'data-video-id=["\']*(VDKA\w+)', - # https://abc.com/shows/the-rookie/episode-guide/season-02/03-the-bet - r'\b(?:video)?id["\']\s*:\s*["\'](VDKA\w+)' - ), webpage, 'video id', default=video_id) + data = self._parse_json( + self._search_regex( + r'["\']__abc_com__["\']\s*\]\s*=\s*({.+?})\s*;', webpage, + 'data', default='{}'), + display_id or video_id, fatal=False) + # https://abc.com/shows/modern-family/episode-guide/season-01/101-pilot + layout = try_get(data, lambda x: x['page']['content']['video']['layout'], dict) + video_id = None + if layout: + video_id = try_get( + layout, + (lambda x: x['videoid'], lambda x: x['video']['id']), + compat_str) + if not video_id: + video_id = self._search_regex( + ( + # There may be inner quotes, e.g. data-video-id="'VDKA3609139'" + # from http://freeform.go.com/shows/shadowhunters/episodes/season-2/1-this-guilty-blood + r'data-video-id=["\']*(VDKA\w+)', + # page.analytics.videoIdCode + r'\bvideoIdCode["\']\s*:\s*["\']((?:vdka|VDKA)\w+)', + # https://abc.com/shows/the-rookie/episode-guide/season-02/03-the-bet + r'\b(?:video)?id["\']\s*:\s*["\'](VDKA\w+)' + ), webpage, 'video id', default=video_id) if not site_info: brand = self._search_regex( (r'data-brand=\s*["\']\s*(\d+)', diff --git a/youtube_dl/extractor/googledrive.py b/youtube_dl/extractor/googledrive.py index de8c80e36..3f2de00f1 100644 --- a/youtube_dl/extractor/googledrive.py +++ b/youtube_dl/extractor/googledrive.py @@ -7,6 +7,7 @@ from ..compat import compat_parse_qs from ..utils import ( determine_ext, ExtractorError, + get_element_by_class, int_or_none, lowercase_escape, try_get, @@ -237,7 +238,7 @@ class GoogleDriveIE(InfoExtractor): if confirmation_webpage: confirm = self._search_regex( r'confirm=([^&"\']+)', confirmation_webpage, - 'confirmation code', fatal=False) + 'confirmation code', default=None) if confirm: confirmed_source_url = update_url_query(source_url, { 'confirm': confirm, @@ -245,6 +246,11 @@ class GoogleDriveIE(InfoExtractor): urlh = request_source_file(confirmed_source_url, 'confirmed source') if urlh and urlh.headers.get('Content-Disposition'): add_source_format(urlh) + else: + self.report_warning( + get_element_by_class('uc-error-subcaption', confirmation_webpage) + or get_element_by_class('uc-error-caption', confirmation_webpage) + or 'unable to extract confirmation code') if not formats and reason: raise ExtractorError(reason, expected=True) diff --git a/youtube_dl/extractor/ign.py b/youtube_dl/extractor/ign.py index a96ea8010..0d9f50ed2 100644 --- a/youtube_dl/extractor/ign.py +++ b/youtube_dl/extractor/ign.py @@ -3,230 +3,255 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import ( + compat_parse_qs, + compat_urllib_parse_urlparse, +) from ..utils import ( + HEADRequest, + determine_ext, int_or_none, parse_iso8601, + strip_or_none, + try_get, ) -class IGNIE(InfoExtractor): +class IGNBaseIE(InfoExtractor): + def _call_api(self, slug): + return self._download_json( + 'http://apis.ign.com/{0}/v3/{0}s/slug/{1}'.format(self._PAGE_TYPE, slug), slug) + + +class IGNIE(IGNBaseIE): """ Extractor for some of the IGN sites, like www.ign.com, es.ign.com de.ign.com. Some videos of it.ign.com are also supported """ - _VALID_URL = r'https?://.+?\.ign\.com/(?:[^/]+/)?(?P<type>videos|show_videos|articles|feature|(?:[^/]+/\d+/video))(/.+)?/(?P<name_or_id>.+)' + _VALID_URL = r'https?://(?:.+?\.ign|www\.pcmag)\.com/videos/(?:\d{4}/\d{2}/\d{2}/)?(?P<id>[^/?&#]+)' IE_NAME = 'ign.com' + _PAGE_TYPE = 'video' - _API_URL_TEMPLATE = 'http://apis.ign.com/video/v3/videos/%s' - _EMBED_RE = r'<iframe[^>]+?["\']((?:https?:)?//.+?\.ign\.com.+?/embed.+?)["\']' - - _TESTS = [ - { - 'url': 'http://www.ign.com/videos/2013/06/05/the-last-of-us-review', - 'md5': 'febda82c4bafecd2d44b6e1a18a595f8', - 'info_dict': { - 'id': '8f862beef863986b2785559b9e1aa599', - 'ext': 'mp4', - 'title': 'The Last of Us Review', - 'description': 'md5:c8946d4260a4d43a00d5ae8ed998870c', - 'timestamp': 1370440800, - 'upload_date': '20130605', - 'uploader_id': 'cberidon@ign.com', - } - }, - { - 'url': 'http://me.ign.com/en/feature/15775/100-little-things-in-gta-5-that-will-blow-your-mind', - 'info_dict': { - 'id': '100-little-things-in-gta-5-that-will-blow-your-mind', - }, - 'playlist': [ - { - 'info_dict': { - 'id': '5ebbd138523268b93c9141af17bec937', - 'ext': 'mp4', - 'title': 'GTA 5 Video Review', - 'description': 'Rockstar drops the mic on this generation of games. Watch our review of the masterly Grand Theft Auto V.', - 'timestamp': 1379339880, - 'upload_date': '20130916', - 'uploader_id': 'danieljkrupa@gmail.com', - }, - }, - { - 'info_dict': { - 'id': '638672ee848ae4ff108df2a296418ee2', - 'ext': 'mp4', - 'title': '26 Twisted Moments from GTA 5 in Slow Motion', - 'description': 'The twisted beauty of GTA 5 in stunning slow motion.', - 'timestamp': 1386878820, - 'upload_date': '20131212', - 'uploader_id': 'togilvie@ign.com', - }, - }, - ], - 'params': { - 'skip_download': True, - }, - }, - { - 'url': 'http://www.ign.com/articles/2014/08/15/rewind-theater-wild-trailer-gamescom-2014?watch', - 'md5': '618fedb9c901fd086f6f093564ef8558', - 'info_dict': { - 'id': '078fdd005f6d3c02f63d795faa1b984f', - 'ext': 'mp4', - 'title': 'Rewind Theater - Wild Trailer Gamescom 2014', - 'description': 'Brian and Jared explore Michel Ancel\'s captivating new preview.', - 'timestamp': 1408047180, - 'upload_date': '20140814', - 'uploader_id': 'jamesduggan1990@gmail.com', - }, - }, - { - 'url': 'http://me.ign.com/en/videos/112203/video/how-hitman-aims-to-be-different-than-every-other-s', - 'only_matching': True, - }, - { - 'url': 'http://me.ign.com/ar/angry-birds-2/106533/video/lrd-ldyy-lwl-lfylm-angry-birds', - 'only_matching': True, - }, - { - # videoId pattern - 'url': 'http://www.ign.com/articles/2017/06/08/new-ducktales-short-donalds-birthday-doesnt-go-as-planned', - 'only_matching': True, - }, - ] - - def _find_video_id(self, webpage): - res_id = [ - r'"video_id"\s*:\s*"(.*?)"', - r'class="hero-poster[^"]*?"[^>]*id="(.+?)"', - r'data-video-id="(.+?)"', - r'<object id="vid_(.+?)"', - r'<meta name="og:image" content=".*/(.+?)-(.+?)/.+.jpg"', - r'videoId"\s*:\s*"(.+?)"', - r'videoId["\']\s*:\s*["\']([^"\']+?)["\']', - ] - return self._search_regex(res_id, webpage, 'video id', default=None) + _TESTS = [{ + 'url': 'http://www.ign.com/videos/2013/06/05/the-last-of-us-review', + 'md5': 'd2e1586d9987d40fad7867bf96a018ea', + 'info_dict': { + 'id': '8f862beef863986b2785559b9e1aa599', + 'ext': 'mp4', + 'title': 'The Last of Us Review', + 'description': 'md5:c8946d4260a4d43a00d5ae8ed998870c', + 'timestamp': 1370440800, + 'upload_date': '20130605', + 'tags': 'count:9', + } + }, { + 'url': 'http://www.pcmag.com/videos/2015/01/06/010615-whats-new-now-is-gogo-snooping-on-your-data', + 'md5': 'f1581a6fe8c5121be5b807684aeac3f6', + 'info_dict': { + 'id': 'ee10d774b508c9b8ec07e763b9125b91', + 'ext': 'mp4', + 'title': 'What\'s New Now: Is GoGo Snooping on Your Data?', + 'description': 'md5:817a20299de610bd56f13175386da6fa', + 'timestamp': 1420571160, + 'upload_date': '20150106', + 'tags': 'count:4', + } + }, { + 'url': 'https://www.ign.com/videos/is-a-resident-evil-4-remake-on-the-way-ign-daily-fix', + 'only_matching': True, + }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - name_or_id = mobj.group('name_or_id') - page_type = mobj.group('type') - webpage = self._download_webpage(url, name_or_id) - if page_type != 'video': - multiple_urls = re.findall( - r'<param name="flashvars"[^>]*value="[^"]*?url=(https?://www\.ign\.com/videos/.*?)["&]', - webpage) - if multiple_urls: - entries = [self.url_result(u, ie='IGN') for u in multiple_urls] - return { - '_type': 'playlist', - 'id': name_or_id, - 'entries': entries, - } - - video_id = self._find_video_id(webpage) - if not video_id: - return self.url_result(self._search_regex( - self._EMBED_RE, webpage, 'embed url')) - return self._get_video_info(video_id) - - def _get_video_info(self, video_id): - api_data = self._download_json( - self._API_URL_TEMPLATE % video_id, video_id) + display_id = self._match_id(url) + video = self._call_api(display_id) + video_id = video['videoId'] + metadata = video['metadata'] + title = metadata.get('longTitle') or metadata.get('title') or metadata['name'] formats = [] - m3u8_url = api_data['refs'].get('m3uUrl') + refs = video.get('refs') or {} + + m3u8_url = refs.get('m3uUrl') if m3u8_url: formats.extend(self._extract_m3u8_formats( m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) - f4m_url = api_data['refs'].get('f4mUrl') + + f4m_url = refs.get('f4mUrl') if f4m_url: formats.extend(self._extract_f4m_formats( f4m_url, video_id, f4m_id='hds', fatal=False)) - for asset in api_data['assets']: + + for asset in (video.get('assets') or []): + asset_url = asset.get('url') + if not asset_url: + continue formats.append({ - 'url': asset['url'], - 'tbr': asset.get('actual_bitrate_kbps'), - 'fps': asset.get('frame_rate'), + 'url': asset_url, + 'tbr': int_or_none(asset.get('bitrate'), 1000), + 'fps': int_or_none(asset.get('frame_rate')), 'height': int_or_none(asset.get('height')), 'width': int_or_none(asset.get('width')), }) + + mezzanine_url = try_get(video, lambda x: x['system']['mezzanineUrl']) + if mezzanine_url: + formats.append({ + 'ext': determine_ext(mezzanine_url, 'mp4'), + 'format_id': 'mezzanine', + 'preference': 1, + 'url': mezzanine_url, + }) + self._sort_formats(formats) - thumbnails = [{ - 'url': thumbnail['url'] - } for thumbnail in api_data.get('thumbnails', [])] + thumbnails = [] + for thumbnail in (video.get('thumbnails') or []): + thumbnail_url = thumbnail.get('url') + if not thumbnail_url: + continue + thumbnails.append({ + 'url': thumbnail_url, + }) - metadata = api_data['metadata'] + tags = [] + for tag in (video.get('tags') or []): + display_name = tag.get('displayName') + if not display_name: + continue + tags.append(display_name) return { - 'id': api_data.get('videoId') or video_id, - 'title': metadata.get('longTitle') or metadata.get('name') or metadata.get['title'], - 'description': metadata.get('description'), + 'id': video_id, + 'title': title, + 'description': strip_or_none(metadata.get('description')), 'timestamp': parse_iso8601(metadata.get('publishDate')), 'duration': int_or_none(metadata.get('duration')), - 'display_id': metadata.get('slug') or video_id, - 'uploader_id': metadata.get('creator'), + 'display_id': display_id, 'thumbnails': thumbnails, 'formats': formats, + 'tags': tags, } -class OneUPIE(IGNIE): - _VALID_URL = r'https?://gamevideos\.1up\.com/(?P<type>video)/id/(?P<name_or_id>.+)\.html' - IE_NAME = '1up.com' - +class IGNVideoIE(InfoExtractor): + _VALID_URL = r'https?://.+?\.ign\.com/(?:[a-z]{2}/)?[^/]+/(?P<id>\d+)/(?:video|trailer)/' _TESTS = [{ - 'url': 'http://gamevideos.1up.com/video/id/34976.html', - 'md5': 'c9cc69e07acb675c31a16719f909e347', + 'url': 'http://me.ign.com/en/videos/112203/video/how-hitman-aims-to-be-different-than-every-other-s', + 'md5': 'dd9aca7ed2657c4e118d8b261e5e9de1', 'info_dict': { - 'id': '34976', + 'id': 'e9be7ea899a9bbfc0674accc22a36cc8', 'ext': 'mp4', - 'title': 'Sniper Elite V2 - Trailer', - 'description': 'md5:bf0516c5ee32a3217aa703e9b1bc7826', - 'timestamp': 1313099220, - 'upload_date': '20110811', - 'uploader_id': 'IGN', + 'title': 'How Hitman Aims to Be Different Than Every Other Stealth Game - NYCC 2015', + 'description': 'Taking out assassination targets in Hitman has never been more stylish.', + 'timestamp': 1444665600, + 'upload_date': '20151012', } + }, { + 'url': 'http://me.ign.com/ar/angry-birds-2/106533/video/lrd-ldyy-lwl-lfylm-angry-birds', + 'only_matching': True, + }, { + # Youtube embed + 'url': 'https://me.ign.com/ar/ratchet-clank-rift-apart/144327/trailer/embed', + 'only_matching': True, + }, { + # Twitter embed + 'url': 'http://adria.ign.com/sherlock-season-4/9687/trailer/embed', + 'only_matching': True, + }, { + # Vimeo embed + 'url': 'https://kr.ign.com/bic-2018/3307/trailer/embed', + 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - result = super(OneUPIE, self)._real_extract(url) - result['id'] = mobj.group('name_or_id') - return result - - -class PCMagIE(IGNIE): - _VALID_URL = r'https?://(?:www\.)?pcmag\.com/(?P<type>videos|article2)(/.+)?/(?P<name_or_id>.+)' - IE_NAME = 'pcmag' + video_id = self._match_id(url) + req = HEADRequest(url.rsplit('/', 1)[0] + '/embed') + url = self._request_webpage(req, video_id).geturl() + ign_url = compat_parse_qs( + compat_urllib_parse_urlparse(url).query).get('url', [None])[0] + if ign_url: + return self.url_result(ign_url, IGNIE.ie_key()) + return self.url_result(url) - _EMBED_RE = r'iframe\.setAttribute\("src",\s*__util.objToUrlString\("http://widgets\.ign\.com/video/embed/content\.html?[^"]*url=([^"]+)["&]' +class IGNArticleIE(IGNBaseIE): + _VALID_URL = r'https?://.+?\.ign\.com/(?:articles(?:/\d{4}/\d{2}/\d{2})?|(?:[a-z]{2}/)?feature/\d+)/(?P<id>[^/?&#]+)' + _PAGE_TYPE = 'article' _TESTS = [{ - 'url': 'http://www.pcmag.com/videos/2015/01/06/010615-whats-new-now-is-gogo-snooping-on-your-data', - 'md5': '212d6154fd0361a2781075f1febbe9ad', + 'url': 'http://me.ign.com/en/feature/15775/100-little-things-in-gta-5-that-will-blow-your-mind', 'info_dict': { - 'id': 'ee10d774b508c9b8ec07e763b9125b91', - 'ext': 'mp4', - 'title': '010615_What\'s New Now: Is GoGo Snooping on Your Data?', - 'description': 'md5:a7071ae64d2f68cc821c729d4ded6bb3', - 'timestamp': 1420571160, - 'upload_date': '20150106', - 'uploader_id': 'cozzipix@gmail.com', - } + 'id': '524497489e4e8ff5848ece34', + 'title': '100 Little Things in GTA 5 That Will Blow Your Mind', + }, + 'playlist': [ + { + 'info_dict': { + 'id': '5ebbd138523268b93c9141af17bec937', + 'ext': 'mp4', + 'title': 'GTA 5 Video Review', + 'description': 'Rockstar drops the mic on this generation of games. Watch our review of the masterly Grand Theft Auto V.', + 'timestamp': 1379339880, + 'upload_date': '20130916', + }, + }, + { + 'info_dict': { + 'id': '638672ee848ae4ff108df2a296418ee2', + 'ext': 'mp4', + 'title': '26 Twisted Moments from GTA 5 in Slow Motion', + 'description': 'The twisted beauty of GTA 5 in stunning slow motion.', + 'timestamp': 1386878820, + 'upload_date': '20131212', + }, + }, + ], + 'params': { + 'playlist_items': '2-3', + 'skip_download': True, + }, }, { - 'url': 'http://www.pcmag.com/article2/0,2817,2470156,00.asp', - 'md5': '94130c1ca07ba0adb6088350681f16c1', + 'url': 'http://www.ign.com/articles/2014/08/15/rewind-theater-wild-trailer-gamescom-2014?watch', 'info_dict': { - 'id': '042e560ba94823d43afcb12ddf7142ca', - 'ext': 'mp4', - 'title': 'HTC\'s Weird New Re Camera - What\'s New Now', - 'description': 'md5:53433c45df96d2ea5d0fda18be2ca908', - 'timestamp': 1412953920, - 'upload_date': '20141010', - 'uploader_id': 'chris_snyder@pcmag.com', - } + 'id': '53ee806780a81ec46e0790f8', + 'title': 'Rewind Theater - Wild Trailer Gamescom 2014', + }, + 'playlist_count': 2, + }, { + # videoId pattern + 'url': 'http://www.ign.com/articles/2017/06/08/new-ducktales-short-donalds-birthday-doesnt-go-as-planned', + 'only_matching': True, + }, { + # Youtube embed + 'url': 'https://www.ign.com/articles/2021-mvp-named-in-puppy-bowl-xvii', + 'only_matching': True, + }, { + # IMDB embed + 'url': 'https://www.ign.com/articles/2014/08/07/sons-of-anarchy-final-season-trailer', + 'only_matching': True, + }, { + # Facebook embed + 'url': 'https://www.ign.com/articles/2017/09/20/marvels-the-punisher-watch-the-new-trailer-for-the-netflix-series', + 'only_matching': True, + }, { + # Brightcove embed + 'url': 'https://www.ign.com/articles/2016/01/16/supergirl-goes-flying-with-martian-manhunter-in-new-clip', + 'only_matching': True, }] + + def _real_extract(self, url): + display_id = self._match_id(url) + article = self._call_api(display_id) + + def entries(): + media_url = try_get(article, lambda x: x['mediaRelations'][0]['media']['metadata']['url']) + if media_url: + yield self.url_result(media_url, IGNIE.ie_key()) + for content in (article.get('content') or []): + for video_url in re.findall(r'(?:\[(?:ignvideo\s+url|youtube\s+clip_id)|<iframe[^>]+src)="([^"]+)"', content): + yield self.url_result(video_url) + + return self.playlist_result( + entries(), article.get('articleId'), + strip_or_none(try_get(article, lambda x: x['metadata']['headline']))) diff --git a/youtube_dl/extractor/instagram.py b/youtube_dl/extractor/instagram.py index 1eeddc3b6..12e10143c 100644 --- a/youtube_dl/extractor/instagram.py +++ b/youtube_dl/extractor/instagram.py @@ -12,6 +12,7 @@ from ..compat import ( ) from ..utils import ( ExtractorError, + float_or_none, get_element_by_attribute, int_or_none, lowercase_escape, @@ -32,6 +33,7 @@ class InstagramIE(InfoExtractor): 'title': 'Video by naomipq', 'description': 'md5:1f17f0ab29bd6fe2bfad705f58de3cb8', 'thumbnail': r're:^https?://.*\.jpg', + 'duration': 0, 'timestamp': 1371748545, 'upload_date': '20130620', 'uploader_id': 'naomipq', @@ -48,6 +50,7 @@ class InstagramIE(InfoExtractor): 'ext': 'mp4', 'title': 'Video by britneyspears', 'thumbnail': r're:^https?://.*\.jpg', + 'duration': 0, 'timestamp': 1453760977, 'upload_date': '20160125', 'uploader_id': 'britneyspears', @@ -87,6 +90,24 @@ class InstagramIE(InfoExtractor): 'description': 'md5:0f9203fc6a2ce4d228da5754bcf54957', }, }, { + # IGTV + 'url': 'https://www.instagram.com/tv/BkfuX9UB-eK/', + 'info_dict': { + 'id': 'BkfuX9UB-eK', + 'ext': 'mp4', + 'title': 'Fingerboarding Tricks with @cass.fb', + 'thumbnail': r're:^https?://.*\.jpg', + 'duration': 53.83, + 'timestamp': 1530032919, + 'upload_date': '20180626', + 'uploader_id': 'instagram', + 'uploader': 'Instagram', + 'like_count': int, + 'comment_count': int, + 'comments': list, + 'description': 'Meet Cass Hirst (@cass.fb), a fingerboarding pro who can perform tiny ollies and kickflips while blindfolded.', + } + }, { 'url': 'https://instagram.com/p/-Cmh1cukG2/', 'only_matching': True, }, { @@ -159,7 +180,9 @@ class InstagramIE(InfoExtractor): description = try_get( media, lambda x: x['edge_media_to_caption']['edges'][0]['node']['text'], compat_str) or media.get('caption') + title = media.get('title') thumbnail = media.get('display_src') or media.get('display_url') + duration = float_or_none(media.get('video_duration')) timestamp = int_or_none(media.get('taken_at_timestamp') or media.get('date')) uploader = media.get('owner', {}).get('full_name') uploader_id = media.get('owner', {}).get('username') @@ -200,9 +223,10 @@ class InstagramIE(InfoExtractor): continue entries.append({ 'id': node.get('shortcode') or node['id'], - 'title': 'Video %d' % edge_num, + 'title': node.get('title') or 'Video %d' % edge_num, 'url': node_video_url, 'thumbnail': node.get('display_url'), + 'duration': float_or_none(node.get('video_duration')), 'width': int_or_none(try_get(node, lambda x: x['dimensions']['width'])), 'height': int_or_none(try_get(node, lambda x: x['dimensions']['height'])), 'view_count': int_or_none(node.get('video_view_count')), @@ -239,8 +263,9 @@ class InstagramIE(InfoExtractor): 'id': video_id, 'formats': formats, 'ext': 'mp4', - 'title': 'Video by %s' % uploader_id, + 'title': title or 'Video by %s' % uploader_id, 'description': description, + 'duration': duration, 'thumbnail': thumbnail, 'timestamp': timestamp, 'uploader_id': uploader_id, diff --git a/youtube_dl/extractor/jamendo.py b/youtube_dl/extractor/jamendo.py index 490efa8fb..1db7c64af 100644 --- a/youtube_dl/extractor/jamendo.py +++ b/youtube_dl/extractor/jamendo.py @@ -29,34 +29,51 @@ class JamendoIE(InfoExtractor): 'id': '196219', 'display_id': 'stories-from-emona-i', 'ext': 'flac', - 'title': 'Maya Filipič - Stories from Emona I', - 'artist': 'Maya Filipič', + # 'title': 'Maya Filipič - Stories from Emona I', + 'title': 'Stories from Emona I', + # 'artist': 'Maya Filipič', 'track': 'Stories from Emona I', 'duration': 210, 'thumbnail': r're:^https?://.*\.jpg', 'timestamp': 1217438117, 'upload_date': '20080730', + 'license': 'by-nc-nd', + 'view_count': int, + 'like_count': int, + 'average_rating': int, + 'tags': ['piano', 'peaceful', 'newage', 'strings', 'upbeat'], } }, { 'url': 'https://licensing.jamendo.com/en/track/1496667/energetic-rock', 'only_matching': True, }] + def _call_api(self, resource, resource_id): + path = '/api/%ss' % resource + rand = compat_str(random.random()) + return self._download_json( + 'https://www.jamendo.com' + path, resource_id, query={ + 'id[]': resource_id, + }, headers={ + 'X-Jam-Call': '$%s*%s~' % (hashlib.sha1((path + rand).encode()).hexdigest(), rand) + })[0] + def _real_extract(self, url): track_id, display_id = self._VALID_URL_RE.match(url).groups() - webpage = self._download_webpage( - 'https://www.jamendo.com/track/' + track_id, track_id) - models = self._parse_json(self._html_search_regex( - r"data-bundled-models='([^']+)", - webpage, 'bundled models'), track_id) - track = models['track']['models'][0] + # webpage = self._download_webpage( + # 'https://www.jamendo.com/track/' + track_id, track_id) + # models = self._parse_json(self._html_search_regex( + # r"data-bundled-models='([^']+)", + # webpage, 'bundled models'), track_id) + # track = models['track']['models'][0] + track = self._call_api('track', track_id) title = track_name = track['name'] - get_model = lambda x: try_get(models, lambda y: y[x]['models'][0], dict) or {} - artist = get_model('artist') - artist_name = artist.get('name') - if artist_name: - title = '%s - %s' % (artist_name, title) - album = get_model('album') + # get_model = lambda x: try_get(models, lambda y: y[x]['models'][0], dict) or {} + # artist = get_model('artist') + # artist_name = artist.get('name') + # if artist_name: + # title = '%s - %s' % (artist_name, title) + # album = get_model('album') formats = [{ 'url': 'https://%s.jamendo.com/?trackid=%s&format=%s&from=app-97dab294' @@ -74,7 +91,7 @@ class JamendoIE(InfoExtractor): urls = [] thumbnails = [] - for _, covers in track.get('cover', {}).items(): + for covers in (track.get('cover') or {}).values(): for cover_id, cover_url in covers.items(): if not cover_url or cover_url in urls: continue @@ -88,13 +105,14 @@ class JamendoIE(InfoExtractor): }) tags = [] - for tag in track.get('tags', []): + for tag in (track.get('tags') or []): tag_name = tag.get('name') if not tag_name: continue tags.append(tag_name) stats = track.get('stats') or {} + license = track.get('licenseCC') or [] return { 'id': track_id, @@ -103,11 +121,11 @@ class JamendoIE(InfoExtractor): 'title': title, 'description': track.get('description'), 'duration': int_or_none(track.get('duration')), - 'artist': artist_name, + # 'artist': artist_name, 'track': track_name, - 'album': album.get('name'), + # 'album': album.get('name'), 'formats': formats, - 'license': '-'.join(track.get('licenseCC', [])) or None, + 'license': '-'.join(license) if license else None, 'timestamp': int_or_none(track.get('dateCreated')), 'view_count': int_or_none(stats.get('listenedAll')), 'like_count': int_or_none(stats.get('favorited')), @@ -116,9 +134,9 @@ class JamendoIE(InfoExtractor): } -class JamendoAlbumIE(InfoExtractor): +class JamendoAlbumIE(JamendoIE): _VALID_URL = r'https?://(?:www\.)?jamendo\.com/album/(?P<id>[0-9]+)' - _TEST = { + _TESTS = [{ 'url': 'https://www.jamendo.com/album/121486/duck-on-cover', 'info_dict': { 'id': '121486', @@ -151,17 +169,7 @@ class JamendoAlbumIE(InfoExtractor): 'params': { 'playlistend': 2 } - } - - def _call_api(self, resource, resource_id): - path = '/api/%ss' % resource - rand = compat_str(random.random()) - return self._download_json( - 'https://www.jamendo.com' + path, resource_id, query={ - 'id[]': resource_id, - }, headers={ - 'X-Jam-Call': '$%s*%s~' % (hashlib.sha1((path + rand).encode()).hexdigest(), rand) - })[0] + }] def _real_extract(self, url): album_id = self._match_id(url) @@ -169,7 +177,7 @@ class JamendoAlbumIE(InfoExtractor): album_name = album.get('name') entries = [] - for track in album.get('tracks', []): + for track in (album.get('tracks') or []): track_id = track.get('id') if not track_id: continue diff --git a/youtube_dl/extractor/kakao.py b/youtube_dl/extractor/kakao.py index 32935bb28..31ce7a85c 100644 --- a/youtube_dl/extractor/kakao.py +++ b/youtube_dl/extractor/kakao.py @@ -3,10 +3,13 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..compat import compat_str +from ..compat import compat_HTTPError from ..utils import ( + ExtractorError, int_or_none, + str_or_none, strip_or_none, + try_get, unified_timestamp, update_url_query, ) @@ -23,7 +26,7 @@ class KakaoIE(InfoExtractor): 'id': '301965083', 'ext': 'mp4', 'title': '乃木坂46 バナナマン 「3期生紹介コーナーが始動!顔高低差GPも!」 『乃木坂工事中』', - 'uploader_id': 2671005, + 'uploader_id': '2671005', 'uploader': '그랑그랑이', 'timestamp': 1488160199, 'upload_date': '20170227', @@ -36,11 +39,15 @@ class KakaoIE(InfoExtractor): 'ext': 'mp4', 'description': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)\r\n\r\n[쇼! 음악중심] 20160611, 507회', 'title': '러블리즈 - Destiny (나의 지구) (Lovelyz - Destiny)', - 'uploader_id': 2653210, + 'uploader_id': '2653210', 'uploader': '쇼! 음악중심', 'timestamp': 1485684628, 'upload_date': '20170129', } + }, { + # geo restricted + 'url': 'https://tv.kakao.com/channel/3643855/cliplink/412069491', + 'only_matching': True, }] def _real_extract(self, url): @@ -68,8 +75,7 @@ class KakaoIE(InfoExtractor): 'fields': ','.join([ '-*', 'tid', 'clipLink', 'displayTitle', 'clip', 'title', 'description', 'channelId', 'createTime', 'duration', 'playCount', - 'likeCount', 'commentCount', 'tagList', 'channel', 'name', - 'clipChapterThumbnailList', 'thumbnailUrl', 'timeInSec', 'isDefault', + 'likeCount', 'commentCount', 'tagList', 'channel', 'name', 'thumbnailUrl', 'videoOutputList', 'width', 'height', 'kbps', 'profile', 'label']) } @@ -82,24 +88,28 @@ class KakaoIE(InfoExtractor): title = clip.get('title') or clip_link.get('displayTitle') - query['tid'] = impress.get('tid', '') + query.update({ + 'fields': '-*,code,message,url', + 'tid': impress.get('tid') or '', + }) formats = [] - for fmt in clip.get('videoOutputList', []): + for fmt in (clip.get('videoOutputList') or []): try: profile_name = fmt['profile'] if profile_name == 'AUDIO': continue - query.update({ - 'profile': profile_name, - 'fields': '-*,url', - }) - fmt_url_json = self._download_json( - api_base + 'raw/videolocation', display_id, - 'Downloading video URL for profile %s' % profile_name, - query=query, headers=player_header, fatal=False) - - if fmt_url_json is None: + query['profile'] = profile_name + try: + fmt_url_json = self._download_json( + api_base + 'raw/videolocation', display_id, + 'Downloading video URL for profile %s' % profile_name, + query=query, headers=player_header) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: + resp = self._parse_json(e.cause.read().decode(), video_id) + if resp.get('code') == 'GeoBlocked': + self.raise_geo_restricted() continue fmt_url = fmt_url_json['url'] @@ -116,27 +126,13 @@ class KakaoIE(InfoExtractor): pass self._sort_formats(formats) - thumbs = [] - for thumb in clip.get('clipChapterThumbnailList', []): - thumbs.append({ - 'url': thumb.get('thumbnailUrl'), - 'id': compat_str(thumb.get('timeInSec')), - 'preference': -1 if thumb.get('isDefault') else 0 - }) - top_thumbnail = clip.get('thumbnailUrl') - if top_thumbnail: - thumbs.append({ - 'url': top_thumbnail, - 'preference': 10, - }) - return { 'id': display_id, 'title': title, 'description': strip_or_none(clip.get('description')), - 'uploader': clip_link.get('channel', {}).get('name'), - 'uploader_id': clip_link.get('channelId'), - 'thumbnails': thumbs, + 'uploader': try_get(clip_link, lambda x: x['channel']['name']), + 'uploader_id': str_or_none(clip_link.get('channelId')), + 'thumbnail': clip.get('thumbnailUrl'), 'timestamp': unified_timestamp(clip_link.get('createTime')), 'duration': int_or_none(clip.get('duration')), 'view_count': int_or_none(clip.get('playCount')), diff --git a/youtube_dl/extractor/kaltura.py b/youtube_dl/extractor/kaltura.py index 49d13460d..c731612c4 100644 --- a/youtube_dl/extractor/kaltura.py +++ b/youtube_dl/extractor/kaltura.py @@ -120,7 +120,7 @@ class KalturaIE(InfoExtractor): def _extract_urls(webpage): # Embed codes: https://knowledge.kaltura.com/embedding-kaltura-media-players-your-site finditer = ( - re.finditer( + list(re.finditer( r"""(?xs) kWidget\.(?:thumb)?[Ee]mbed\( \{.*? @@ -128,8 +128,8 @@ class KalturaIE(InfoExtractor): (?P<q2>['"])_?(?P<partner_id>(?:(?!(?P=q2)).)+)(?P=q2),.*? (?P<q3>['"])entry_?[Ii]d(?P=q3)\s*:\s* (?P<q4>['"])(?P<id>(?:(?!(?P=q4)).)+)(?P=q4)(?:,|\s*\}) - """, webpage) - or re.finditer( + """, webpage)) + or list(re.finditer( r'''(?xs) (?P<q1>["']) (?:https?:)?//cdnapi(?:sec)?\.kaltura\.com(?::\d+)?/(?:(?!(?P=q1)).)*\b(?:p|partner_id)/(?P<partner_id>\d+)(?:(?!(?P=q1)).)* @@ -142,16 +142,16 @@ class KalturaIE(InfoExtractor): \[\s*(?P<q2_1>["'])entry_?[Ii]d(?P=q2_1)\s*\]\s*=\s* ) (?P<q3>["'])(?P<id>(?:(?!(?P=q3)).)+)(?P=q3) - ''', webpage) - or re.finditer( + ''', webpage)) + or list(re.finditer( r'''(?xs) - <(?:iframe[^>]+src|meta[^>]+\bcontent)=(?P<q1>["']) + <(?:iframe[^>]+src|meta[^>]+\bcontent)=(?P<q1>["'])\s* (?:https?:)?//(?:(?:www|cdnapi(?:sec)?)\.)?kaltura\.com/(?:(?!(?P=q1)).)*\b(?:p|partner_id)/(?P<partner_id>\d+) (?:(?!(?P=q1)).)* [?&;]entry_id=(?P<id>(?:(?!(?P=q1))[^&])+) (?:(?!(?P=q1)).)* (?P=q1) - ''', webpage) + ''', webpage)) ) urls = [] for mobj in finditer: diff --git a/youtube_dl/extractor/khanacademy.py b/youtube_dl/extractor/khanacademy.py index 61739efa7..87e520378 100644 --- a/youtube_dl/extractor/khanacademy.py +++ b/youtube_dl/extractor/khanacademy.py @@ -1,82 +1,107 @@ from __future__ import unicode_literals -import re +import json from .common import InfoExtractor from ..utils import ( - unified_strdate, + int_or_none, + parse_iso8601, + try_get, ) -class KhanAcademyIE(InfoExtractor): - _VALID_URL = r'^https?://(?:(?:www|api)\.)?khanacademy\.org/(?P<key>[^/]+)/(?:[^/]+/){,2}(?P<id>[^?#/]+)(?:$|[?#])' - IE_NAME = 'KhanAcademy' +class KhanAcademyBaseIE(InfoExtractor): + _VALID_URL_TEMPL = r'https?://(?:www\.)?khanacademy\.org/(?P<id>(?:[^/]+/){%s}%s[^?#/&]+)' - _TESTS = [{ - 'url': 'http://www.khanacademy.org/video/one-time-pad', - 'md5': '7b391cce85e758fb94f763ddc1bbb979', + def _parse_video(self, video): + return { + '_type': 'url_transparent', + 'url': video['youtubeId'], + 'id': video.get('slug'), + 'title': video.get('title'), + 'thumbnail': video.get('imageUrl') or video.get('thumbnailUrl'), + 'duration': int_or_none(video.get('duration')), + 'description': video.get('description'), + 'ie_key': 'Youtube', + } + + def _real_extract(self, url): + display_id = self._match_id(url) + component_props = self._parse_json(self._download_json( + 'https://www.khanacademy.org/api/internal/graphql', + display_id, query={ + 'hash': 1604303425, + 'variables': json.dumps({ + 'path': display_id, + 'queryParams': '', + }), + })['data']['contentJson'], display_id)['componentProps'] + return self._parse_component_props(component_props) + + +class KhanAcademyIE(KhanAcademyBaseIE): + IE_NAME = 'khanacademy' + _VALID_URL = KhanAcademyBaseIE._VALID_URL_TEMPL % ('4', 'v/') + _TEST = { + 'url': 'https://www.khanacademy.org/computing/computer-science/cryptography/crypt/v/one-time-pad', + 'md5': '9c84b7b06f9ebb80d22a5c8dedefb9a0', 'info_dict': { - 'id': 'one-time-pad', - 'ext': 'webm', + 'id': 'FlIG3TvQCBQ', + 'ext': 'mp4', 'title': 'The one-time pad', 'description': 'The perfect cipher', 'duration': 176, 'uploader': 'Brit Cruise', 'uploader_id': 'khanacademy', 'upload_date': '20120411', + 'timestamp': 1334170113, + 'license': 'cc-by-nc-sa', }, 'add_ie': ['Youtube'], - }, { - 'url': 'https://www.khanacademy.org/math/applied-math/cryptography', + } + + def _parse_component_props(self, component_props): + video = component_props['tutorialPageData']['contentModel'] + info = self._parse_video(video) + author_names = video.get('authorNames') + info.update({ + 'uploader': ', '.join(author_names) if author_names else None, + 'timestamp': parse_iso8601(video.get('dateAdded')), + 'license': video.get('kaUserLicense'), + }) + return info + + +class KhanAcademyUnitIE(KhanAcademyBaseIE): + IE_NAME = 'khanacademy:unit' + _VALID_URL = (KhanAcademyBaseIE._VALID_URL_TEMPL % ('2', '')) + '/?(?:[?#&]|$)' + _TEST = { + 'url': 'https://www.khanacademy.org/computing/computer-science/cryptography', 'info_dict': { 'id': 'cryptography', - 'title': 'Journey into cryptography', + 'title': 'Cryptography', 'description': 'How have humans protected their secret messages through history? What has changed today?', }, - 'playlist_mincount': 3, - }] - - def _real_extract(self, url): - m = re.match(self._VALID_URL, url) - video_id = m.group('id') + 'playlist_mincount': 31, + } - if m.group('key') == 'video': - data = self._download_json( - 'http://api.khanacademy.org/api/v1/videos/' + video_id, - video_id, 'Downloading video info') + def _parse_component_props(self, component_props): + curation = component_props['curation'] - upload_date = unified_strdate(data['date_added']) - uploader = ', '.join(data['author_names']) - return { - '_type': 'url_transparent', - 'url': data['url'], - 'id': video_id, - 'title': data['title'], - 'thumbnail': data['image_url'], - 'duration': data['duration'], - 'description': data['description'], - 'uploader': uploader, - 'upload_date': upload_date, + entries = [] + tutorials = try_get(curation, lambda x: x['tabs'][0]['modules'][0]['tutorials'], list) or [] + for tutorial_number, tutorial in enumerate(tutorials, 1): + chapter_info = { + 'chapter': tutorial.get('title'), + 'chapter_number': tutorial_number, + 'chapter_id': tutorial.get('id'), } - else: - # topic - data = self._download_json( - 'http://api.khanacademy.org/api/v1/topic/' + video_id, - video_id, 'Downloading topic info') + for content_item in (tutorial.get('contentItems') or []): + if content_item.get('kind') == 'Video': + info = self._parse_video(content_item) + info.update(chapter_info) + entries.append(info) - entries = [ - { - '_type': 'url', - 'url': c['url'], - 'id': c['id'], - 'title': c['title'], - } - for c in data['children'] if c['kind'] in ('Video', 'Topic')] - - return { - '_type': 'playlist', - 'id': video_id, - 'title': data['title'], - 'description': data['description'], - 'entries': entries, - } + return self.playlist_result( + entries, curation.get('unit'), curation.get('title'), + curation.get('description')) diff --git a/youtube_dl/extractor/lbry.py b/youtube_dl/extractor/lbry.py index 41cc245eb..cfd6b8393 100644 --- a/youtube_dl/extractor/lbry.py +++ b/youtube_dl/extractor/lbry.py @@ -5,7 +5,12 @@ import functools import json from .common import InfoExtractor -from ..compat import compat_str +from ..compat import ( + compat_parse_qs, + compat_str, + compat_urllib_parse_unquote, + compat_urllib_parse_urlparse, +) from ..utils import ( determine_ext, ExtractorError, @@ -57,6 +62,7 @@ class LBRYBaseIE(InfoExtractor): 'description': stream_value.get('description'), 'license': stream_value.get('license'), 'timestamp': int_or_none(stream.get('timestamp')), + 'release_timestamp': int_or_none(stream_value.get('release_time')), 'tags': stream_value.get('tags'), 'duration': int_or_none(media.get('duration')), 'channel': try_get(signing_channel, lambda x: x['value']['title']), @@ -89,6 +95,8 @@ class LBRYIE(LBRYBaseIE): 'description': 'md5:f6cb5c704b332d37f5119313c2c98f51', 'timestamp': 1595694354, 'upload_date': '20200725', + 'release_timestamp': 1595340697, + 'release_date': '20200721', 'width': 1280, 'height': 720, } @@ -103,6 +111,8 @@ class LBRYIE(LBRYBaseIE): 'description': 'md5:661ac4f1db09f31728931d7b88807a61', 'timestamp': 1591312601, 'upload_date': '20200604', + 'release_timestamp': 1591312421, + 'release_date': '20200604', 'tags': list, 'duration': 2570, 'channel': 'The LBRY Foundation', @@ -111,6 +121,26 @@ class LBRYIE(LBRYBaseIE): 'vcodec': 'none', } }, { + # HLS + 'url': 'https://odysee.com/@gardeningincanada:b/plants-i-will-never-grow-again.-the:e', + 'md5': 'fc82f45ea54915b1495dd7cb5cc1289f', + 'info_dict': { + 'id': 'e51671357333fe22ae88aad320bde2f6f96b1410', + 'ext': 'mp4', + 'title': 'PLANTS I WILL NEVER GROW AGAIN. THE BLACK LIST PLANTS FOR A CANADIAN GARDEN | Gardening in Canada 🍁', + 'description': 'md5:9c539c6a03fb843956de61a4d5288d5e', + 'timestamp': 1618254123, + 'upload_date': '20210412', + 'release_timestamp': 1618254002, + 'release_date': '20210412', + 'tags': list, + 'duration': 554, + 'channel': 'Gardening In Canada', + 'channel_id': 'b8be0e93b423dad221abe29545fbe8ec36e806bc', + 'channel_url': 'https://odysee.com/@gardeningincanada:b8be0e93b423dad221abe29545fbe8ec36e806bc', + 'formats': 'mincount:3', + } + }, { 'url': 'https://odysee.com/@BrodieRobertson:5/apple-is-tracking-everything-you-do-on:e', 'only_matching': True, }, { @@ -131,6 +161,9 @@ class LBRYIE(LBRYBaseIE): }, { 'url': 'https://lbry.tv/$/download/Episode-1/e7d93d772bd87e2b62d5ab993c1c3ced86ebb396', 'only_matching': True, + }, { + 'url': 'https://lbry.tv/@lacajadepandora:a/TRUMP-EST%C3%81-BIEN-PUESTO-con-Pilar-Baselga,-Carlos-Senra,-Luis-Palacios-(720p_30fps_H264-192kbit_AAC):1', + 'only_matching': True, }] def _real_extract(self, url): @@ -139,6 +172,7 @@ class LBRYIE(LBRYBaseIE): display_id = display_id.split('/', 2)[-1].replace('/', ':') else: display_id = display_id.replace(':', '#') + display_id = compat_urllib_parse_unquote(display_id) uri = 'lbry://' + display_id result = self._resolve_url(uri, display_id, 'stream') result_value = result['value'] @@ -149,10 +183,18 @@ class LBRYIE(LBRYBaseIE): streaming_url = self._call_api_proxy( 'get', claim_id, {'uri': uri}, 'streaming url')['streaming_url'] info = self._parse_stream(result, url) + urlh = self._request_webpage( + streaming_url, display_id, note='Downloading streaming redirect url info') + if determine_ext(urlh.geturl()) == 'm3u8': + info['formats'] = self._extract_m3u8_formats( + urlh.geturl(), display_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls') + self._sort_formats(info['formats']) + else: + info['url'] = streaming_url info.update({ 'id': claim_id, 'title': title, - 'url': streaming_url, }) return info @@ -174,17 +216,18 @@ class LBRYChannelIE(LBRYBaseIE): }] _PAGE_SIZE = 50 - def _fetch_page(self, claim_id, url, page): + def _fetch_page(self, claim_id, url, params, page): page += 1 + page_params = { + 'channel_ids': [claim_id], + 'claim_type': 'stream', + 'no_totals': True, + 'page': page, + 'page_size': self._PAGE_SIZE, + } + page_params.update(params) result = self._call_api_proxy( - 'claim_search', claim_id, { - 'channel_ids': [claim_id], - 'claim_type': 'stream', - 'no_totals': True, - 'page': page, - 'page_size': self._PAGE_SIZE, - 'stream_types': self._SUPPORTED_STREAM_TYPES, - }, 'page %d' % page) + 'claim_search', claim_id, page_params, 'page %d' % page) for item in (result.get('items') or []): stream_claim_name = item.get('name') stream_claim_id = item.get('claim_id') @@ -205,8 +248,31 @@ class LBRYChannelIE(LBRYBaseIE): result = self._resolve_url( 'lbry://' + display_id, display_id, 'channel') claim_id = result['claim_id'] + qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query) + content = qs.get('content', [None])[0] + params = { + 'fee_amount': qs.get('fee_amount', ['>=0'])[0], + 'order_by': { + 'new': ['release_time'], + 'top': ['effective_amount'], + 'trending': ['trending_group', 'trending_mixed'], + }[qs.get('order', ['new'])[0]], + 'stream_types': [content] if content in ['audio', 'video'] else self._SUPPORTED_STREAM_TYPES, + } + duration = qs.get('duration', [None])[0] + if duration: + params['duration'] = { + 'long': '>=1200', + 'short': '<=240', + }[duration] + language = qs.get('language', ['all'])[0] + if language != 'all': + languages = [language] + if language == 'en': + languages.append('none') + params['any_languages'] = languages entries = OnDemandPagedList( - functools.partial(self._fetch_page, claim_id, url), + functools.partial(self._fetch_page, claim_id, url, params), self._PAGE_SIZE) result_value = result.get('value') or {} return self.playlist_result( diff --git a/youtube_dl/extractor/line.py b/youtube_dl/extractor/line.py index 7f5fa446e..2526daa77 100644 --- a/youtube_dl/extractor/line.py +++ b/youtube_dl/extractor/line.py @@ -4,7 +4,13 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import js_to_json +from ..compat import compat_str +from ..utils import ( + ExtractorError, + int_or_none, + js_to_json, + str_or_none, +) class LineTVIE(InfoExtractor): @@ -88,3 +94,137 @@ class LineTVIE(InfoExtractor): for thumbnail in video_info.get('thumbnails', {}).get('list', [])], 'view_count': video_info.get('meta', {}).get('count'), } + + +class LineLiveBaseIE(InfoExtractor): + _API_BASE_URL = 'https://live-api.line-apps.com/web/v4.0/channel/' + + def _parse_broadcast_item(self, item): + broadcast_id = compat_str(item['id']) + title = item['title'] + is_live = item.get('isBroadcastingNow') + + thumbnails = [] + for thumbnail_id, thumbnail_url in (item.get('thumbnailURLs') or {}).items(): + if not thumbnail_url: + continue + thumbnails.append({ + 'id': thumbnail_id, + 'url': thumbnail_url, + }) + + channel = item.get('channel') or {} + channel_id = str_or_none(channel.get('id')) + + return { + 'id': broadcast_id, + 'title': self._live_title(title) if is_live else title, + 'thumbnails': thumbnails, + 'timestamp': int_or_none(item.get('createdAt')), + 'channel': channel.get('name'), + 'channel_id': channel_id, + 'channel_url': 'https://live.line.me/channels/' + channel_id if channel_id else None, + 'duration': int_or_none(item.get('archiveDuration')), + 'view_count': int_or_none(item.get('viewerCount')), + 'comment_count': int_or_none(item.get('chatCount')), + 'is_live': is_live, + } + + +class LineLiveIE(LineLiveBaseIE): + _VALID_URL = r'https?://live\.line\.me/channels/(?P<channel_id>\d+)/broadcast/(?P<id>\d+)' + _TESTS = [{ + 'url': 'https://live.line.me/channels/4867368/broadcast/16331360', + 'md5': 'bc931f26bf1d4f971e3b0982b3fab4a3', + 'info_dict': { + 'id': '16331360', + 'title': '振りコピ講座😙😙😙', + 'ext': 'mp4', + 'timestamp': 1617095132, + 'upload_date': '20210330', + 'channel': '白川ゆめか', + 'channel_id': '4867368', + 'view_count': int, + 'comment_count': int, + 'is_live': False, + } + }, { + # archiveStatus == 'DELETED' + 'url': 'https://live.line.me/channels/4778159/broadcast/16378488', + 'only_matching': True, + }] + + def _real_extract(self, url): + channel_id, broadcast_id = re.match(self._VALID_URL, url).groups() + broadcast = self._download_json( + self._API_BASE_URL + '%s/broadcast/%s' % (channel_id, broadcast_id), + broadcast_id) + item = broadcast['item'] + info = self._parse_broadcast_item(item) + protocol = 'm3u8' if info['is_live'] else 'm3u8_native' + formats = [] + for k, v in (broadcast.get(('live' if info['is_live'] else 'archived') + 'HLSURLs') or {}).items(): + if not v: + continue + if k == 'abr': + formats.extend(self._extract_m3u8_formats( + v, broadcast_id, 'mp4', protocol, + m3u8_id='hls', fatal=False)) + continue + f = { + 'ext': 'mp4', + 'format_id': 'hls-' + k, + 'protocol': protocol, + 'url': v, + } + if not k.isdigit(): + f['vcodec'] = 'none' + formats.append(f) + if not formats: + archive_status = item.get('archiveStatus') + if archive_status != 'ARCHIVED': + raise ExtractorError('this video has been ' + archive_status.lower(), expected=True) + self._sort_formats(formats) + info['formats'] = formats + return info + + +class LineLiveChannelIE(LineLiveBaseIE): + _VALID_URL = r'https?://live\.line\.me/channels/(?P<id>\d+)(?!/broadcast/\d+)(?:[/?&#]|$)' + _TEST = { + 'url': 'https://live.line.me/channels/5893542', + 'info_dict': { + 'id': '5893542', + 'title': 'いくらちゃん', + 'description': 'md5:c3a4af801f43b2fac0b02294976580be', + }, + 'playlist_mincount': 29 + } + + def _archived_broadcasts_entries(self, archived_broadcasts, channel_id): + while True: + for row in (archived_broadcasts.get('rows') or []): + share_url = str_or_none(row.get('shareURL')) + if not share_url: + continue + info = self._parse_broadcast_item(row) + info.update({ + '_type': 'url', + 'url': share_url, + 'ie_key': LineLiveIE.ie_key(), + }) + yield info + if not archived_broadcasts.get('hasNextPage'): + return + archived_broadcasts = self._download_json( + self._API_BASE_URL + channel_id + '/archived_broadcasts', + channel_id, query={ + 'lastId': info['id'], + }) + + def _real_extract(self, url): + channel_id = self._match_id(url) + channel = self._download_json(self._API_BASE_URL + channel_id, channel_id) + return self.playlist_result( + self._archived_broadcasts_entries(channel.get('archivedBroadcasts') or {}, channel_id), + channel_id, channel.get('title'), channel.get('information')) diff --git a/youtube_dl/extractor/liveleak.py b/youtube_dl/extractor/liveleak.py deleted file mode 100644 index 4ac437c8b..000000000 --- a/youtube_dl/extractor/liveleak.py +++ /dev/null @@ -1,191 +0,0 @@ -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..utils import int_or_none - - -class LiveLeakIE(InfoExtractor): - _VALID_URL = r'https?://(?:\w+\.)?liveleak\.com/view\?.*?\b[it]=(?P<id>[\w_]+)' - _TESTS = [{ - 'url': 'http://www.liveleak.com/view?i=757_1364311680', - 'md5': '0813c2430bea7a46bf13acf3406992f4', - 'info_dict': { - 'id': '757_1364311680', - 'ext': 'mp4', - 'description': 'extremely bad day for this guy..!', - 'uploader': 'ljfriel2', - 'title': 'Most unlucky car accident', - 'thumbnail': r're:^https?://.*\.jpg$' - } - }, { - 'url': 'http://www.liveleak.com/view?i=f93_1390833151', - 'md5': 'd3f1367d14cc3c15bf24fbfbe04b9abf', - 'info_dict': { - 'id': 'f93_1390833151', - 'ext': 'mp4', - 'description': 'German Television Channel NDR does an exclusive interview with Edward Snowden.\r\nUploaded on LiveLeak cause German Television thinks the rest of the world isn\'t intereseted in Edward Snowden.', - 'uploader': 'ARD_Stinkt', - 'title': 'German Television does first Edward Snowden Interview (ENGLISH)', - 'thumbnail': r're:^https?://.*\.jpg$' - } - }, { - # Prochan embed - 'url': 'http://www.liveleak.com/view?i=4f7_1392687779', - 'md5': '42c6d97d54f1db107958760788c5f48f', - 'info_dict': { - 'id': '4f7_1392687779', - 'ext': 'mp4', - 'description': "The guy with the cigarette seems amazingly nonchalant about the whole thing... I really hope my friends' reactions would be a bit stronger.\r\n\r\nAction-go to 0:55.", - 'uploader': 'CapObveus', - 'title': 'Man is Fatally Struck by Reckless Car While Packing up a Moving Truck', - 'age_limit': 18, - }, - 'skip': 'Video is dead', - }, { - # Covers https://github.com/ytdl-org/youtube-dl/pull/5983 - # Multiple resolutions - 'url': 'http://www.liveleak.com/view?i=801_1409392012', - 'md5': 'c3a449dbaca5c0d1825caecd52a57d7b', - 'info_dict': { - 'id': '801_1409392012', - 'ext': 'mp4', - 'description': 'Happened on 27.7.2014. \r\nAt 0:53 you can see people still swimming at near beach.', - 'uploader': 'bony333', - 'title': 'Crazy Hungarian tourist films close call waterspout in Croatia', - 'thumbnail': r're:^https?://.*\.jpg$' - } - }, { - # Covers https://github.com/ytdl-org/youtube-dl/pull/10664#issuecomment-247439521 - 'url': 'http://m.liveleak.com/view?i=763_1473349649', - 'add_ie': ['Youtube'], - 'info_dict': { - 'id': '763_1473349649', - 'ext': 'mp4', - 'title': 'Reporters and public officials ignore epidemic of black on asian violence in Sacramento | Colin Flaherty', - 'description': 'Colin being the warrior he is and showing the injustice Asians in Sacramento are being subjected to.', - 'uploader': 'Ziz', - 'upload_date': '20160908', - 'uploader_id': 'UCEbta5E_jqlZmEJsriTEtnw' - }, - 'params': { - 'skip_download': True, - }, - }, { - 'url': 'https://www.liveleak.com/view?i=677_1439397581', - 'info_dict': { - 'id': '677_1439397581', - 'title': 'Fuel Depot in China Explosion caught on video', - }, - 'playlist_count': 3, - }, { - 'url': 'https://www.liveleak.com/view?t=HvHi_1523016227', - 'only_matching': True, - }, { - # No original video - 'url': 'https://www.liveleak.com/view?t=C26ZZ_1558612804', - 'only_matching': True, - }] - - @staticmethod - def _extract_urls(webpage): - return re.findall( - r'<iframe[^>]+src="(https?://(?:\w+\.)?liveleak\.com/ll_embed\?[^"]*[ift]=[\w_]+[^"]+)"', - webpage) - - def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - - video_title = self._og_search_title(webpage).replace('LiveLeak.com -', '').strip() - video_description = self._og_search_description(webpage) - video_uploader = self._html_search_regex( - r'By:.*?(\w+)</a>', webpage, 'uploader', fatal=False) - age_limit = int_or_none(self._search_regex( - r'you confirm that you are ([0-9]+) years and over.', - webpage, 'age limit', default=None)) - video_thumbnail = self._og_search_thumbnail(webpage) - - entries = self._parse_html5_media_entries(url, webpage, video_id) - if not entries: - # Maybe an embed? - embed_url = self._search_regex( - r'<iframe[^>]+src="((?:https?:)?//(?:www\.)?(?:prochan|youtube)\.com/embed[^"]+)"', - webpage, 'embed URL') - return { - '_type': 'url_transparent', - 'url': embed_url, - 'id': video_id, - 'title': video_title, - 'description': video_description, - 'uploader': video_uploader, - 'age_limit': age_limit, - } - - for idx, info_dict in enumerate(entries): - formats = [] - for a_format in info_dict['formats']: - if not a_format.get('height'): - a_format['height'] = int_or_none(self._search_regex( - r'([0-9]+)p\.mp4', a_format['url'], 'height label', - default=None)) - formats.append(a_format) - - # Removing '.*.mp4' gives the raw video, which is essentially - # the same video without the LiveLeak logo at the top (see - # https://github.com/ytdl-org/youtube-dl/pull/4768) - orig_url = re.sub(r'\.mp4\.[^.]+', '', a_format['url']) - if a_format['url'] != orig_url: - format_id = a_format.get('format_id') - format_id = 'original' + ('-' + format_id if format_id else '') - if self._is_valid_url(orig_url, video_id, format_id): - formats.append({ - 'format_id': format_id, - 'url': orig_url, - 'preference': 1, - }) - self._sort_formats(formats) - info_dict['formats'] = formats - - # Don't append entry ID for one-video pages to keep backward compatibility - if len(entries) > 1: - info_dict['id'] = '%s_%s' % (video_id, idx + 1) - else: - info_dict['id'] = video_id - - info_dict.update({ - 'title': video_title, - 'description': video_description, - 'uploader': video_uploader, - 'age_limit': age_limit, - 'thumbnail': video_thumbnail, - }) - - return self.playlist_result(entries, video_id, video_title) - - -class LiveLeakEmbedIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?liveleak\.com/ll_embed\?.*?\b(?P<kind>[ift])=(?P<id>[\w_]+)' - - # See generic.py for actual test cases - _TESTS = [{ - 'url': 'https://www.liveleak.com/ll_embed?i=874_1459135191', - 'only_matching': True, - }, { - 'url': 'https://www.liveleak.com/ll_embed?f=ab065df993c1', - 'only_matching': True, - }] - - def _real_extract(self, url): - kind, video_id = re.match(self._VALID_URL, url).groups() - - if kind == 'f': - webpage = self._download_webpage(url, video_id) - liveleak_url = self._search_regex( - r'(?:logourl\s*:\s*|window\.open\()(?P<q1>[\'"])(?P<url>%s)(?P=q1)' % LiveLeakIE._VALID_URL, - webpage, 'LiveLeak URL', group='url') - else: - liveleak_url = 'http://www.liveleak.com/view?%s=%s' % (kind, video_id) - - return self.url_result(liveleak_url, ie=LiveLeakIE.ie_key()) diff --git a/youtube_dl/extractor/maoritv.py b/youtube_dl/extractor/maoritv.py new file mode 100644 index 000000000..0d23fec75 --- /dev/null +++ b/youtube_dl/extractor/maoritv.py @@ -0,0 +1,31 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor + + +class MaoriTVIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?maoritelevision\.com/shows/(?:[^/]+/)+(?P<id>[^/?&#]+)' + _TEST = { + 'url': 'https://www.maoritelevision.com/shows/korero-mai/S01E054/korero-mai-series-1-episode-54', + 'md5': '5ade8ef53851b6a132c051b1cd858899', + 'info_dict': { + 'id': '4774724855001', + 'ext': 'mp4', + 'title': 'Kōrero Mai, Series 1 Episode 54', + 'upload_date': '20160226', + 'timestamp': 1456455018, + 'description': 'md5:59bde32fd066d637a1a55794c56d8dcb', + 'uploader_id': '1614493167001', + }, + } + BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/1614493167001/HJlhIQhQf_default/index.html?videoId=%s' + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + brightcove_id = self._search_regex( + r'data-main-video-id=["\'](\d+)', webpage, 'brightcove id') + return self.url_result( + self.BRIGHTCOVE_URL_TEMPLATE % brightcove_id, + 'BrightcoveNew', brightcove_id) diff --git a/youtube_dl/extractor/medaltv.py b/youtube_dl/extractor/medaltv.py index 1603b55f6..67bb4debb 100644 --- a/youtube_dl/extractor/medaltv.py +++ b/youtube_dl/extractor/medaltv.py @@ -15,33 +15,39 @@ from ..utils import ( class MedalTVIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?medal\.tv/clips/(?P<id>[0-9]+)' + _VALID_URL = r'https?://(?:www\.)?medal\.tv/clips/(?P<id>[^/?#&]+)' _TESTS = [{ - 'url': 'https://medal.tv/clips/34934644/3Is9zyGMoBMr', + 'url': 'https://medal.tv/clips/2mA60jWAGQCBH', 'md5': '7b07b064331b1cf9e8e5c52a06ae68fa', 'info_dict': { - 'id': '34934644', + 'id': '2mA60jWAGQCBH', 'ext': 'mp4', 'title': 'Quad Cold', 'description': 'Medal,https://medal.tv/desktop/', 'uploader': 'MowgliSB', 'timestamp': 1603165266, 'upload_date': '20201020', - 'uploader_id': 10619174, + 'uploader_id': '10619174', } }, { - 'url': 'https://medal.tv/clips/36787208', + 'url': 'https://medal.tv/clips/2um24TWdty0NA', 'md5': 'b6dc76b78195fff0b4f8bf4a33ec2148', 'info_dict': { - 'id': '36787208', + 'id': '2um24TWdty0NA', 'ext': 'mp4', 'title': 'u tk me i tk u bigger', 'description': 'Medal,https://medal.tv/desktop/', 'uploader': 'Mimicc', 'timestamp': 1605580939, 'upload_date': '20201117', - 'uploader_id': 5156321, + 'uploader_id': '5156321', } + }, { + 'url': 'https://medal.tv/clips/37rMeFpryCC-9', + 'only_matching': True, + }, { + 'url': 'https://medal.tv/clips/2WRj40tpY_EU9', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/medialaan.py b/youtube_dl/extractor/medialaan.py index 50d5db802..788acf7fb 100644 --- a/youtube_dl/extractor/medialaan.py +++ b/youtube_dl/extractor/medialaan.py @@ -2,268 +2,113 @@ from __future__ import unicode_literals import re -from .gigya import GigyaBaseIE - -from ..compat import compat_str +from .common import InfoExtractor from ..utils import ( + extract_attributes, int_or_none, - parse_duration, - try_get, - unified_timestamp, + mimetype2ext, + parse_iso8601, ) -class MedialaanIE(GigyaBaseIE): +class MedialaanIE(InfoExtractor): _VALID_URL = r'''(?x) https?:// - (?:www\.|nieuws\.)? (?: - (?P<site_id>vtm|q2|vtmkzoom)\.be/ - (?: - video(?:/[^/]+/id/|/?\?.*?\baid=)| - (?:[^/]+/)* - ) + (?:embed\.)?mychannels.video/embed/| + embed\.mychannels\.video/(?:s(?:dk|cript)/)?production/| + (?:www\.)?(?: + (?: + 7sur7| + demorgen| + hln| + joe| + qmusic + )\.be| + (?: + [abe]d| + bndestem| + destentor| + gelderlander| + pzc| + tubantia| + volkskrant + )\.nl + )/video/(?:[^/]+/)*[^/?&#]+~p ) - (?P<id>[^/?#&]+) + (?P<id>\d+) ''' - _NETRC_MACHINE = 'medialaan' - _APIKEY = '3_HZ0FtkMW_gOyKlqQzW5_0FHRC7Nd5XpXJZcDdXY4pk5eES2ZWmejRW5egwVm4ug-' - _SITE_TO_APP_ID = { - 'vtm': 'vtm_watch', - 'q2': 'q2', - 'vtmkzoom': 'vtmkzoom', - } _TESTS = [{ - # vod - 'url': 'http://vtm.be/video/volledige-afleveringen/id/vtm_20170219_VM0678361_vtmwatch', + 'url': 'https://www.bndestem.nl/video/de-terugkeer-van-ally-de-aap-en-wie-vertrekt-er-nog-bij-nac~p193993', 'info_dict': { - 'id': 'vtm_20170219_VM0678361_vtmwatch', + 'id': '193993', 'ext': 'mp4', - 'title': 'Allemaal Chris afl. 6', - 'description': 'md5:4be86427521e7b07e0adb0c9c554ddb2', - 'timestamp': 1487533280, - 'upload_date': '20170219', - 'duration': 2562, - 'series': 'Allemaal Chris', - 'season': 'Allemaal Chris', - 'season_number': 1, - 'season_id': '256936078124527', - 'episode': 'Allemaal Chris afl. 6', - 'episode_number': 6, - 'episode_id': '256936078591527', + 'title': 'De terugkeer van Ally de Aap en wie vertrekt er nog bij NAC?', + 'timestamp': 1611663540, + 'upload_date': '20210126', + 'duration': 238, }, 'params': { 'skip_download': True, }, - 'skip': 'Requires account credentials', - }, { - # clip - 'url': 'http://vtm.be/video?aid=168332', - 'info_dict': { - 'id': '168332', - 'ext': 'mp4', - 'title': '"Veronique liegt!"', - 'description': 'md5:1385e2b743923afe54ba4adc38476155', - 'timestamp': 1489002029, - 'upload_date': '20170308', - 'duration': 96, - }, }, { - # vod - 'url': 'http://vtm.be/video/volledige-afleveringen/id/257107153551000', + 'url': 'https://www.gelderlander.nl/video/kanalen/degelderlander~c320/series/snel-nieuws~s984/noodbevel-in-doetinchem-politie-stuurt-mensen-centrum-uit~p194093', 'only_matching': True, }, { - # vod - 'url': 'http://vtm.be/video?aid=163157', + 'url': 'https://embed.mychannels.video/sdk/production/193993?options=TFTFF_default', 'only_matching': True, }, { - # vod - 'url': 'http://www.q2.be/video/volledige-afleveringen/id/2be_20170301_VM0684442_q2', + 'url': 'https://embed.mychannels.video/script/production/193993', 'only_matching': True, }, { - # clip - 'url': 'http://vtmkzoom.be/k3-dansstudio/een-nieuw-seizoen-van-k3-dansstudio', + 'url': 'https://embed.mychannels.video/production/193993', 'only_matching': True, }, { - # http/s redirect - 'url': 'https://vtmkzoom.be/video?aid=45724', - 'info_dict': { - 'id': '257136373657000', - 'ext': 'mp4', - 'title': 'K3 Dansstudio Ushuaia afl.6', - }, - 'params': { - 'skip_download': True, - }, - 'skip': 'Requires account credentials', + 'url': 'https://mychannels.video/embed/193993', + 'only_matching': True, }, { - # nieuws.vtm.be - 'url': 'https://nieuws.vtm.be/stadion/stadion/genk-nog-moeilijk-programma', + 'url': 'https://embed.mychannels.video/embed/193993', 'only_matching': True, }] - def _real_initialize(self): - self._logged_in = False - - def _login(self): - username, password = self._get_login_info() - if username is None: - self.raise_login_required() - - auth_data = { - 'APIKey': self._APIKEY, - 'sdk': 'js_6.1', - 'format': 'json', - 'loginID': username, - 'password': password, - } - - auth_info = self._gigya_login(auth_data) - - self._uid = auth_info['UID'] - self._uid_signature = auth_info['UIDSignature'] - self._signature_timestamp = auth_info['signatureTimestamp'] - - self._logged_in = True + @staticmethod + def _extract_urls(webpage): + entries = [] + for element in re.findall(r'(<div[^>]+data-mychannels-type="video"[^>]*>)', webpage): + mychannels_id = extract_attributes(element).get('data-mychannels-id') + if mychannels_id: + entries.append('https://mychannels.video/embed/' + mychannels_id) + return entries def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id, site_id = mobj.group('id', 'site_id') - - webpage = self._download_webpage(url, video_id) - - config = self._parse_json( - self._search_regex( - r'videoJSConfig\s*=\s*JSON\.parse\(\'({.+?})\'\);', - webpage, 'config', default='{}'), video_id, - transform_source=lambda s: s.replace( - '\\\\', '\\').replace(r'\"', '"').replace(r"\'", "'")) - - vod_id = config.get('vodId') or self._search_regex( - (r'\\"vodId\\"\s*:\s*\\"(.+?)\\"', - r'"vodId"\s*:\s*"(.+?)"', - r'<[^>]+id=["\']vod-(\d+)'), - webpage, 'video_id', default=None) - - # clip, no authentication required - if not vod_id: - player = self._parse_json( - self._search_regex( - r'vmmaplayer\(({.+?})\);', webpage, 'vmma player', - default=''), - video_id, transform_source=lambda s: '[%s]' % s, fatal=False) - if player: - video = player[-1] - if video['videoUrl'] in ('http', 'https'): - return self.url_result(video['url'], MedialaanIE.ie_key()) - info = { - 'id': video_id, - 'url': video['videoUrl'], - 'title': video['title'], - 'thumbnail': video.get('imageUrl'), - 'timestamp': int_or_none(video.get('createdDate')), - 'duration': int_or_none(video.get('duration')), - } + production_id = self._match_id(url) + production = self._download_json( + 'https://embed.mychannels.video/sdk/production/' + production_id, + production_id, query={'options': 'UUUU_default'})['productions'][0] + title = production['title'] + + formats = [] + for source in (production.get('sources') or []): + src = source.get('src') + if not src: + continue + ext = mimetype2ext(source.get('type')) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + src, production_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) else: - info = self._parse_html5_media_entries( - url, webpage, video_id, m3u8_id='hls')[0] - info.update({ - 'id': video_id, - 'title': self._html_search_meta('description', webpage), - 'duration': parse_duration(self._html_search_meta('duration', webpage)), - }) - # vod, authentication required - else: - if not self._logged_in: - self._login() - - settings = self._parse_json( - self._search_regex( - r'jQuery\.extend\(Drupal\.settings\s*,\s*({.+?})\);', - webpage, 'drupal settings', default='{}'), - video_id) - - def get(container, item): - return try_get( - settings, lambda x: x[container][item], - compat_str) or self._search_regex( - r'"%s"\s*:\s*"([^"]+)' % item, webpage, item, - default=None) - - app_id = get('vod', 'app_id') or self._SITE_TO_APP_ID.get(site_id, 'vtm_watch') - sso = get('vod', 'gigyaDatabase') or 'vtm-sso' - - data = self._download_json( - 'http://vod.medialaan.io/api/1.0/item/%s/video' % vod_id, - video_id, query={ - 'app_id': app_id, - 'user_network': sso, - 'UID': self._uid, - 'UIDSignature': self._uid_signature, - 'signatureTimestamp': self._signature_timestamp, + formats.append({ + 'ext': ext, + 'url': src, }) - - formats = self._extract_m3u8_formats( - data['response']['uri'], video_id, entry_protocol='m3u8_native', - ext='mp4', m3u8_id='hls') - - self._sort_formats(formats) - - info = { - 'id': vod_id, - 'formats': formats, - } - - api_key = get('vod', 'apiKey') - channel = get('medialaanGigya', 'channel') - - if api_key: - videos = self._download_json( - 'http://vod.medialaan.io/vod/v2/videos', video_id, fatal=False, - query={ - 'channels': channel, - 'ids': vod_id, - 'limit': 1, - 'apikey': api_key, - }) - if videos: - video = try_get( - videos, lambda x: x['response']['videos'][0], dict) - if video: - def get(container, item, expected_type=None): - return try_get( - video, lambda x: x[container][item], expected_type) - - def get_string(container, item): - return get(container, item, compat_str) - - info.update({ - 'series': get_string('program', 'title'), - 'season': get_string('season', 'title'), - 'season_number': int_or_none(get('season', 'number')), - 'season_id': get_string('season', 'id'), - 'episode': get_string('episode', 'title'), - 'episode_number': int_or_none(get('episode', 'number')), - 'episode_id': get_string('episode', 'id'), - 'duration': int_or_none( - video.get('duration')) or int_or_none( - video.get('durationMillis'), scale=1000), - 'title': get_string('episode', 'title'), - 'description': get_string('episode', 'text'), - 'timestamp': unified_timestamp(get_string( - 'publication', 'begin')), - }) - - if not info.get('title'): - info['title'] = try_get( - config, lambda x: x['videoConfig']['title'], - compat_str) or self._html_search_regex( - r'\\"title\\"\s*:\s*\\"(.+?)\\"', webpage, 'title', - default=None) or self._og_search_title(webpage) - - if not info.get('description'): - info['description'] = self._html_search_regex( - r'<div[^>]+class="field-item\s+even">\s*<p>(.+?)</p>', - webpage, 'description', default=None) - - return info + self._sort_formats(formats) + + return { + 'id': production_id, + 'title': title, + 'formats': formats, + 'thumbnail': production.get('posterUrl'), + 'timestamp': parse_iso8601(production.get('publicationDate'), ' '), + 'duration': int_or_none(production.get('duration')) or None, + } diff --git a/youtube_dl/extractor/minds.py b/youtube_dl/extractor/minds.py new file mode 100644 index 000000000..8e9f0f825 --- /dev/null +++ b/youtube_dl/extractor/minds.py @@ -0,0 +1,196 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + clean_html, + int_or_none, + str_or_none, + strip_or_none, +) + + +class MindsBaseIE(InfoExtractor): + _VALID_URL_BASE = r'https?://(?:www\.)?minds\.com/' + + def _call_api(self, path, video_id, resource, query=None): + api_url = 'https://www.minds.com/api/' + path + token = self._get_cookies(api_url).get('XSRF-TOKEN') + return self._download_json( + api_url, video_id, 'Downloading %s JSON metadata' % resource, headers={ + 'Referer': 'https://www.minds.com/', + 'X-XSRF-TOKEN': token.value if token else '', + }, query=query) + + +class MindsIE(MindsBaseIE): + IE_NAME = 'minds' + _VALID_URL = MindsBaseIE._VALID_URL_BASE + r'(?:media|newsfeed|archive/view)/(?P<id>[0-9]+)' + _TESTS = [{ + 'url': 'https://www.minds.com/media/100000000000086822', + 'md5': '215a658184a419764852239d4970b045', + 'info_dict': { + 'id': '100000000000086822', + 'ext': 'mp4', + 'title': 'Minds intro sequence', + 'thumbnail': r're:https?://.+\.png', + 'uploader_id': 'ottman', + 'upload_date': '20130524', + 'timestamp': 1369404826, + 'uploader': 'Bill Ottman', + 'view_count': int, + 'like_count': int, + 'dislike_count': int, + 'tags': ['animation'], + 'comment_count': int, + 'license': 'attribution-cc', + }, + }, { + # entity.type == 'activity' and empty title + 'url': 'https://www.minds.com/newsfeed/798025111988506624', + 'md5': 'b2733a74af78d7fd3f541c4cbbaa5950', + 'info_dict': { + 'id': '798022190320226304', + 'ext': 'mp4', + 'title': '798022190320226304', + 'uploader': 'ColinFlaherty', + 'upload_date': '20180111', + 'timestamp': 1515639316, + 'uploader_id': 'ColinFlaherty', + }, + }, { + 'url': 'https://www.minds.com/archive/view/715172106794442752', + 'only_matching': True, + }, { + # youtube perma_url + 'url': 'https://www.minds.com/newsfeed/1197131838022602752', + 'only_matching': True, + }] + + def _real_extract(self, url): + entity_id = self._match_id(url) + entity = self._call_api( + 'v1/entities/entity/' + entity_id, entity_id, 'entity')['entity'] + if entity.get('type') == 'activity': + if entity.get('custom_type') == 'video': + video_id = entity['entity_guid'] + else: + return self.url_result(entity['perma_url']) + else: + assert(entity['subtype'] == 'video') + video_id = entity_id + # 1080p and webm formats available only on the sources array + video = self._call_api( + 'v2/media/video/' + video_id, video_id, 'video') + + formats = [] + for source in (video.get('sources') or []): + src = source.get('src') + if not src: + continue + formats.append({ + 'format_id': source.get('label'), + 'height': int_or_none(source.get('size')), + 'url': src, + }) + self._sort_formats(formats) + + entity = video.get('entity') or entity + owner = entity.get('ownerObj') or {} + uploader_id = owner.get('username') + + tags = entity.get('tags') + if tags and isinstance(tags, compat_str): + tags = [tags] + + thumbnail = None + poster = video.get('poster') or entity.get('thumbnail_src') + if poster: + urlh = self._request_webpage(poster, video_id, fatal=False) + if urlh: + thumbnail = urlh.geturl() + + return { + 'id': video_id, + 'title': entity.get('title') or video_id, + 'formats': formats, + 'description': clean_html(entity.get('description')) or None, + 'license': str_or_none(entity.get('license')), + 'timestamp': int_or_none(entity.get('time_created')), + 'uploader': strip_or_none(owner.get('name')), + 'uploader_id': uploader_id, + 'uploader_url': 'https://www.minds.com/' + uploader_id if uploader_id else None, + 'view_count': int_or_none(entity.get('play:count')), + 'like_count': int_or_none(entity.get('thumbs:up:count')), + 'dislike_count': int_or_none(entity.get('thumbs:down:count')), + 'tags': tags, + 'comment_count': int_or_none(entity.get('comments:count')), + 'thumbnail': thumbnail, + } + + +class MindsFeedBaseIE(MindsBaseIE): + _PAGE_SIZE = 150 + + def _entries(self, feed_id): + query = {'limit': self._PAGE_SIZE, 'sync': 1} + i = 1 + while True: + data = self._call_api( + 'v2/feeds/container/%s/videos' % feed_id, + feed_id, 'page %s' % i, query) + entities = data.get('entities') or [] + for entity in entities: + guid = entity.get('guid') + if not guid: + continue + yield self.url_result( + 'https://www.minds.com/newsfeed/' + guid, + MindsIE.ie_key(), guid) + query['from_timestamp'] = data['load-next'] + if not (query['from_timestamp'] and len(entities) == self._PAGE_SIZE): + break + i += 1 + + def _real_extract(self, url): + feed_id = self._match_id(url) + feed = self._call_api( + 'v1/%s/%s' % (self._FEED_PATH, feed_id), + feed_id, self._FEED_TYPE)[self._FEED_TYPE] + + return self.playlist_result( + self._entries(feed['guid']), feed_id, + strip_or_none(feed.get('name')), + feed.get('briefdescription')) + + +class MindsChannelIE(MindsFeedBaseIE): + _FEED_TYPE = 'channel' + IE_NAME = 'minds:' + _FEED_TYPE + _VALID_URL = MindsBaseIE._VALID_URL_BASE + r'(?!(?:newsfeed|media|api|archive|groups)/)(?P<id>[^/?&#]+)' + _FEED_PATH = 'channel' + _TEST = { + 'url': 'https://www.minds.com/ottman', + 'info_dict': { + 'id': 'ottman', + 'title': 'Bill Ottman', + 'description': 'Co-creator & CEO @minds', + }, + 'playlist_mincount': 54, + } + + +class MindsGroupIE(MindsFeedBaseIE): + _FEED_TYPE = 'group' + IE_NAME = 'minds:' + _FEED_TYPE + _VALID_URL = MindsBaseIE._VALID_URL_BASE + r'groups/profile/(?P<id>[0-9]+)' + _FEED_PATH = 'groups/group' + _TEST = { + 'url': 'https://www.minds.com/groups/profile/785582576369672204/feed/videos', + 'info_dict': { + 'id': '785582576369672204', + 'title': 'Cooking Videos', + }, + 'playlist_mincount': 1, + } diff --git a/youtube_dl/extractor/mixcloud.py b/youtube_dl/extractor/mixcloud.py index 9759560f1..69319857d 100644 --- a/youtube_dl/extractor/mixcloud.py +++ b/youtube_dl/extractor/mixcloud.py @@ -251,8 +251,11 @@ class MixcloudPlaylistBaseIE(MixcloudBaseIE): cloudcast_url = cloudcast.get('url') if not cloudcast_url: continue + slug = try_get(cloudcast, lambda x: x['slug'], compat_str) + owner_username = try_get(cloudcast, lambda x: x['owner']['username'], compat_str) + video_id = '%s_%s' % (owner_username, slug) if slug and owner_username else None entries.append(self.url_result( - cloudcast_url, MixcloudIE.ie_key(), cloudcast.get('slug'))) + cloudcast_url, MixcloudIE.ie_key(), video_id)) page_info = items['pageInfo'] has_next_page = page_info['hasNextPage'] @@ -321,7 +324,8 @@ class MixcloudUserIE(MixcloudPlaylistBaseIE): _DESCRIPTION_KEY = 'biog' _ROOT_TYPE = 'user' _NODE_TEMPLATE = '''slug - url''' + url + owner { username }''' def _get_playlist_title(self, title, slug): return '%s (%s)' % (title, slug) @@ -345,6 +349,7 @@ class MixcloudPlaylistIE(MixcloudPlaylistBaseIE): _NODE_TEMPLATE = '''cloudcast { slug url + owner { username } }''' def _get_cloudcast(self, node): diff --git a/youtube_dl/extractor/mlb.py b/youtube_dl/extractor/mlb.py index b907f6b49..b69301d97 100644 --- a/youtube_dl/extractor/mlb.py +++ b/youtube_dl/extractor/mlb.py @@ -1,15 +1,91 @@ from __future__ import unicode_literals -from .nhl import NHLBaseIE +import re +from .common import InfoExtractor +from ..utils import ( + determine_ext, + int_or_none, + parse_duration, + parse_iso8601, + try_get, +) -class MLBIE(NHLBaseIE): + +class MLBBaseIE(InfoExtractor): + def _real_extract(self, url): + display_id = self._match_id(url) + video = self._download_video_data(display_id) + video_id = video['id'] + title = video['title'] + feed = self._get_feed(video) + + formats = [] + for playback in (feed.get('playbacks') or []): + playback_url = playback.get('url') + if not playback_url: + continue + name = playback.get('name') + ext = determine_ext(playback_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + playback_url, video_id, 'mp4', + 'm3u8_native', m3u8_id=name, fatal=False)) + else: + f = { + 'format_id': name, + 'url': playback_url, + } + mobj = re.search(r'_(\d+)K_(\d+)X(\d+)', name) + if mobj: + f.update({ + 'height': int(mobj.group(3)), + 'tbr': int(mobj.group(1)), + 'width': int(mobj.group(2)), + }) + mobj = re.search(r'_(\d+)x(\d+)_(\d+)_(\d+)K\.mp4', playback_url) + if mobj: + f.update({ + 'fps': int(mobj.group(3)), + 'height': int(mobj.group(2)), + 'tbr': int(mobj.group(4)), + 'width': int(mobj.group(1)), + }) + formats.append(f) + self._sort_formats(formats) + + thumbnails = [] + for cut in (try_get(feed, lambda x: x['image']['cuts'], list) or []): + src = cut.get('src') + if not src: + continue + thumbnails.append({ + 'height': int_or_none(cut.get('height')), + 'url': src, + 'width': int_or_none(cut.get('width')), + }) + + language = (video.get('language') or 'EN').lower() + + return { + 'id': video_id, + 'title': title, + 'formats': formats, + 'description': video.get('description'), + 'duration': parse_duration(feed.get('duration')), + 'thumbnails': thumbnails, + 'timestamp': parse_iso8601(video.get(self._TIMESTAMP_KEY)), + 'subtitles': self._extract_mlb_subtitles(feed, language), + } + + +class MLBIE(MLBBaseIE): _VALID_URL = r'''(?x) https?:// - (?:[\da-z_-]+\.)*(?P<site>mlb)\.com/ + (?:[\da-z_-]+\.)*mlb\.com/ (?: (?: - (?:[^/]+/)*c-| + (?:[^/]+/)*video/[^/]+/c-| (?: shared/video/embed/(?:embed|m-internal-embed)\.html| (?:[^/]+/)+(?:play|index)\.jsp| @@ -18,7 +94,6 @@ class MLBIE(NHLBaseIE): (?P<id>\d+) ) ''' - _CONTENT_DOMAIN = 'content.mlb.com' _TESTS = [ { 'url': 'https://www.mlb.com/mariners/video/ackleys-spectacular-catch/c-34698933', @@ -77,18 +152,6 @@ class MLBIE(NHLBaseIE): }, }, { - 'url': 'https://www.mlb.com/news/blue-jays-kevin-pillar-goes-spidey-up-the-wall-to-rob-tim-beckham-of-a-homer/c-118550098', - 'md5': 'e09e37b552351fddbf4d9e699c924d68', - 'info_dict': { - 'id': '75609783', - 'ext': 'mp4', - 'title': 'Must C: Pillar climbs for catch', - 'description': '4/15/15: Blue Jays outfielder Kevin Pillar continues his defensive dominance by climbing the wall in left to rob Tim Beckham of a home run', - 'timestamp': 1429139220, - 'upload_date': '20150415', - } - }, - { 'url': 'https://www.mlb.com/video/hargrove-homers-off-caldwell/c-1352023483?tid=67793694', 'only_matching': True, }, @@ -113,8 +176,92 @@ class MLBIE(NHLBaseIE): 'url': 'http://mlb.mlb.com/shared/video/embed/m-internal-embed.html?content_id=75609783&property=mlb&autoplay=true&hashmode=false&siteSection=mlb/multimedia/article_118550098/article_embed&club=mlb', 'only_matching': True, }, - { - 'url': 'https://www.mlb.com/cut4/carlos-gomez-borrowed-sunglasses-from-an-as-fan/c-278912842', - 'only_matching': True, - } ] + _TIMESTAMP_KEY = 'date' + + @staticmethod + def _get_feed(video): + return video + + @staticmethod + def _extract_mlb_subtitles(feed, language): + subtitles = {} + for keyword in (feed.get('keywordsAll') or []): + keyword_type = keyword.get('type') + if keyword_type and keyword_type.startswith('closed_captions_location_'): + cc_location = keyword.get('value') + if cc_location: + subtitles.setdefault(language, []).append({ + 'url': cc_location, + }) + return subtitles + + def _download_video_data(self, display_id): + return self._download_json( + 'http://content.mlb.com/mlb/item/id/v1/%s/details/web-v1.json' % display_id, + display_id) + + +class MLBVideoIE(MLBBaseIE): + _VALID_URL = r'https?://(?:www\.)?mlb\.com/(?:[^/]+/)*video/(?P<id>[^/?&#]+)' + _TEST = { + 'url': 'https://www.mlb.com/mariners/video/ackley-s-spectacular-catch-c34698933', + 'md5': '632358dacfceec06bad823b83d21df2d', + 'info_dict': { + 'id': 'c04a8863-f569-42e6-9f87-992393657614', + 'ext': 'mp4', + 'title': "Ackley's spectacular catch", + 'description': 'md5:7f5a981eb4f3cbc8daf2aeffa2215bf0', + 'duration': 66, + 'timestamp': 1405995000, + 'upload_date': '20140722', + 'thumbnail': r're:^https?://.+', + }, + } + _TIMESTAMP_KEY = 'timestamp' + + @classmethod + def suitable(cls, url): + return False if MLBIE.suitable(url) else super(MLBVideoIE, cls).suitable(url) + + @staticmethod + def _get_feed(video): + return video['feeds'][0] + + @staticmethod + def _extract_mlb_subtitles(feed, language): + subtitles = {} + for cc_location in (feed.get('closedCaptions') or []): + subtitles.setdefault(language, []).append({ + 'url': cc_location, + }) + + def _download_video_data(self, display_id): + # https://www.mlb.com/data-service/en/videos/[SLUG] + return self._download_json( + 'https://fastball-gateway.mlb.com/graphql', + display_id, query={ + 'query': '''{ + mediaPlayback(ids: "%s") { + description + feeds(types: CMS) { + closedCaptions + duration + image { + cuts { + width + height + src + } + } + playbacks { + name + url + } + } + id + timestamp + title + } +}''' % display_id, + })['data']['mediaPlayback'][0] diff --git a/youtube_dl/extractor/mtv.py b/youtube_dl/extractor/mtv.py index df1034fc5..5a5205c0e 100644 --- a/youtube_dl/extractor/mtv.py +++ b/youtube_dl/extractor/mtv.py @@ -253,6 +253,12 @@ class MTVServicesInfoExtractor(InfoExtractor): return try_get(feed, lambda x: x['result']['data']['id'], compat_str) + @staticmethod + def _extract_child_with_type(parent, t): + for c in parent['children']: + if c.get('type') == t: + return c + def _extract_mgid(self, webpage): try: # the url can be http://media.mtvnservices.com/fb/{mgid}.swf @@ -278,6 +284,14 @@ class MTVServicesInfoExtractor(InfoExtractor): if not mgid: mgid = self._extract_triforce_mgid(webpage) + if not mgid: + data = self._parse_json(self._search_regex( + r'__DATA__\s*=\s*({.+?});', webpage, 'data'), None) + main_container = self._extract_child_with_type(data, 'MainContainer') + ab_testing = self._extract_child_with_type(main_container, 'ABTesting') + video_player = self._extract_child_with_type(ab_testing or main_container, 'VideoPlayer') + mgid = video_player['props']['media']['video']['config']['uri'] + return mgid def _real_extract(self, url): @@ -309,7 +323,7 @@ class MTVServicesEmbeddedIE(MTVServicesInfoExtractor): @staticmethod def _extract_url(webpage): mobj = re.search( - r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media.mtvnservices.com/embed/.+?)\1', webpage) + r'<iframe[^>]+?src=(["\'])(?P<url>(?:https?:)?//media\.mtvnservices\.com/embed/.+?)\1', webpage) if mobj: return mobj.group('url') @@ -349,18 +363,6 @@ class MTVIE(MTVServicesInfoExtractor): 'only_matching': True, }] - @staticmethod - def extract_child_with_type(parent, t): - children = parent['children'] - return next(c for c in children if c.get('type') == t) - - def _extract_mgid(self, webpage): - data = self._parse_json(self._search_regex( - r'__DATA__\s*=\s*({.+?});', webpage, 'data'), None) - main_container = self.extract_child_with_type(data, 'MainContainer') - video_player = self.extract_child_with_type(main_container, 'VideoPlayer') - return video_player['props']['media']['video']['config']['uri'] - class MTVJapanIE(MTVServicesInfoExtractor): IE_NAME = 'mtvjapan' diff --git a/youtube_dl/extractor/ninecninemedia.py b/youtube_dl/extractor/ninecninemedia.py index a569c889e..cfc220314 100644 --- a/youtube_dl/extractor/ninecninemedia.py +++ b/youtube_dl/extractor/ninecninemedia.py @@ -23,11 +23,9 @@ class NineCNineMediaIE(InfoExtractor): destination_code, content_id = re.match(self._VALID_URL, url).groups() api_base_url = self._API_BASE_TEMPLATE % (destination_code, content_id) content = self._download_json(api_base_url, content_id, query={ - '$include': '[Media,Season,ContentPackages]', + '$include': '[Media.Name,Season,ContentPackages.Duration,ContentPackages.Id]', }) title = content['Name'] - if len(content['ContentPackages']) > 1: - raise ExtractorError('multiple content packages') content_package = content['ContentPackages'][0] package_id = content_package['Id'] content_package_url = api_base_url + 'contentpackages/%s/' % package_id diff --git a/youtube_dl/extractor/ninegag.py b/youtube_dl/extractor/ninegag.py index dc6a27d36..14390823b 100644 --- a/youtube_dl/extractor/ninegag.py +++ b/youtube_dl/extractor/ninegag.py @@ -1,104 +1,130 @@ from __future__ import unicode_literals -import re - from .common import InfoExtractor -from ..utils import str_to_int +from ..utils import ( + ExtractorError, + determine_ext, + int_or_none, + try_get, + unescapeHTML, + url_or_none, +) class NineGagIE(InfoExtractor): IE_NAME = '9gag' - _VALID_URL = r'https?://(?:www\.)?9gag(?:\.com/tv|\.tv)/(?:p|embed)/(?P<id>[a-zA-Z0-9]+)(?:/(?P<display_id>[^?#/]+))?' + _VALID_URL = r'https?://(?:www\.)?9gag\.com/gag/(?P<id>[^/?&#]+)' _TESTS = [{ - 'url': 'http://9gag.com/tv/p/Kk2X5/people-are-awesome-2013-is-absolutely-awesome', - 'info_dict': { - 'id': 'kXzwOKyGlSA', - 'ext': 'mp4', - 'description': 'This 3-minute video will make you smile and then make you feel untalented and insignificant. Anyway, you should share this awesomeness. (Thanks, Dino!)', - 'title': '\"People Are Awesome 2013\" Is Absolutely Awesome', - 'uploader_id': 'UCdEH6EjDKwtTe-sO2f0_1XA', - 'uploader': 'CompilationChannel', - 'upload_date': '20131110', - 'view_count': int, - }, - 'add_ie': ['Youtube'], - }, { - 'url': 'http://9gag.com/tv/p/aKolP3', + 'url': 'https://9gag.com/gag/ae5Ag7B', 'info_dict': { - 'id': 'aKolP3', + 'id': 'ae5Ag7B', 'ext': 'mp4', - 'title': 'This Guy Travelled 11 countries In 44 days Just To Make This Amazing Video', - 'description': "I just saw more in 1 minute than I've seen in 1 year. This guy's video is epic!!", - 'uploader_id': 'rickmereki', - 'uploader': 'Rick Mereki', - 'upload_date': '20110803', - 'view_count': int, - }, - 'add_ie': ['Vimeo'], - }, { - 'url': 'http://9gag.com/tv/p/KklwM', - 'only_matching': True, - }, { - 'url': 'http://9gag.tv/p/Kk2X5', - 'only_matching': True, + 'title': 'Capybara Agility Training', + 'upload_date': '20191108', + 'timestamp': 1573237208, + 'categories': ['Awesome'], + 'tags': ['Weimaraner', 'American Pit Bull Terrier'], + 'duration': 44, + 'like_count': int, + 'dislike_count': int, + 'comment_count': int, + } }, { - 'url': 'http://9gag.com/tv/embed/a5Dmvl', + # HTML escaped title + 'url': 'https://9gag.com/gag/av5nvyb', 'only_matching': True, }] - _EXTERNAL_VIDEO_PROVIDER = { - '1': { - 'url': '%s', - 'ie_key': 'Youtube', - }, - '2': { - 'url': 'http://player.vimeo.com/video/%s', - 'ie_key': 'Vimeo', - }, - '3': { - 'url': 'http://instagram.com/p/%s', - 'ie_key': 'Instagram', - }, - '4': { - 'url': 'http://vine.co/v/%s', - 'ie_key': 'Vine', - }, - } - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - display_id = mobj.group('display_id') or video_id + post_id = self._match_id(url) + post = self._download_json( + 'https://9gag.com/v1/post', post_id, query={ + 'id': post_id + })['data']['post'] + + if post.get('type') != 'Animated': + raise ExtractorError( + 'The given url does not contain a video', + expected=True) + + title = unescapeHTML(post['title']) + + duration = None + formats = [] + thumbnails = [] + for key, image in (post.get('images') or {}).items(): + image_url = url_or_none(image.get('url')) + if not image_url: + continue + ext = determine_ext(image_url) + image_id = key.strip('image') + common = { + 'url': image_url, + 'width': int_or_none(image.get('width')), + 'height': int_or_none(image.get('height')), + } + if ext in ('jpg', 'png'): + webp_url = image.get('webpUrl') + if webp_url: + t = common.copy() + t.update({ + 'id': image_id + '-webp', + 'url': webp_url, + }) + thumbnails.append(t) + common.update({ + 'id': image_id, + 'ext': ext, + }) + thumbnails.append(common) + elif ext in ('webm', 'mp4'): + if not duration: + duration = int_or_none(image.get('duration')) + common['acodec'] = 'none' if image.get('hasAudio') == 0 else None + for vcodec in ('vp8', 'vp9', 'h265'): + c_url = image.get(vcodec + 'Url') + if not c_url: + continue + c_f = common.copy() + c_f.update({ + 'format_id': image_id + '-' + vcodec, + 'url': c_url, + 'vcodec': vcodec, + }) + formats.append(c_f) + common.update({ + 'ext': ext, + 'format_id': image_id, + }) + formats.append(common) + self._sort_formats(formats) - webpage = self._download_webpage(url, display_id) + section = try_get(post, lambda x: x['postSection']['name']) - post_view = self._parse_json( - self._search_regex( - r'var\s+postView\s*=\s*new\s+app\.PostView\({\s*post:\s*({.+?})\s*,\s*posts:\s*prefetchedCurrentPost', - webpage, 'post view'), - display_id) + tags = None + post_tags = post.get('tags') + if post_tags: + tags = [] + for tag in post_tags: + tag_key = tag.get('key') + if not tag_key: + continue + tags.append(tag_key) - ie_key = None - source_url = post_view.get('sourceUrl') - if not source_url: - external_video_id = post_view['videoExternalId'] - external_video_provider = post_view['videoExternalProvider'] - source_url = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['url'] % external_video_id - ie_key = self._EXTERNAL_VIDEO_PROVIDER[external_video_provider]['ie_key'] - title = post_view['title'] - description = post_view.get('description') - view_count = str_to_int(post_view.get('externalView')) - thumbnail = post_view.get('thumbnail_700w') or post_view.get('ogImageUrl') or post_view.get('thumbnail_300w') + get_count = lambda x: int_or_none(post.get(x + 'Count')) return { - '_type': 'url_transparent', - 'url': source_url, - 'ie_key': ie_key, - 'id': video_id, - 'display_id': display_id, + 'id': post_id, 'title': title, - 'description': description, - 'view_count': view_count, - 'thumbnail': thumbnail, + 'timestamp': int_or_none(post.get('creationTs')), + 'duration': duration, + 'formats': formats, + 'thumbnails': thumbnails, + 'like_count': get_count('upVote'), + 'dislike_count': get_count('downVote'), + 'comment_count': get_count('comments'), + 'age_limit': 18 if post.get('nsfw') == 1 else None, + 'categories': [section] if section else None, + 'tags': tags, } diff --git a/youtube_dl/extractor/njpwworld.py b/youtube_dl/extractor/njpwworld.py index 025c5d249..3639d142f 100644 --- a/youtube_dl/extractor/njpwworld.py +++ b/youtube_dl/extractor/njpwworld.py @@ -6,30 +6,40 @@ import re from .common import InfoExtractor from ..compat import compat_urlparse from ..utils import ( - extract_attributes, get_element_by_class, urlencode_postdata, ) class NJPWWorldIE(InfoExtractor): - _VALID_URL = r'https?://njpwworld\.com/p/(?P<id>[a-z0-9_]+)' + _VALID_URL = r'https?://(front\.)?njpwworld\.com/p/(?P<id>[a-z0-9_]+)' IE_DESC = '新日本プロレスワールド' _NETRC_MACHINE = 'njpwworld' - _TEST = { + _TESTS = [{ 'url': 'http://njpwworld.com/p/s_series_00155_1_9/', 'info_dict': { 'id': 's_series_00155_1_9', 'ext': 'mp4', - 'title': '第9試合 ランディ・サベージ vs リック・スタイナー', + 'title': '闘強導夢2000 2000年1月4日 東京ドーム 第9試合 ランディ・サベージ VS リック・スタイナー', 'tags': list, }, 'params': { 'skip_download': True, # AES-encrypted m3u8 }, 'skip': 'Requires login', - } + }, { + 'url': 'https://front.njpwworld.com/p/s_series_00563_16_bs', + 'info_dict': { + 'id': 's_series_00563_16_bs', + 'ext': 'mp4', + 'title': 'WORLD TAG LEAGUE 2020 & BEST OF THE SUPER Jr.27 2020年12月6日 福岡・福岡国際センター バックステージコメント(字幕あり)', + 'tags': ["福岡・福岡国際センター", "バックステージコメント", "2020", "20年代"], + }, + 'params': { + 'skip_download': True, + }, + }] _LOGIN_URL = 'https://front.njpwworld.com/auth/login' @@ -64,35 +74,27 @@ class NJPWWorldIE(InfoExtractor): webpage = self._download_webpage(url, video_id) formats = [] - for mobj in re.finditer(r'<a[^>]+\bhref=(["\'])/player.+?[^>]*>', webpage): - player = extract_attributes(mobj.group(0)) - player_path = player.get('href') - if not player_path: - continue - kind = self._search_regex( - r'(low|high)$', player.get('class') or '', 'kind', - default='low') + for kind, vid in re.findall(r'if\s+\(\s*imageQualityType\s*==\s*\'([^\']+)\'\s*\)\s*{\s*video_id\s*=\s*"(\d+)"', webpage): + player_path = '/intent?id=%s&type=url' % vid player_url = compat_urlparse.urljoin(url, player_path) - player_page = self._download_webpage( - player_url, video_id, note='Downloading player page') - entries = self._parse_html5_media_entries( - player_url, player_page, video_id, m3u8_id='hls-%s' % kind, - m3u8_entry_protocol='m3u8_native') - kind_formats = entries[0]['formats'] - for f in kind_formats: - f['quality'] = 2 if kind == 'high' else 1 - formats.extend(kind_formats) + formats.append({ + 'url': player_url, + 'format_id': kind, + 'ext': 'mp4', + 'protocol': 'm3u8', + 'quality': 2 if kind == 'high' else 1, + }) self._sort_formats(formats) - post_content = get_element_by_class('post-content', webpage) + tag_block = get_element_by_class('tag-block', webpage) tags = re.findall( - r'<li[^>]+class="tag-[^"]+"><a[^>]*>([^<]+)</a></li>', post_content - ) if post_content else None + r'<a[^>]+class="tag-[^"]+"[^>]*>([^<]+)</a>', tag_block + ) if tag_block else None return { 'id': video_id, - 'title': self._og_search_title(webpage), + 'title': get_element_by_class('article-title', webpage) or self._og_search_title(webpage), 'formats': formats, 'tags': tags, } diff --git a/youtube_dl/extractor/nrk.py b/youtube_dl/extractor/nrk.py index 40dee2162..6d01a25c3 100644 --- a/youtube_dl/extractor/nrk.py +++ b/youtube_dl/extractor/nrk.py @@ -58,7 +58,7 @@ class NRKBaseIE(InfoExtractor): def _call_api(self, path, video_id, item=None, note=None, fatal=True, query=None): return self._download_json( - urljoin('http://psapi.nrk.no/', path), + urljoin('https://psapi.nrk.no/', path), video_id, note or 'Downloading %s JSON' % item, fatal=fatal, query=query, headers={'Accept-Encoding': 'gzip, deflate, br'}) diff --git a/youtube_dl/extractor/orf.py b/youtube_dl/extractor/orf.py index 700ce448c..8d537d7ae 100644 --- a/youtube_dl/extractor/orf.py +++ b/youtube_dl/extractor/orf.py @@ -98,6 +98,9 @@ class ORFTVthekIE(InfoExtractor): elif ext == 'f4m': formats.extend(self._extract_f4m_formats( src, video_id, f4m_id=format_id, fatal=False)) + elif ext == 'mpd': + formats.extend(self._extract_mpd_formats( + src, video_id, mpd_id=format_id, fatal=False)) else: formats.append({ 'format_id': format_id, @@ -140,6 +143,25 @@ class ORFTVthekIE(InfoExtractor): }) upload_date = unified_strdate(sd.get('created_date')) + + thumbnails = [] + preview = sd.get('preview_image_url') + if preview: + thumbnails.append({ + 'id': 'preview', + 'url': preview, + 'preference': 0, + }) + image = sd.get('image_full_url') + if not image and len(data_jsb) == 1: + image = self._og_search_thumbnail(webpage) + if image: + thumbnails.append({ + 'id': 'full', + 'url': image, + 'preference': 1, + }) + entries.append({ '_type': 'video', 'id': video_id, @@ -149,7 +171,7 @@ class ORFTVthekIE(InfoExtractor): 'description': sd.get('description'), 'duration': int_or_none(sd.get('duration_in_seconds')), 'upload_date': upload_date, - 'thumbnail': sd.get('image_full_url'), + 'thumbnails': thumbnails, }) return { @@ -182,7 +204,7 @@ class ORFRadioIE(InfoExtractor): duration = end - start if end and start else None entries.append({ 'id': loop_stream_id.replace('.mp3', ''), - 'url': 'http://loopstream01.apa.at/?channel=%s&id=%s' % (self._LOOP_STATION, loop_stream_id), + 'url': 'https://loopstream01.apa.at/?channel=%s&id=%s' % (self._LOOP_STATION, loop_stream_id), 'title': title, 'description': clean_html(data.get('subtitle')), 'duration': duration, diff --git a/youtube_dl/extractor/palcomp3.py b/youtube_dl/extractor/palcomp3.py new file mode 100644 index 000000000..fb29d83f9 --- /dev/null +++ b/youtube_dl/extractor/palcomp3.py @@ -0,0 +1,148 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + int_or_none, + str_or_none, + try_get, +) + + +class PalcoMP3BaseIE(InfoExtractor): + _GQL_QUERY_TMPL = '''{ + artist(slug: "%s") { + %s + } +}''' + _ARTIST_FIELDS_TMPL = '''music(slug: "%%s") { + %s + }''' + _MUSIC_FIELDS = '''duration + hls + mp3File + musicID + plays + title''' + + def _call_api(self, artist_slug, artist_fields): + return self._download_json( + 'https://www.palcomp3.com.br/graphql/', artist_slug, query={ + 'query': self._GQL_QUERY_TMPL % (artist_slug, artist_fields), + })['data'] + + def _parse_music(self, music): + music_id = compat_str(music['musicID']) + title = music['title'] + + formats = [] + hls_url = music.get('hls') + if hls_url: + formats.append({ + 'url': hls_url, + 'protocol': 'm3u8_native', + 'ext': 'mp4', + }) + mp3_file = music.get('mp3File') + if mp3_file: + formats.append({ + 'url': mp3_file, + }) + + return { + 'id': music_id, + 'title': title, + 'formats': formats, + 'duration': int_or_none(music.get('duration')), + 'view_count': int_or_none(music.get('plays')), + } + + def _real_initialize(self): + self._ARTIST_FIELDS_TMPL = self._ARTIST_FIELDS_TMPL % self._MUSIC_FIELDS + + def _real_extract(self, url): + artist_slug, music_slug = re.match(self._VALID_URL, url).groups() + artist_fields = self._ARTIST_FIELDS_TMPL % music_slug + music = self._call_api(artist_slug, artist_fields)['artist']['music'] + return self._parse_music(music) + + +class PalcoMP3IE(PalcoMP3BaseIE): + IE_NAME = 'PalcoMP3:song' + _VALID_URL = r'https?://(?:www\.)?palcomp3\.com(?:\.br)?/(?P<artist>[^/]+)/(?P<id>[^/?&#]+)' + _TESTS = [{ + 'url': 'https://www.palcomp3.com/maiaraemaraisaoficial/nossas-composicoes-cuida-bem-dela/', + 'md5': '99fd6405b2d8fd589670f6db1ba3b358', + 'info_dict': { + 'id': '3162927', + 'ext': 'mp3', + 'title': 'Nossas Composições - CUIDA BEM DELA', + 'duration': 210, + 'view_count': int, + } + }] + + @classmethod + def suitable(cls, url): + return False if PalcoMP3VideoIE.suitable(url) else super(PalcoMP3IE, cls).suitable(url) + + +class PalcoMP3ArtistIE(PalcoMP3BaseIE): + IE_NAME = 'PalcoMP3:artist' + _VALID_URL = r'https?://(?:www\.)?palcomp3\.com(?:\.br)?/(?P<id>[^/?&#]+)' + _TESTS = [{ + 'url': 'https://www.palcomp3.com.br/condedoforro/', + 'info_dict': { + 'id': '358396', + 'title': 'Conde do Forró', + }, + 'playlist_mincount': 188, + }] + _ARTIST_FIELDS_TMPL = '''artistID + musics { + nodes { + %s + } + } + name''' + + @ classmethod + def suitable(cls, url): + return False if re.match(PalcoMP3IE._VALID_URL, url) else super(PalcoMP3ArtistIE, cls).suitable(url) + + def _real_extract(self, url): + artist_slug = self._match_id(url) + artist = self._call_api(artist_slug, self._ARTIST_FIELDS_TMPL)['artist'] + + def entries(): + for music in (try_get(artist, lambda x: x['musics']['nodes'], list) or []): + yield self._parse_music(music) + + return self.playlist_result( + entries(), str_or_none(artist.get('artistID')), artist.get('name')) + + +class PalcoMP3VideoIE(PalcoMP3BaseIE): + IE_NAME = 'PalcoMP3:video' + _VALID_URL = r'https?://(?:www\.)?palcomp3\.com(?:\.br)?/(?P<artist>[^/]+)/(?P<id>[^/?&#]+)/?#clipe' + _TESTS = [{ + 'url': 'https://www.palcomp3.com/maiaraemaraisaoficial/maiara-e-maraisa-voce-faz-falta-aqui-ao-vivo-em-vicosa-mg/#clipe', + 'add_ie': ['Youtube'], + 'info_dict': { + 'id': '_pD1nR2qqPg', + 'ext': 'mp4', + 'title': 'Maiara e Maraisa - Você Faz Falta Aqui - DVD Ao Vivo Em Campo Grande', + 'description': 'md5:7043342c09a224598e93546e98e49282', + 'upload_date': '20161107', + 'uploader_id': 'maiaramaraisaoficial', + 'uploader': 'Maiara e Maraisa', + } + }] + _MUSIC_FIELDS = 'youtubeID' + + def _parse_music(self, music): + youtube_id = music['youtubeID'] + return self.url_result(youtube_id, 'Youtube', youtube_id) diff --git a/youtube_dl/extractor/peertube.py b/youtube_dl/extractor/peertube.py index c39d12728..3af533925 100644 --- a/youtube_dl/extractor/peertube.py +++ b/youtube_dl/extractor/peertube.py @@ -413,7 +413,8 @@ class PeerTubeIE(InfoExtractor): peertube3\.cpy\.re| peertube2\.cpy\.re| videos\.tcit\.fr| - peertube\.cpy\.re + peertube\.cpy\.re| + canard\.tube )''' _UUID_RE = r'[\da-fA-F]{8}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{12}' _API_BASE = 'https://%s/api/v1/videos/%s/%s' @@ -451,6 +452,18 @@ class PeerTubeIE(InfoExtractor): 'categories': ['Science & Technology'], } }, { + # Issue #26002 + 'url': 'peertube:spacepub.space:d8943b2d-8280-497b-85ec-bc282ec2afdc', + 'info_dict': { + 'id': 'd8943b2d-8280-497b-85ec-bc282ec2afdc', + 'ext': 'mp4', + 'title': 'Dot matrix printer shell demo', + 'uploader_id': '3', + 'timestamp': 1587401293, + 'upload_date': '20200420', + 'uploader': 'Drew DeVault', + } + }, { 'url': 'https://peertube.tamanoir.foucry.net/videos/watch/0b04f13d-1e18-4f1d-814e-4979aa7c9c44', 'only_matching': True, }, { @@ -526,7 +539,15 @@ class PeerTubeIE(InfoExtractor): title = video['name'] formats = [] - for file_ in video['files']: + files = video.get('files') or [] + for playlist in (video.get('streamingPlaylists') or []): + if not isinstance(playlist, dict): + continue + playlist_files = playlist.get('files') + if not (playlist_files and isinstance(playlist_files, list)): + continue + files.extend(playlist_files) + for file_ in files: if not isinstance(file_, dict): continue file_url = url_or_none(file_.get('fileUrl')) @@ -548,15 +569,15 @@ class PeerTubeIE(InfoExtractor): formats.append(f) self._sort_formats(formats) - full_description = self._call_api( - host, video_id, 'description', note='Downloading description JSON', - fatal=False) + description = video.get('description') + if len(description) >= 250: + # description is shortened + full_description = self._call_api( + host, video_id, 'description', note='Downloading description JSON', + fatal=False) - description = None - if isinstance(full_description, dict): - description = str_or_none(full_description.get('description')) - if not description: - description = video.get('description') + if isinstance(full_description, dict): + description = str_or_none(full_description.get('description')) or description subtitles = self.extract_subtitles(host, video_id) @@ -578,11 +599,13 @@ class PeerTubeIE(InfoExtractor): else: age_limit = None + webpage_url = 'https://%s/videos/watch/%s' % (host, video_id) + return { 'id': video_id, 'title': title, 'description': description, - 'thumbnail': urljoin(url, video.get('thumbnailPath')), + 'thumbnail': urljoin(webpage_url, video.get('thumbnailPath')), 'timestamp': unified_timestamp(video.get('publishedAt')), 'uploader': account_data('displayName', compat_str), 'uploader_id': str_or_none(account_data('id', int)), @@ -600,5 +623,6 @@ class PeerTubeIE(InfoExtractor): 'tags': try_get(video, lambda x: x['tags'], list), 'categories': categories, 'formats': formats, - 'subtitles': subtitles + 'subtitles': subtitles, + 'webpage_url': webpage_url, } diff --git a/youtube_dl/extractor/periscope.py b/youtube_dl/extractor/periscope.py index b15906390..b93a02b7d 100644 --- a/youtube_dl/extractor/periscope.py +++ b/youtube_dl/extractor/periscope.py @@ -12,6 +12,10 @@ from ..utils import ( class PeriscopeBaseIE(InfoExtractor): + _M3U8_HEADERS = { + 'Referer': 'https://www.periscope.tv/' + } + def _call_api(self, method, query, item_id): return self._download_json( 'https://api.periscope.tv/api/v2/%s' % method, @@ -54,9 +58,11 @@ class PeriscopeBaseIE(InfoExtractor): m3u8_url, video_id, 'mp4', entry_protocol='m3u8_native' if state in ('ended', 'timed_out') else 'm3u8', - m3u8_id=format_id, fatal=fatal) + m3u8_id=format_id, fatal=fatal, headers=self._M3U8_HEADERS) if len(m3u8_formats) == 1: self._add_width_and_height(m3u8_formats[0], width, height) + for f in m3u8_formats: + f.setdefault('http_headers', {}).update(self._M3U8_HEADERS) return m3u8_formats diff --git a/youtube_dl/extractor/phoenix.py b/youtube_dl/extractor/phoenix.py index e435c28e1..e3ea01443 100644 --- a/youtube_dl/extractor/phoenix.py +++ b/youtube_dl/extractor/phoenix.py @@ -1,45 +1,133 @@ +# coding: utf-8 from __future__ import unicode_literals -from .dreisat import DreiSatIE +import re +from .youtube import YoutubeIE +from .zdf import ZDFBaseIE +from ..compat import compat_str +from ..utils import ( + int_or_none, + merge_dicts, + try_get, + unified_timestamp, + urljoin, +) -class PhoenixIE(DreiSatIE): + +class PhoenixIE(ZDFBaseIE): IE_NAME = 'phoenix.de' - _VALID_URL = r'''(?x)https?://(?:www\.)?phoenix\.de/content/ - (?: - phoenix/die_sendungen/(?:[^/]+/)? - )? - (?P<id>[0-9]+)''' - _TESTS = [ - { - 'url': 'http://www.phoenix.de/content/884301', - 'md5': 'ed249f045256150c92e72dbb70eadec6', - 'info_dict': { - 'id': '884301', - 'ext': 'mp4', - 'title': 'Michael Krons mit Hans-Werner Sinn', - 'description': 'Im Dialog - Sa. 25.10.14, 00.00 - 00.35 Uhr', - 'upload_date': '20141025', - 'uploader': 'Im Dialog', - } + _VALID_URL = r'https?://(?:www\.)?phoenix\.de/(?:[^/]+/)*[^/?#&]*-a-(?P<id>\d+)\.html' + _TESTS = [{ + # Same as https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html + 'url': 'https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html', + 'md5': '34ec321e7eb34231fd88616c65c92db0', + 'info_dict': { + 'id': '210222_phx_nachgehakt_corona_protest', + 'ext': 'mp4', + 'title': 'Wohin führt der Protest in der Pandemie?', + 'description': 'md5:7d643fe7f565e53a24aac036b2122fbd', + 'duration': 1691, + 'timestamp': 1613902500, + 'upload_date': '20210221', + 'uploader': 'Phoenix', + 'series': 'corona nachgehakt', + 'episode': 'Wohin führt der Protest in der Pandemie?', }, - { - 'url': 'http://www.phoenix.de/content/phoenix/die_sendungen/869815', - 'only_matching': True, + }, { + # Youtube embed + 'url': 'https://www.phoenix.de/sendungen/gespraeche/phoenix-streitgut-brennglas-corona-a-1965505.html', + 'info_dict': { + 'id': 'hMQtqFYjomk', + 'ext': 'mp4', + 'title': 'phoenix streitgut: Brennglas Corona - Wie gerecht ist unsere Gesellschaft?', + 'description': 'md5:ac7a02e2eb3cb17600bc372e4ab28fdd', + 'duration': 3509, + 'upload_date': '20201219', + 'uploader': 'phoenix', + 'uploader_id': 'phoenix', }, - { - 'url': 'http://www.phoenix.de/content/phoenix/die_sendungen/diskussionen/928234', - 'only_matching': True, + 'params': { + 'skip_download': True, }, - ] + }, { + 'url': 'https://www.phoenix.de/entwicklungen-in-russland-a-2044720.html', + 'only_matching': True, + }, { + # no media + 'url': 'https://www.phoenix.de/sendungen/dokumentationen/mit-dem-jumbo-durch-die-nacht-a-89625.html', + 'only_matching': True, + }, { + # Same as https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html + 'url': 'https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche', + 'only_matching': True, + }] def _real_extract(self, url): - video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) + article_id = self._match_id(url) + + article = self._download_json( + 'https://www.phoenix.de/response/id/%s' % article_id, article_id, + 'Downloading article JSON') + + video = article['absaetze'][0] + title = video.get('titel') or article.get('subtitel') + + if video.get('typ') == 'video-youtube': + video_id = video['id'] + return self.url_result( + video_id, ie=YoutubeIE.ie_key(), video_id=video_id, + video_title=title) + + video_id = compat_str(video.get('basename') or video.get('content')) - internal_id = self._search_regex( - r'<div class="phx_vod" id="phx_vod_([0-9]+)"', - webpage, 'internal video ID') + details = self._download_json( + 'https://www.phoenix.de/php/mediaplayer/data/beitrags_details.php', + video_id, 'Downloading details JSON', query={ + 'ak': 'web', + 'ptmd': 'true', + 'id': video_id, + 'profile': 'player2', + }) + + title = title or details['title'] + content_id = details['tracking']['nielsen']['content']['assetid'] + + info = self._extract_ptmd( + 'https://tmd.phoenix.de/tmd/2/ngplayer_2_3/vod/ptmd/phoenix/%s' % content_id, + content_id, None, url) + + duration = int_or_none(try_get( + details, lambda x: x['tracking']['nielsen']['content']['length'])) + timestamp = unified_timestamp(details.get('editorialDate')) + series = try_get( + details, lambda x: x['tracking']['nielsen']['content']['program'], + compat_str) + episode = title if details.get('contentType') == 'episode' else None + + thumbnails = [] + teaser_images = try_get(details, lambda x: x['teaserImageRef']['layouts'], dict) or {} + for thumbnail_key, thumbnail_url in teaser_images.items(): + thumbnail_url = urljoin(url, thumbnail_url) + if not thumbnail_url: + continue + thumbnail = { + 'url': thumbnail_url, + } + m = re.match('^([0-9]+)x([0-9]+)$', thumbnail_key) + if m: + thumbnail['width'] = int(m.group(1)) + thumbnail['height'] = int(m.group(2)) + thumbnails.append(thumbnail) - api_url = 'http://www.phoenix.de/php/mediaplayer/data/beitrags_details.php?ak=web&id=%s' % internal_id - return self.extract_from_xml_url(video_id, api_url) + return merge_dicts(info, { + 'id': content_id, + 'title': title, + 'description': details.get('leadParagraph'), + 'duration': duration, + 'thumbnails': thumbnails, + 'timestamp': timestamp, + 'uploader': details.get('tvService'), + 'series': series, + 'episode': episode, + }) diff --git a/youtube_dl/extractor/picarto.py b/youtube_dl/extractor/picarto.py index 8099ef1d6..e6c51e16b 100644 --- a/youtube_dl/extractor/picarto.py +++ b/youtube_dl/extractor/picarto.py @@ -1,22 +1,15 @@ # coding: utf-8 from __future__ import unicode_literals -import re -import time - from .common import InfoExtractor -from ..compat import compat_str from ..utils import ( ExtractorError, js_to_json, - try_get, - update_url_query, - urlencode_postdata, ) class PicartoIE(InfoExtractor): - _VALID_URL = r'https?://(?:www.)?picarto\.tv/(?P<id>[a-zA-Z0-9]+)(?:/(?P<token>[a-zA-Z0-9]+))?' + _VALID_URL = r'https?://(?:www.)?picarto\.tv/(?P<id>[a-zA-Z0-9]+)' _TEST = { 'url': 'https://picarto.tv/Setz', 'info_dict': { @@ -34,65 +27,46 @@ class PicartoIE(InfoExtractor): return False if PicartoVodIE.suitable(url) else super(PicartoIE, cls).suitable(url) def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - channel_id = mobj.group('id') - - metadata = self._download_json( - 'https://api.picarto.tv/v1/channel/name/' + channel_id, - channel_id) - - if metadata.get('online') is False: + channel_id = self._match_id(url) + + data = self._download_json( + 'https://ptvintern.picarto.tv/ptvapi', channel_id, query={ + 'query': '''{ + channel(name: "%s") { + adult + id + online + stream_name + title + } + getLoadBalancerUrl(channel_name: "%s") { + url + } +}''' % (channel_id, channel_id), + })['data'] + metadata = data['channel'] + + if metadata.get('online') == 0: raise ExtractorError('Stream is offline', expected=True) + title = metadata['title'] cdn_data = self._download_json( - 'https://picarto.tv/process/channel', channel_id, - data=urlencode_postdata({'loadbalancinginfo': channel_id}), - note='Downloading load balancing info') - - token = mobj.group('token') or 'public' - params = { - 'con': int(time.time() * 1000), - 'token': token, - } + data['getLoadBalancerUrl']['url'] + '/stream/json_' + metadata['stream_name'] + '.js', + channel_id, 'Downloading load balancing info') - prefered_edge = cdn_data.get('preferedEdge') formats = [] - - for edge in cdn_data['edges']: - edge_ep = edge.get('ep') - if not edge_ep or not isinstance(edge_ep, compat_str): + for source in (cdn_data.get('source') or []): + source_url = source.get('url') + if not source_url: continue - edge_id = edge.get('id') - for tech in cdn_data['techs']: - tech_label = tech.get('label') - tech_type = tech.get('type') - preference = 0 - if edge_id == prefered_edge: - preference += 1 - format_id = [] - if edge_id: - format_id.append(edge_id) - if tech_type == 'application/x-mpegurl' or tech_label == 'HLS': - format_id.append('hls') - formats.extend(self._extract_m3u8_formats( - update_url_query( - 'https://%s/hls/%s/index.m3u8' - % (edge_ep, channel_id), params), - channel_id, 'mp4', preference=preference, - m3u8_id='-'.join(format_id), fatal=False)) - continue - elif tech_type == 'video/mp4' or tech_label == 'MP4': - format_id.append('mp4') - formats.append({ - 'url': update_url_query( - 'https://%s/mp4/%s.mp4' % (edge_ep, channel_id), - params), - 'format_id': '-'.join(format_id), - 'preference': preference, - }) - else: - # rtmp format does not seem to work - continue + source_type = source.get('type') + if source_type == 'html5/application/vnd.apple.mpegurl': + formats.extend(self._extract_m3u8_formats( + source_url, channel_id, 'mp4', m3u8_id='hls', fatal=False)) + elif source_type == 'html5/video/mp4': + formats.append({ + 'url': source_url, + }) self._sort_formats(formats) mature = metadata.get('adult') @@ -103,10 +77,10 @@ class PicartoIE(InfoExtractor): return { 'id': channel_id, - 'title': self._live_title(metadata.get('title') or channel_id), + 'title': self._live_title(title.strip()), 'is_live': True, - 'thumbnail': try_get(metadata, lambda x: x['thumbnails']['web']), 'channel': channel_id, + 'channel_id': metadata.get('id'), 'channel_url': 'https://picarto.tv/%s' % channel_id, 'age_limit': age_limit, 'formats': formats, diff --git a/youtube_dl/extractor/pinterest.py b/youtube_dl/extractor/pinterest.py index b249c9eda..42528d746 100644 --- a/youtube_dl/extractor/pinterest.py +++ b/youtube_dl/extractor/pinterest.py @@ -31,6 +31,7 @@ class PinterestBaseIE(InfoExtractor): title = (data.get('title') or data.get('grid_title') or video_id).strip() + urls = [] formats = [] duration = None if extract_formats: @@ -38,8 +39,9 @@ class PinterestBaseIE(InfoExtractor): if not isinstance(format_dict, dict): continue format_url = url_or_none(format_dict.get('url')) - if not format_url: + if not format_url or format_url in urls: continue + urls.append(format_url) duration = float_or_none(format_dict.get('duration'), scale=1000) ext = determine_ext(format_url) if 'hls' in format_id.lower() or ext == 'm3u8': diff --git a/youtube_dl/extractor/playstuff.py b/youtube_dl/extractor/playstuff.py new file mode 100644 index 000000000..5a329957f --- /dev/null +++ b/youtube_dl/extractor/playstuff.py @@ -0,0 +1,65 @@ +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..compat import compat_str +from ..utils import ( + smuggle_url, + try_get, +) + + +class PlayStuffIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?play\.stuff\.co\.nz/details/(?P<id>[^/?#&]+)' + _TESTS = [{ + 'url': 'https://play.stuff.co.nz/details/608778ac1de1c4001a3fa09a', + 'md5': 'c82d3669e5247c64bc382577843e5bd0', + 'info_dict': { + 'id': '6250584958001', + 'ext': 'mp4', + 'title': 'Episode 1: Rotorua/Mt Maunganui/Tauranga', + 'description': 'md5:c154bafb9f0dd02d01fd4100fb1c1913', + 'uploader_id': '6005208634001', + 'timestamp': 1619491027, + 'upload_date': '20210427', + }, + 'add_ie': ['BrightcoveNew'], + }, { + # geo restricted, bypassable + 'url': 'https://play.stuff.co.nz/details/_6155660351001', + 'only_matching': True, + }] + BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/%s_default/index.html?videoId=%s' + + def _real_extract(self, url): + video_id = self._match_id(url) + + webpage = self._download_webpage(url, video_id) + + state = self._parse_json( + self._search_regex( + r'__INITIAL_STATE__\s*=\s*({.+?})\s*;', webpage, 'state'), + video_id) + + account_id = try_get( + state, lambda x: x['configurations']['accountId'], + compat_str) or '6005208634001' + player_id = try_get( + state, lambda x: x['configurations']['playerId'], + compat_str) or 'default' + + entries = [] + for item_id, video in state['items'].items(): + if not isinstance(video, dict): + continue + asset_id = try_get( + video, lambda x: x['content']['attributes']['assetId'], + compat_str) + if not asset_id: + continue + entries.append(self.url_result( + smuggle_url( + self.BRIGHTCOVE_URL_TEMPLATE % (account_id, player_id, asset_id), + {'geo_countries': ['NZ']}), + 'BrightcoveNew', video_id)) + + return self.playlist_result(entries, video_id) diff --git a/youtube_dl/extractor/pluralsight.py b/youtube_dl/extractor/pluralsight.py index abd08bc28..2d63855df 100644 --- a/youtube_dl/extractor/pluralsight.py +++ b/youtube_dl/extractor/pluralsight.py @@ -393,7 +393,7 @@ query viewClip { # To somewhat reduce the probability of these consequences # we will sleep random amount of time before each call to ViewClip. self._sleep( - random.randint(2, 5), display_id, + random.randint(5, 10), display_id, '%(video_id)s: Waiting for %(timeout)s seconds to avoid throttling') if not viewclip: diff --git a/youtube_dl/extractor/pornhub.py b/youtube_dl/extractor/pornhub.py index 2fcbd186f..e2e1500ff 100644 --- a/youtube_dl/extractor/pornhub.py +++ b/youtube_dl/extractor/pornhub.py @@ -22,11 +22,16 @@ from ..utils import ( orderedSet, remove_quotes, str_to_int, + update_url_query, + urlencode_postdata, url_or_none, ) class PornHubBaseIE(InfoExtractor): + _NETRC_MACHINE = 'pornhub' + _PORNHUB_HOST_RE = r'(?:(?P<host>pornhub(?:premium)?\.(?:com|net|org))|pornhubthbh7ap3u\.onion)' + def _download_webpage_handle(self, *args, **kwargs): def dl(*args, **kwargs): return super(PornHubBaseIE, self)._download_webpage_handle(*args, **kwargs) @@ -52,17 +57,79 @@ class PornHubBaseIE(InfoExtractor): return webpage, urlh + def _real_initialize(self): + self._logged_in = False + + def _login(self, host): + if self._logged_in: + return + + site = host.split('.')[0] + + # Both sites pornhub and pornhubpremium have separate accounts + # so there should be an option to provide credentials for both. + # At the same time some videos are available under the same video id + # on both sites so that we have to identify them as the same video. + # For that purpose we have to keep both in the same extractor + # but under different netrc machines. + username, password = self._get_login_info(netrc_machine=site) + if username is None: + return + + login_url = 'https://www.%s/%slogin' % (host, 'premium/' if 'premium' in host else '') + login_page = self._download_webpage( + login_url, None, 'Downloading %s login page' % site) + + def is_logged(webpage): + return any(re.search(p, webpage) for p in ( + r'class=["\']signOut', + r'>Sign\s+[Oo]ut\s*<')) + + if is_logged(login_page): + self._logged_in = True + return + + login_form = self._hidden_inputs(login_page) + + login_form.update({ + 'username': username, + 'password': password, + }) + + response = self._download_json( + 'https://www.%s/front/authenticate' % host, None, + 'Logging in to %s' % site, + data=urlencode_postdata(login_form), + headers={ + 'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8', + 'Referer': login_url, + 'X-Requested-With': 'XMLHttpRequest', + }) + + if response.get('success') == '1': + self._logged_in = True + return + + message = response.get('message') + if message is not None: + raise ExtractorError( + 'Unable to login: %s' % message, expected=True) + + raise ExtractorError('Unable to log in') + class PornHubIE(PornHubBaseIE): IE_DESC = 'PornHub and Thumbzilla' _VALID_URL = r'''(?x) https?:// (?: - (?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:view_video\.php|video/show)\?viewkey=|embed/)| + (?:[^/]+\.)? + %s + /(?:(?:view_video\.php|video/show)\?viewkey=|embed/)| (?:www\.)?thumbzilla\.com/video/ ) (?P<id>[\da-z]+) - ''' + ''' % PornHubBaseIE._PORNHUB_HOST_RE _TESTS = [{ 'url': 'http://www.pornhub.com/view_video.php?viewkey=648719015', 'md5': 'a6391306d050e4547f62b3f485dd9ba9', @@ -103,6 +170,7 @@ class PornHubIE(PornHubBaseIE): 'params': { 'skip_download': True, }, + 'skip': 'Video has been flagged for verification in accordance with our trust and safety policy', }, { # subtitles 'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5af5fef7c2aa7', @@ -163,12 +231,27 @@ class PornHubIE(PornHubBaseIE): }, { 'url': 'https://www.pornhubpremium.com/view_video.php?viewkey=ph5e4acdae54a82', 'only_matching': True, + }, { + # Some videos are available with the same id on both premium + # and non-premium sites (e.g. this and the following test) + 'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5f75b0f4b18e3', + 'only_matching': True, + }, { + 'url': 'https://www.pornhubpremium.com/view_video.php?viewkey=ph5f75b0f4b18e3', + 'only_matching': True, + }, { + # geo restricted + 'url': 'https://www.pornhub.com/view_video.php?viewkey=ph5a9813bfa7156', + 'only_matching': True, + }, { + 'url': 'http://pornhubthbh7ap3u.onion/view_video.php?viewkey=ph5a9813bfa7156', + 'only_matching': True, }] @staticmethod def _extract_urls(webpage): return re.findall( - r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub\.(?:com|net|org)/embed/[\da-z]+)', + r'<iframe[^>]+?src=["\'](?P<url>(?:https?:)?//(?:www\.)?pornhub(?:premium)?\.(?:com|net|org)/embed/[\da-z]+)', webpage) def _extract_count(self, pattern, webpage, name): @@ -180,12 +263,7 @@ class PornHubIE(PornHubBaseIE): host = mobj.group('host') or 'pornhub.com' video_id = mobj.group('id') - if 'premium' in host: - if not self._downloader.params.get('cookiefile'): - raise ExtractorError( - 'PornHub Premium requires authentication.' - ' You may want to use --cookies.', - expected=True) + self._login(host) self._set_cookie(host, 'age_verified', '1') @@ -198,7 +276,8 @@ class PornHubIE(PornHubBaseIE): webpage = dl_webpage('pc') error_msg = self._html_search_regex( - r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>', + (r'(?s)<div[^>]+class=(["\'])(?:(?!\1).)*\b(?:removed|userMessageSection)\b(?:(?!\1).)*\1[^>]*>(?P<error>.+?)</div>', + r'(?s)<section[^>]+class=["\']noVideo["\'][^>]*>(?P<error>.+?)</section>'), webpage, 'error message', default=None, group='error') if error_msg: error_msg = re.sub(r'\s+', ' ', error_msg) @@ -206,6 +285,11 @@ class PornHubIE(PornHubBaseIE): 'PornHub said: %s' % error_msg, expected=True, video_id=video_id) + if any(re.search(p, webpage) for p in ( + r'class=["\']geoBlocked["\']', + r'>\s*This content is unavailable in your country')): + self.raise_geo_restricted() + # video_title from flashvars contains whitespace instead of non-ASCII (see # http://www.pornhub.com/view_video.php?viewkey=1331683002), not relying # on that anymore. @@ -327,35 +411,49 @@ class PornHubIE(PornHubBaseIE): upload_date = None formats = [] - for video_url, height in video_urls: - if not upload_date: - upload_date = self._search_regex( - r'/(\d{6}/\d{2})/', video_url, 'upload data', default=None) - if upload_date: - upload_date = upload_date.replace('/', '') - ext = determine_ext(video_url) + + def add_format(format_url, height=None): + ext = determine_ext(format_url) if ext == 'mpd': formats.extend(self._extract_mpd_formats( - video_url, video_id, mpd_id='dash', fatal=False)) - continue - elif ext == 'm3u8': + format_url, video_id, mpd_id='dash', fatal=False)) + return + if ext == 'm3u8': formats.extend(self._extract_m3u8_formats( - video_url, video_id, 'mp4', entry_protocol='m3u8_native', + format_url, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls', fatal=False)) - continue - tbr = None - mobj = re.search(r'(?P<height>\d+)[pP]?_(?P<tbr>\d+)[kK]', video_url) - if mobj: - if not height: - height = int(mobj.group('height')) - tbr = int(mobj.group('tbr')) + return + if not height: + height = int_or_none(self._search_regex( + r'(?P<height>\d+)[pP]?_\d+[kK]', format_url, 'height', + default=None)) formats.append({ - 'url': video_url, + 'url': format_url, 'format_id': '%dp' % height if height else None, 'height': height, - 'tbr': tbr, }) - self._sort_formats(formats) + + for video_url, height in video_urls: + if not upload_date: + upload_date = self._search_regex( + r'/(\d{6}/\d{2})/', video_url, 'upload data', default=None) + if upload_date: + upload_date = upload_date.replace('/', '') + if '/video/get_media' in video_url: + medias = self._download_json(video_url, video_id, fatal=False) + if isinstance(medias, list): + for media in medias: + if not isinstance(media, dict): + continue + video_url = url_or_none(media.get('videoUrl')) + if not video_url: + continue + height = int_or_none(media.get('quality')) + add_format(video_url, height) + continue + add_format(video_url) + self._sort_formats( + formats, field_preference=('height', 'width', 'fps', 'format_id')) video_uploader = self._html_search_regex( r'(?s)From: .+?<(?:a\b[^>]+\bhref=["\']/(?:(?:user|channel)s|model|pornstar)/|span\b[^>]+\bclass=["\']username)[^>]+>(.+?)<', @@ -405,6 +503,10 @@ class PornHubIE(PornHubBaseIE): class PornHubPlaylistBaseIE(PornHubBaseIE): + def _extract_page(self, url): + return int_or_none(self._search_regex( + r'\bpage=(\d+)', url, 'page', default=None)) + def _extract_entries(self, webpage, host): # Only process container div with main playlist content skipping # drop-down menu that uses similar pattern for videos (see @@ -422,29 +524,9 @@ class PornHubPlaylistBaseIE(PornHubBaseIE): container)) ] - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - host = mobj.group('host') - playlist_id = mobj.group('id') - - webpage = self._download_webpage(url, playlist_id) - - entries = self._extract_entries(webpage, host) - - playlist = self._parse_json( - self._search_regex( - r'(?:playlistObject|PLAYLIST_VIEW)\s*=\s*({.+?});', webpage, - 'playlist', default='{}'), - playlist_id, fatal=False) - title = playlist.get('title') or self._search_regex( - r'>Videos\s+in\s+(.+?)\s+[Pp]laylist<', webpage, 'title', fatal=False) - - return self.playlist_result( - entries, playlist_id, title, playlist.get('description')) - class PornHubUserIE(PornHubPlaylistBaseIE): - _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)' + _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?%s/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/?#&]+))(?:[?#&]|/(?!videos)|$)' % PornHubBaseIE._PORNHUB_HOST_RE _TESTS = [{ 'url': 'https://www.pornhub.com/model/zoe_ph', 'playlist_mincount': 118, @@ -463,14 +545,30 @@ class PornHubUserIE(PornHubPlaylistBaseIE): }, { 'url': 'https://www.pornhub.com/model/zoe_ph?abc=1', 'only_matching': True, + }, { + # Unavailable via /videos page, but available with direct pagination + # on pornstar page (see [1]), requires premium + # 1. https://github.com/ytdl-org/youtube-dl/issues/27853 + 'url': 'https://www.pornhubpremium.com/pornstar/sienna-west', + 'only_matching': True, + }, { + # Same as before, multi page + 'url': 'https://www.pornhubpremium.com/pornstar/lily-labeau', + 'only_matching': True, + }, { + 'url': 'https://pornhubthbh7ap3u.onion/model/zoe_ph', + 'only_matching': True, }] def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) user_id = mobj.group('id') + videos_url = '%s/videos' % mobj.group('url') + page = self._extract_page(url) + if page: + videos_url = update_url_query(videos_url, {'page': page}) return self.url_result( - '%s/videos' % mobj.group('url'), ie=PornHubPagedVideoListIE.ie_key(), - video_id=user_id) + videos_url, ie=PornHubPagedVideoListIE.ie_key(), video_id=user_id) class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE): @@ -483,36 +581,59 @@ class PornHubPagedPlaylistBaseIE(PornHubPlaylistBaseIE): <button[^>]+\bid=["\']moreDataBtn ''', webpage) is not None - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - host = mobj.group('host') - item_id = mobj.group('id') + def _entries(self, url, host, item_id): + page = self._extract_page(url) - page = int_or_none(self._search_regex( - r'\bpage=(\d+)', url, 'page', default=None)) + VIDEOS = '/videos' - entries = [] - for page_num in (page, ) if page is not None else itertools.count(1): + def download_page(base_url, num, fallback=False): + note = 'Downloading page %d%s' % (num, ' (switch to fallback)' if fallback else '') + return self._download_webpage( + base_url, item_id, note, query={'page': num}) + + def is_404(e): + return isinstance(e.cause, compat_HTTPError) and e.cause.code == 404 + + base_url = url + has_page = page is not None + first_page = page if has_page else 1 + for page_num in (first_page, ) if has_page else itertools.count(first_page): try: - webpage = self._download_webpage( - url, item_id, 'Downloading page %d' % page_num, - query={'page': page_num}) + try: + webpage = download_page(base_url, page_num) + except ExtractorError as e: + # Some sources may not be available via /videos page, + # trying to fallback to main page pagination (see [1]) + # 1. https://github.com/ytdl-org/youtube-dl/issues/27853 + if is_404(e) and page_num == first_page and VIDEOS in base_url: + base_url = base_url.replace(VIDEOS, '') + webpage = download_page(base_url, page_num, fallback=True) + else: + raise except ExtractorError as e: - if isinstance(e.cause, compat_HTTPError) and e.cause.code == 404: + if is_404(e) and page_num != first_page: break raise page_entries = self._extract_entries(webpage, host) if not page_entries: break - entries.extend(page_entries) + for e in page_entries: + yield e if not self._has_more(webpage): break - return self.playlist_result(orderedSet(entries), item_id) + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + host = mobj.group('host') + item_id = mobj.group('id') + + self._login(host) + + return self.playlist_result(self._entries(url, host, item_id), item_id) class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE): - _VALID_URL = r'https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?P<id>(?:[^/]+/)*[^/?#&]+)' + _VALID_URL = r'https?://(?:[^/]+\.)?%s/(?P<id>(?:[^/]+/)*[^/?#&]+)' % PornHubBaseIE._PORNHUB_HOST_RE _TESTS = [{ 'url': 'https://www.pornhub.com/model/zoe_ph/videos', 'only_matching': True, @@ -617,6 +738,9 @@ class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE): }, { 'url': 'https://de.pornhub.com/playlist/4667351', 'only_matching': True, + }, { + 'url': 'https://pornhubthbh7ap3u.onion/model/zoe_ph/videos', + 'only_matching': True, }] @classmethod @@ -627,7 +751,7 @@ class PornHubPagedVideoListIE(PornHubPagedPlaylistBaseIE): class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE): - _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?(?P<host>pornhub(?:premium)?\.(?:com|net|org))/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)' + _VALID_URL = r'(?P<url>https?://(?:[^/]+\.)?%s/(?:(?:user|channel)s|model|pornstar)/(?P<id>[^/]+)/videos/upload)' % PornHubBaseIE._PORNHUB_HOST_RE _TESTS = [{ 'url': 'https://www.pornhub.com/pornstar/jenny-blighe/videos/upload', 'info_dict': { @@ -637,4 +761,7 @@ class PornHubUserVideosUploadIE(PornHubPagedPlaylistBaseIE): }, { 'url': 'https://www.pornhub.com/model/zoe_ph/videos/upload', 'only_matching': True, + }, { + 'url': 'http://pornhubthbh7ap3u.onion/pornstar/jenny-blighe/videos/upload', + 'only_matching': True, }] diff --git a/youtube_dl/extractor/rds.py b/youtube_dl/extractor/rds.py index 8c016a77d..0c497856e 100644 --- a/youtube_dl/extractor/rds.py +++ b/youtube_dl/extractor/rds.py @@ -15,17 +15,17 @@ class RDSIE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?rds\.ca/vid(?:[eé]|%C3%A9)os/(?:[^/]+/)*(?P<id>[^/]+)-\d+\.\d+' _TESTS = [{ - 'url': 'http://www.rds.ca/videos/football/nfl/fowler-jr-prend-la-direction-de-jacksonville-3.1132799', + # has two 9c9media ContentPackages, the web player selects the first ContentPackage + 'url': 'https://www.rds.ca/videos/Hockey/NationalHockeyLeague/teams/9/forum-du-5-a-7-jesperi-kotkaniemi-de-retour-de-finlande-3.1377606', 'info_dict': { - 'id': '604333', - 'display_id': 'fowler-jr-prend-la-direction-de-jacksonville', + 'id': '2083309', + 'display_id': 'forum-du-5-a-7-jesperi-kotkaniemi-de-retour-de-finlande', 'ext': 'flv', - 'title': 'Fowler Jr. prend la direction de Jacksonville', - 'description': 'Dante Fowler Jr. est le troisième choix du repêchage 2015 de la NFL. ', - 'timestamp': 1430397346, - 'upload_date': '20150430', - 'duration': 154.354, - 'age_limit': 0, + 'title': 'Forum du 5 à 7 : Kotkaniemi de retour de Finlande', + 'description': 'md5:83fa38ecc4a79b19e433433254077f25', + 'timestamp': 1606129030, + 'upload_date': '20201123', + 'duration': 773.039, } }, { 'url': 'http://www.rds.ca/vid%C3%A9os/un-voyage-positif-3.877934', diff --git a/youtube_dl/extractor/redbulltv.py b/youtube_dl/extractor/redbulltv.py index 3aae79f5d..6d000b372 100644 --- a/youtube_dl/extractor/redbulltv.py +++ b/youtube_dl/extractor/redbulltv.py @@ -133,8 +133,10 @@ class RedBullEmbedIE(RedBullTVIE): rrn_id = self._match_id(url) asset_id = self._download_json( 'https://edge-graphql.crepo-production.redbullaws.com/v1/graphql', - rrn_id, headers={'API-KEY': 'e90a1ff11335423998b100c929ecc866'}, - query={ + rrn_id, headers={ + 'Accept': 'application/json', + 'API-KEY': 'e90a1ff11335423998b100c929ecc866', + }, query={ 'query': '''{ resource(id: "%s", enforceGeoBlocking: false) { %s diff --git a/youtube_dl/extractor/rts.py b/youtube_dl/extractor/rts.py index 48f17b828..aed35f8a9 100644 --- a/youtube_dl/extractor/rts.py +++ b/youtube_dl/extractor/rts.py @@ -6,11 +6,12 @@ import re from .srgssr import SRGSSRIE from ..compat import compat_str from ..utils import ( + determine_ext, int_or_none, parse_duration, parse_iso8601, unescapeHTML, - determine_ext, + urljoin, ) @@ -21,7 +22,7 @@ class RTSIE(SRGSSRIE): _TESTS = [ { 'url': 'http://www.rts.ch/archives/tv/divers/3449373-les-enfants-terribles.html', - 'md5': 'ff7f8450a90cf58dacb64e29707b4a8e', + 'md5': '753b877968ad8afaeddccc374d4256a5', 'info_dict': { 'id': '3449373', 'display_id': 'les-enfants-terribles', @@ -35,6 +36,7 @@ class RTSIE(SRGSSRIE): 'thumbnail': r're:^https?://.*\.image', 'view_count': int, }, + 'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'], }, { 'url': 'http://www.rts.ch/emissions/passe-moi-les-jumelles/5624067-entre-ciel-et-mer.html', @@ -63,11 +65,12 @@ class RTSIE(SRGSSRIE): # m3u8 download 'skip_download': True, }, + 'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'], 'skip': 'Blocked outside Switzerland', }, { 'url': 'http://www.rts.ch/video/info/journal-continu/5745356-londres-cachee-par-un-epais-smog.html', - 'md5': '1bae984fe7b1f78e94abc74e802ed99f', + 'md5': '9bb06503773c07ce83d3cbd793cebb91', 'info_dict': { 'id': '5745356', 'display_id': 'londres-cachee-par-un-epais-smog', @@ -81,6 +84,7 @@ class RTSIE(SRGSSRIE): 'thumbnail': r're:^https?://.*\.image', 'view_count': int, }, + 'expected_warnings': ['Unable to download f4m manifest', 'Failed to download m3u8 information'], }, { 'url': 'http://www.rts.ch/audio/couleur3/programmes/la-belle-video-de-stephane-laurenceau/5706148-urban-hippie-de-damien-krisl-03-04-2014.html', @@ -160,7 +164,7 @@ class RTSIE(SRGSSRIE): media_type = 'video' if 'video' in all_info else 'audio' # check for errors - self.get_media_data('rts', media_type, media_id) + self._get_media_data('rts', media_type, media_id) info = all_info['video']['JSONinfo'] if 'video' in all_info else all_info['audio'] @@ -194,6 +198,7 @@ class RTSIE(SRGSSRIE): 'tbr': extract_bitrate(format_url), }) + download_base = 'http://rtsww%s-d.rts.ch/' % ('-a' if media_type == 'audio' else '') for media in info.get('media', []): media_url = media.get('url') if not media_url or re.match(r'https?://', media_url): @@ -205,7 +210,7 @@ class RTSIE(SRGSSRIE): format_id += '-%dk' % rate formats.append({ 'format_id': format_id, - 'url': 'http://download-video.rts.ch/' + media_url, + 'url': urljoin(download_base, media_url), 'tbr': rate or extract_bitrate(media_url), }) diff --git a/youtube_dl/extractor/rtve.py b/youtube_dl/extractor/rtve.py index ce9db0629..d2fb754cf 100644 --- a/youtube_dl/extractor/rtve.py +++ b/youtube_dl/extractor/rtve.py @@ -2,8 +2,9 @@ from __future__ import unicode_literals import base64 +import io import re -import time +import sys from .common import InfoExtractor from ..compat import ( @@ -14,56 +15,13 @@ from ..utils import ( determine_ext, ExtractorError, float_or_none, + qualities, remove_end, remove_start, - sanitized_Request, std_headers, ) - -def _decrypt_url(png): - encrypted_data = compat_b64decode(png) - text_index = encrypted_data.find(b'tEXt') - text_chunk = encrypted_data[text_index - 4:] - length = compat_struct_unpack('!I', text_chunk[:4])[0] - # Use bytearray to get integers when iterating in both python 2.x and 3.x - data = bytearray(text_chunk[8:8 + length]) - data = [chr(b) for b in data if b != 0] - hash_index = data.index('#') - alphabet_data = data[:hash_index] - url_data = data[hash_index + 1:] - if url_data[0] == 'H' and url_data[3] == '%': - # remove useless HQ%% at the start - url_data = url_data[4:] - - alphabet = [] - e = 0 - d = 0 - for l in alphabet_data: - if d == 0: - alphabet.append(l) - d = e = (e + 1) % 4 - else: - d -= 1 - url = '' - f = 0 - e = 3 - b = 1 - for letter in url_data: - if f == 0: - l = int(letter) * 10 - f = 1 - else: - if e == 0: - l += int(letter) - url += alphabet[l] - e = (b + 3) % 4 - f = 0 - b += 1 - else: - e -= 1 - - return url +_bytes_to_chr = (lambda x: x) if sys.version_info[0] == 2 else (lambda x: map(chr, x)) class RTVEALaCartaIE(InfoExtractor): @@ -79,28 +37,31 @@ class RTVEALaCartaIE(InfoExtractor): 'ext': 'mp4', 'title': 'Balonmano - Swiss Cup masculina. Final: España-Suecia', 'duration': 5024.566, + 'series': 'Balonmano', }, + 'expected_warnings': ['Failed to download MPD manifest', 'Failed to download m3u8 information'], }, { 'note': 'Live stream', 'url': 'http://www.rtve.es/alacarta/videos/television/24h-live/1694255/', 'info_dict': { 'id': '1694255', - 'ext': 'flv', - 'title': 'TODO', + 'ext': 'mp4', + 'title': 're:^24H LIVE [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'is_live': True, + }, + 'params': { + 'skip_download': 'live stream', }, - 'skip': 'The f4m manifest can\'t be used yet', }, { 'url': 'http://www.rtve.es/alacarta/videos/servir-y-proteger/servir-proteger-capitulo-104/4236788/', - 'md5': 'e55e162379ad587e9640eda4f7353c0f', + 'md5': 'd850f3c8731ea53952ebab489cf81cbf', 'info_dict': { 'id': '4236788', 'ext': 'mp4', - 'title': 'Servir y proteger - Capítulo 104 ', + 'title': 'Servir y proteger - Capítulo 104', 'duration': 3222.0, }, - 'params': { - 'skip_download': True, # requires ffmpeg - }, + 'expected_warnings': ['Failed to download MPD manifest', 'Failed to download m3u8 information'], }, { 'url': 'http://www.rtve.es/m/alacarta/videos/cuentame-como-paso/cuentame-como-paso-t16-ultimo-minuto-nuestra-vida-capitulo-276/2969138/?media=tve', 'only_matching': True, @@ -111,58 +72,102 @@ class RTVEALaCartaIE(InfoExtractor): def _real_initialize(self): user_agent_b64 = base64.b64encode(std_headers['User-Agent'].encode('utf-8')).decode('utf-8') - manager_info = self._download_json( + self._manager = self._download_json( 'http://www.rtve.es/odin/loki/' + user_agent_b64, - None, 'Fetching manager info') - self._manager = manager_info['manager'] + None, 'Fetching manager info')['manager'] + + @staticmethod + def _decrypt_url(png): + encrypted_data = io.BytesIO(compat_b64decode(png)[8:]) + while True: + length = compat_struct_unpack('!I', encrypted_data.read(4))[0] + chunk_type = encrypted_data.read(4) + if chunk_type == b'IEND': + break + data = encrypted_data.read(length) + if chunk_type == b'tEXt': + alphabet_data, text = data.split(b'\0') + quality, url_data = text.split(b'%%') + alphabet = [] + e = 0 + d = 0 + for l in _bytes_to_chr(alphabet_data): + if d == 0: + alphabet.append(l) + d = e = (e + 1) % 4 + else: + d -= 1 + url = '' + f = 0 + e = 3 + b = 1 + for letter in _bytes_to_chr(url_data): + if f == 0: + l = int(letter) * 10 + f = 1 + else: + if e == 0: + l += int(letter) + url += alphabet[l] + e = (b + 3) % 4 + f = 0 + b += 1 + else: + e -= 1 + + yield quality.decode(), url + encrypted_data.read(4) # CRC + + def _extract_png_formats(self, video_id): + png = self._download_webpage( + 'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id), + video_id, 'Downloading url information', query={'q': 'v2'}) + q = qualities(['Media', 'Alta', 'HQ', 'HD_READY', 'HD_FULL']) + formats = [] + for quality, video_url in self._decrypt_url(png): + ext = determine_ext(video_url) + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + video_url, video_id, 'mp4', 'm3u8_native', + m3u8_id='hls', fatal=False)) + elif ext == 'mpd': + formats.extend(self._extract_mpd_formats( + video_url, video_id, 'dash', fatal=False)) + else: + formats.append({ + 'format_id': quality, + 'quality': q(quality), + 'url': video_url, + }) + self._sort_formats(formats) + return formats def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') + video_id = self._match_id(url) info = self._download_json( 'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id, video_id)['page']['items'][0] if info['state'] == 'DESPU': raise ExtractorError('The video is no longer available', expected=True) - title = info['title'] - png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/%s/videos/%s.png' % (self._manager, video_id) - png_request = sanitized_Request(png_url) - png_request.add_header('Referer', url) - png = self._download_webpage(png_request, video_id, 'Downloading url information') - video_url = _decrypt_url(png) - ext = determine_ext(video_url) - - formats = [] - if not video_url.endswith('.f4m') and ext != 'm3u8': - if '?' not in video_url: - video_url = video_url.replace('resources/', 'auth/resources/') - video_url = video_url.replace('.net.rtve', '.multimedia.cdn.rtve') - - if ext == 'm3u8': - formats.extend(self._extract_m3u8_formats( - video_url, video_id, ext='mp4', entry_protocol='m3u8_native', - m3u8_id='hls', fatal=False)) - elif ext == 'f4m': - formats.extend(self._extract_f4m_formats( - video_url, video_id, f4m_id='hds', fatal=False)) - else: - formats.append({ - 'url': video_url, - }) - self._sort_formats(formats) + title = info['title'].strip() + formats = self._extract_png_formats(video_id) subtitles = None - if info.get('sbtFile') is not None: - subtitles = self.extract_subtitles(video_id, info['sbtFile']) + sbt_file = info.get('sbtFile') + if sbt_file: + subtitles = self.extract_subtitles(video_id, sbt_file) + + is_live = info.get('live') is True return { 'id': video_id, - 'title': title, + 'title': self._live_title(title) if is_live else title, 'formats': formats, 'thumbnail': info.get('image'), - 'page_url': url, 'subtitles': subtitles, - 'duration': float_or_none(info.get('duration'), scale=1000), + 'duration': float_or_none(info.get('duration'), 1000), + 'is_live': is_live, + 'series': info.get('programTitle'), } def _get_subtitles(self, video_id, sub_file): @@ -174,48 +179,26 @@ class RTVEALaCartaIE(InfoExtractor): for s in subs) -class RTVEInfantilIE(InfoExtractor): +class RTVEInfantilIE(RTVEALaCartaIE): IE_NAME = 'rtve.es:infantil' IE_DESC = 'RTVE infantil' - _VALID_URL = r'https?://(?:www\.)?rtve\.es/infantil/serie/(?P<show>[^/]*)/video/(?P<short_title>[^/]*)/(?P<id>[0-9]+)/' + _VALID_URL = r'https?://(?:www\.)?rtve\.es/infantil/serie/[^/]+/video/[^/]+/(?P<id>[0-9]+)/' _TESTS = [{ 'url': 'http://www.rtve.es/infantil/serie/cleo/video/maneras-vivir/3040283/', - 'md5': '915319587b33720b8e0357caaa6617e6', + 'md5': '5747454717aedf9f9fdf212d1bcfc48d', 'info_dict': { 'id': '3040283', 'ext': 'mp4', 'title': 'Maneras de vivir', - 'thumbnail': 'http://www.rtve.es/resources/jpg/6/5/1426182947956.JPG', + 'thumbnail': r're:https?://.+/1426182947956\.JPG', 'duration': 357.958, }, + 'expected_warnings': ['Failed to download MPD manifest', 'Failed to download m3u8 information'], }] - def _real_extract(self, url): - video_id = self._match_id(url) - info = self._download_json( - 'http://www.rtve.es/api/videos/%s/config/alacarta_videos.json' % video_id, - video_id)['page']['items'][0] - - webpage = self._download_webpage(url, video_id) - vidplayer_id = self._search_regex( - r' id="vidplayer([0-9]+)"', webpage, 'internal video ID') - png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/default/videos/%s.png' % vidplayer_id - png = self._download_webpage(png_url, video_id, 'Downloading url information') - video_url = _decrypt_url(png) - - return { - 'id': video_id, - 'ext': 'mp4', - 'title': info['title'], - 'url': video_url, - 'thumbnail': info.get('image'), - 'duration': float_or_none(info.get('duration'), scale=1000), - } - - -class RTVELiveIE(InfoExtractor): +class RTVELiveIE(RTVEALaCartaIE): IE_NAME = 'rtve.es:live' IE_DESC = 'RTVE.es live streams' _VALID_URL = r'https?://(?:www\.)?rtve\.es/directo/(?P<id>[a-zA-Z0-9-]+)' @@ -225,7 +208,7 @@ class RTVELiveIE(InfoExtractor): 'info_dict': { 'id': 'la-1', 'ext': 'mp4', - 'title': 're:^La 1 [0-9]{4}-[0-9]{2}-[0-9]{2}Z[0-9]{6}$', + 'title': 're:^La 1 [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', }, 'params': { 'skip_download': 'live stream', @@ -234,29 +217,22 @@ class RTVELiveIE(InfoExtractor): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - start_time = time.gmtime() video_id = mobj.group('id') webpage = self._download_webpage(url, video_id) title = remove_end(self._og_search_title(webpage), ' en directo en RTVE.es') title = remove_start(title, 'Estoy viendo ') - title += ' ' + time.strftime('%Y-%m-%dZ%H%M%S', start_time) vidplayer_id = self._search_regex( (r'playerId=player([0-9]+)', r'class=["\'].*?\blive_mod\b.*?["\'][^>]+data-assetid=["\'](\d+)', r'data-id=["\'](\d+)'), webpage, 'internal video ID') - png_url = 'http://www.rtve.es/ztnr/movil/thumbnail/amonet/videos/%s.png' % vidplayer_id - png = self._download_webpage(png_url, video_id, 'Downloading url information') - m3u8_url = _decrypt_url(png) - formats = self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4') - self._sort_formats(formats) return { 'id': video_id, - 'title': title, - 'formats': formats, + 'title': self._live_title(title), + 'formats': self._extract_png_formats(vidplayer_id), 'is_live': True, } diff --git a/youtube_dl/extractor/samplefocus.py b/youtube_dl/extractor/samplefocus.py new file mode 100644 index 000000000..806c3c354 --- /dev/null +++ b/youtube_dl/extractor/samplefocus.py @@ -0,0 +1,100 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + extract_attributes, + get_element_by_attribute, + int_or_none, +) + + +class SampleFocusIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?samplefocus\.com/samples/(?P<id>[^/?&#]+)' + _TESTS = [{ + 'url': 'https://samplefocus.com/samples/lil-peep-sad-emo-guitar', + 'md5': '48c8d62d60be467293912e0e619a5120', + 'info_dict': { + 'id': '40316', + 'display_id': 'lil-peep-sad-emo-guitar', + 'ext': 'mp3', + 'title': 'Lil Peep Sad Emo Guitar', + 'thumbnail': r're:^https?://.+\.png', + 'license': 'Standard License', + 'uploader': 'CapsCtrl', + 'uploader_id': 'capsctrl', + 'like_count': int, + 'comment_count': int, + 'categories': ['Samples', 'Guitar', 'Electric guitar'], + }, + }, { + 'url': 'https://samplefocus.com/samples/dababy-style-bass-808', + 'only_matching': True + }, { + 'url': 'https://samplefocus.com/samples/young-chop-kick', + 'only_matching': True + }] + + def _real_extract(self, url): + display_id = self._match_id(url) + webpage = self._download_webpage(url, display_id) + + sample_id = self._search_regex( + r'<input[^>]+id=(["\'])sample_id\1[^>]+value=(?:["\'])(?P<id>\d+)', + webpage, 'sample id', group='id') + + title = self._og_search_title(webpage, fatal=False) or self._html_search_regex( + r'<h1>(.+?)</h1>', webpage, 'title') + + mp3_url = self._search_regex( + r'<input[^>]+id=(["\'])sample_mp3\1[^>]+value=(["\'])(?P<url>(?:(?!\2).)+)', + webpage, 'mp3', fatal=False, group='url') or extract_attributes(self._search_regex( + r'<meta[^>]+itemprop=(["\'])contentUrl\1[^>]*>', + webpage, 'mp3 url', group=0))['content'] + + thumbnail = self._og_search_thumbnail(webpage) or self._html_search_regex( + r'<img[^>]+class=(?:["\'])waveform responsive-img[^>]+src=(["\'])(?P<url>(?:(?!\1).)+)', + webpage, 'mp3', fatal=False, group='url') + + comments = [] + for author_id, author, body in re.findall(r'(?s)<p[^>]+class="comment-author"><a[^>]+href="/users/([^"]+)">([^"]+)</a>.+?<p[^>]+class="comment-body">([^>]+)</p>', webpage): + comments.append({ + 'author': author, + 'author_id': author_id, + 'text': body, + }) + + uploader_id = uploader = None + mobj = re.search(r'>By <a[^>]+href="/users/([^"]+)"[^>]*>([^<]+)', webpage) + if mobj: + uploader_id, uploader = mobj.groups() + + breadcrumb = get_element_by_attribute('typeof', 'BreadcrumbList', webpage) + categories = [] + if breadcrumb: + for _, name in re.findall(r'<span[^>]+property=(["\'])name\1[^>]*>([^<]+)', breadcrumb): + categories.append(name) + + def extract_count(klass): + return int_or_none(self._html_search_regex( + r'<span[^>]+class=(?:["\'])?%s-count[^>]*>(\d+)' % klass, + webpage, klass, fatal=False)) + + return { + 'id': sample_id, + 'title': title, + 'url': mp3_url, + 'display_id': display_id, + 'thumbnail': thumbnail, + 'uploader': uploader, + 'license': self._html_search_regex( + r'<a[^>]+href=(["\'])/license\1[^>]*>(?P<license>[^<]+)<', + webpage, 'license', fatal=False, group='license'), + 'uploader_id': uploader_id, + 'like_count': extract_count('sample-%s-favorites' % sample_id), + 'comment_count': extract_count('comments'), + 'comments': comments, + 'categories': categories, + } diff --git a/youtube_dl/extractor/sbs.py b/youtube_dl/extractor/sbs.py index f722528cd..0a806ee4e 100644 --- a/youtube_dl/extractor/sbs.py +++ b/youtube_dl/extractor/sbs.py @@ -10,7 +10,7 @@ from ..utils import ( class SBSIE(InfoExtractor): IE_DESC = 'sbs.com.au' - _VALID_URL = r'https?://(?:www\.)?sbs\.com\.au/(?:ondemand(?:/video/(?:single/)?|.*?\bplay=)|news/(?:embeds/)?video/)(?P<id>[0-9]+)' + _VALID_URL = r'https?://(?:www\.)?sbs\.com\.au/(?:ondemand(?:/video/(?:single/)?|.*?\bplay=|/watch/)|news/(?:embeds/)?video/)(?P<id>[0-9]+)' _TESTS = [{ # Original URL is handled by the generic IE which finds the iframe: @@ -43,6 +43,9 @@ class SBSIE(InfoExtractor): }, { 'url': 'https://www.sbs.com.au/news/embeds/video/1840778819866', 'only_matching': True, + }, { + 'url': 'https://www.sbs.com.au/ondemand/watch/1698704451971', + 'only_matching': True, }] def _real_extract(self, url): diff --git a/youtube_dl/extractor/screencastomatic.py b/youtube_dl/extractor/screencastomatic.py index b5e76c9af..0afdc1715 100644 --- a/youtube_dl/extractor/screencastomatic.py +++ b/youtube_dl/extractor/screencastomatic.py @@ -2,12 +2,18 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import js_to_json +from ..utils import ( + get_element_by_class, + int_or_none, + remove_start, + strip_or_none, + unified_strdate, +) class ScreencastOMaticIE(InfoExtractor): - _VALID_URL = r'https?://screencast-o-matic\.com/watch/(?P<id>[0-9a-zA-Z]+)' - _TEST = { + _VALID_URL = r'https?://screencast-o-matic\.com/(?:(?:watch|player)/|embed\?.*?\bsc=)(?P<id>[0-9a-zA-Z]+)' + _TESTS = [{ 'url': 'http://screencast-o-matic.com/watch/c2lD3BeOPl', 'md5': '483583cb80d92588f15ccbedd90f0c18', 'info_dict': { @@ -16,22 +22,30 @@ class ScreencastOMaticIE(InfoExtractor): 'title': 'Welcome to 3-4 Philosophy @ DECV!', 'thumbnail': r're:^https?://.*\.jpg$', 'description': 'as the title says! also: some general info re 1) VCE philosophy and 2) distance learning.', - 'duration': 369.163, + 'duration': 369, + 'upload_date': '20141216', } - } + }, { + 'url': 'http://screencast-o-matic.com/player/c2lD3BeOPl', + 'only_matching': True, + }, { + 'url': 'http://screencast-o-matic.com/embed?ff=true&sc=cbV2r4Q5TL&fromPH=true&a=1', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) - webpage = self._download_webpage(url, video_id) - - jwplayer_data = self._parse_json( - self._search_regex( - r"(?s)jwplayer\('mp4Player'\).setup\((\{.*?\})\);", webpage, 'setup code'), - video_id, transform_source=js_to_json) - - info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False) - info_dict.update({ - 'title': self._og_search_title(webpage), - 'description': self._og_search_description(webpage), + webpage = self._download_webpage( + 'https://screencast-o-matic.com/player/' + video_id, video_id) + info = self._parse_html5_media_entries(url, webpage, video_id)[0] + info.update({ + 'id': video_id, + 'title': get_element_by_class('overlayTitle', webpage), + 'description': strip_or_none(get_element_by_class('overlayDescription', webpage)) or None, + 'duration': int_or_none(self._search_regex( + r'player\.duration\s*=\s*function\(\)\s*{\s*return\s+(\d+);\s*};', + webpage, 'duration', default=None)), + 'upload_date': unified_strdate(remove_start( + get_element_by_class('overlayPublished', webpage), 'Published: ')), }) - return info_dict + return info diff --git a/youtube_dl/extractor/shahid.py b/youtube_dl/extractor/shahid.py index 5c2a6206b..88b938e05 100644 --- a/youtube_dl/extractor/shahid.py +++ b/youtube_dl/extractor/shahid.py @@ -21,6 +21,7 @@ from ..utils import ( class ShahidBaseIE(AWSIE): _AWS_PROXY_HOST = 'api2.shahid.net' _AWS_API_KEY = '2RRtuMHx95aNI1Kvtn2rChEuwsCogUd4samGPjLh' + _VALID_URL_BASE = r'https?://shahid\.mbc\.net/[a-z]{2}/' def _handle_error(self, e): fail_data = self._parse_json( @@ -49,15 +50,18 @@ class ShahidBaseIE(AWSIE): class ShahidIE(ShahidBaseIE): _NETRC_MACHINE = 'shahid' - _VALID_URL = r'https?://shahid\.mbc\.net/ar/(?:serie|show|movie)s/[^/]+/(?P<type>episode|clip|movie)-(?P<id>\d+)' + _VALID_URL = ShahidBaseIE._VALID_URL_BASE + r'(?:serie|show|movie)s/[^/]+/(?P<type>episode|clip|movie)-(?P<id>\d+)' _TESTS = [{ - 'url': 'https://shahid.mbc.net/ar/shows/%D9%85%D8%AC%D9%84%D8%B3-%D8%A7%D9%84%D8%B4%D8%A8%D8%A7%D8%A8-%D8%A7%D9%84%D9%85%D9%88%D8%B3%D9%85-1-%D9%83%D9%84%D9%8A%D8%A8-1/clip-275286', + 'url': 'https://shahid.mbc.net/ar/shows/%D9%85%D8%AA%D8%AD%D9%81-%D8%A7%D9%84%D8%AF%D8%AD%D9%8A%D8%AD-%D8%A7%D9%84%D9%85%D9%88%D8%B3%D9%85-1-%D9%83%D9%84%D9%8A%D8%A8-1/clip-816924', 'info_dict': { - 'id': '275286', + 'id': '816924', 'ext': 'mp4', - 'title': 'مجلس الشباب الموسم 1 كليب 1', - 'timestamp': 1506988800, - 'upload_date': '20171003', + 'title': 'متحف الدحيح الموسم 1 كليب 1', + 'timestamp': 1602806400, + 'upload_date': '20201016', + 'description': 'برومو', + 'duration': 22, + 'categories': ['كوميديا'], }, 'params': { # m3u8 download @@ -70,6 +74,9 @@ class ShahidIE(ShahidBaseIE): # shahid plus subscriber only 'url': 'https://shahid.mbc.net/ar/series/%D9%85%D8%B1%D8%A7%D9%8A%D8%A7-2011-%D8%A7%D9%84%D9%85%D9%88%D8%B3%D9%85-1-%D8%A7%D9%84%D8%AD%D9%84%D9%82%D8%A9-1/episode-90511', 'only_matching': True + }, { + 'url': 'https://shahid.mbc.net/en/shows/Ramez-Fi-Al-Shallal-season-1-episode-1/episode-359319', + 'only_matching': True }] def _real_initialize(self): @@ -109,12 +116,15 @@ class ShahidIE(ShahidBaseIE): page_type = 'episode' playout = self._call_api( - 'playout/url/' + video_id, video_id)['playout'] + 'playout/new/url/' + video_id, video_id)['playout'] if playout.get('drm'): raise ExtractorError('This video is DRM protected.', expected=True) - formats = self._extract_m3u8_formats(playout['url'], video_id, 'mp4') + formats = self._extract_m3u8_formats(re.sub( + # https://docs.aws.amazon.com/mediapackage/latest/ug/manifest-filtering.html + r'aws\.manifestfilter=[\w:;,-]+&?', + '', playout['url']), video_id, 'mp4') self._sort_formats(formats) # video = self._call_api( @@ -162,7 +172,7 @@ class ShahidIE(ShahidBaseIE): class ShahidShowIE(ShahidBaseIE): - _VALID_URL = r'https?://shahid\.mbc\.net/ar/(?:show|serie)s/[^/]+/(?:show|series)-(?P<id>\d+)' + _VALID_URL = ShahidBaseIE._VALID_URL_BASE + r'(?:show|serie)s/[^/]+/(?:show|series)-(?P<id>\d+)' _TESTS = [{ 'url': 'https://shahid.mbc.net/ar/shows/%D8%B1%D8%A7%D9%85%D8%B2-%D9%82%D8%B1%D8%B4-%D8%A7%D9%84%D8%A8%D8%AD%D8%B1/show-79187', 'info_dict': { diff --git a/youtube_dl/extractor/shared.py b/youtube_dl/extractor/shared.py index 02295d1a4..93ab2a167 100644 --- a/youtube_dl/extractor/shared.py +++ b/youtube_dl/extractor/shared.py @@ -86,10 +86,10 @@ class SharedIE(SharedBaseIE): class VivoIE(SharedBaseIE): IE_DESC = 'vivo.sx' - _VALID_URL = r'https?://vivo\.sx/(?P<id>[\da-z]{10})' + _VALID_URL = r'https?://vivo\.s[xt]/(?P<id>[\da-z]{10})' _FILE_NOT_FOUND = '>The file you have requested does not exists or has been removed' - _TEST = { + _TESTS = [{ 'url': 'http://vivo.sx/d7ddda0e78', 'md5': '15b3af41be0b4fe01f4df075c2678b2c', 'info_dict': { @@ -98,7 +98,10 @@ class VivoIE(SharedBaseIE): 'title': 'Chicken', 'filesize': 515659, }, - } + }, { + 'url': 'http://vivo.st/d7ddda0e78', + 'only_matching': True, + }] def _extract_title(self, webpage): title = self._html_search_regex( diff --git a/youtube_dl/extractor/simplecast.py b/youtube_dl/extractor/simplecast.py new file mode 100644 index 000000000..2d0b3c06d --- /dev/null +++ b/youtube_dl/extractor/simplecast.py @@ -0,0 +1,160 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + clean_podcast_url, + int_or_none, + parse_iso8601, + strip_or_none, + try_get, + urlencode_postdata, +) + + +class SimplecastBaseIE(InfoExtractor): + _UUID_REGEX = r'[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12}' + _API_BASE = 'https://api.simplecast.com/' + + def _call_api(self, path_tmpl, video_id): + return self._download_json( + self._API_BASE + path_tmpl % video_id, video_id) + + def _call_search_api(self, resource, resource_id, resource_url): + return self._download_json( + 'https://api.simplecast.com/%ss/search' % resource, resource_id, + data=urlencode_postdata({'url': resource_url})) + + def _parse_episode(self, episode): + episode_id = episode['id'] + title = episode['title'].strip() + audio_file = episode.get('audio_file') or {} + audio_file_url = audio_file.get('url') or episode.get('audio_file_url') or episode['enclosure_url'] + + season = episode.get('season') or {} + season_href = season.get('href') + season_id = None + if season_href: + season_id = self._search_regex( + r'https?://api.simplecast.com/seasons/(%s)' % self._UUID_REGEX, + season_href, 'season id', default=None) + + webpage_url = episode.get('episode_url') + channel_url = None + if webpage_url: + channel_url = self._search_regex( + r'(https?://[^/]+\.simplecast\.com)', + webpage_url, 'channel url', default=None) + + return { + 'id': episode_id, + 'display_id': episode.get('slug'), + 'title': title, + 'url': clean_podcast_url(audio_file_url), + 'webpage_url': webpage_url, + 'channel_url': channel_url, + 'series': try_get(episode, lambda x: x['podcast']['title']), + 'season_number': int_or_none(season.get('number')), + 'season_id': season_id, + 'thumbnail': episode.get('image_url'), + 'episode_id': episode_id, + 'episode_number': int_or_none(episode.get('number')), + 'description': strip_or_none(episode.get('description')), + 'timestamp': parse_iso8601(episode.get('published_at')), + 'duration': int_or_none(episode.get('duration')), + 'filesize': int_or_none(audio_file.get('size') or episode.get('audio_file_size')), + } + + +class SimplecastIE(SimplecastBaseIE): + IE_NAME = 'simplecast' + _VALID_URL = r'https?://(?:api\.simplecast\.com/episodes|player\.simplecast\.com)/(?P<id>%s)' % SimplecastBaseIE._UUID_REGEX + _COMMON_TEST_INFO = { + 'display_id': 'errant-signal-chris-franklin-new-wave-video-essays', + 'id': 'b6dc49a2-9404-4853-9aa9-9cfc097be876', + 'ext': 'mp3', + 'title': 'Errant Signal - Chris Franklin & New Wave Video Essays', + 'episode_number': 1, + 'episode_id': 'b6dc49a2-9404-4853-9aa9-9cfc097be876', + 'description': 'md5:34752789d3d2702e2d2c975fbd14f357', + 'season_number': 1, + 'season_id': 'e23df0da-bae4-4531-8bbf-71364a88dc13', + 'series': 'The RE:BIND.io Podcast', + 'duration': 5343, + 'timestamp': 1580979475, + 'upload_date': '20200206', + 'webpage_url': r're:^https?://the-re-bind-io-podcast\.simplecast\.com/episodes/errant-signal-chris-franklin-new-wave-video-essays', + 'channel_url': r're:^https?://the-re-bind-io-podcast\.simplecast\.com$', + } + _TESTS = [{ + 'url': 'https://api.simplecast.com/episodes/b6dc49a2-9404-4853-9aa9-9cfc097be876', + 'md5': '8c93be7be54251bf29ee97464eabe61c', + 'info_dict': _COMMON_TEST_INFO, + }, { + 'url': 'https://player.simplecast.com/b6dc49a2-9404-4853-9aa9-9cfc097be876', + 'only_matching': True, + }] + + @staticmethod + def _extract_urls(webpage): + return re.findall( + r'''(?x)<iframe[^>]+src=["\'] + ( + https?://(?:embed\.simplecast\.com/[0-9a-f]{8}| + player\.simplecast\.com/%s + ))''' % SimplecastBaseIE._UUID_REGEX, webpage) + + def _real_extract(self, url): + episode_id = self._match_id(url) + episode = self._call_api('episodes/%s', episode_id) + return self._parse_episode(episode) + + +class SimplecastEpisodeIE(SimplecastBaseIE): + IE_NAME = 'simplecast:episode' + _VALID_URL = r'https?://(?!api\.)[^/]+\.simplecast\.com/episodes/(?P<id>[^/?&#]+)' + _TEST = { + 'url': 'https://the-re-bind-io-podcast.simplecast.com/episodes/errant-signal-chris-franklin-new-wave-video-essays', + 'md5': '8c93be7be54251bf29ee97464eabe61c', + 'info_dict': SimplecastIE._COMMON_TEST_INFO, + } + + def _real_extract(self, url): + mobj = re.match(self._VALID_URL, url) + episode = self._call_search_api( + 'episode', mobj.group(1), mobj.group(0)) + return self._parse_episode(episode) + + +class SimplecastPodcastIE(SimplecastBaseIE): + IE_NAME = 'simplecast:podcast' + _VALID_URL = r'https?://(?!(?:api|cdn|embed|feeds|player)\.)(?P<id>[^/]+)\.simplecast\.com(?!/episodes/[^/?&#]+)' + _TESTS = [{ + 'url': 'https://the-re-bind-io-podcast.simplecast.com', + 'playlist_mincount': 33, + 'info_dict': { + 'id': '07d28d26-7522-42eb-8c53-2bdcfc81c43c', + 'title': 'The RE:BIND.io Podcast', + }, + }, { + 'url': 'https://the-re-bind-io-podcast.simplecast.com/episodes', + 'only_matching': True, + }] + + def _real_extract(self, url): + subdomain = self._match_id(url) + site = self._call_search_api('site', subdomain, url) + podcast = site['podcast'] + podcast_id = podcast['id'] + podcast_title = podcast.get('title') + + def entries(): + episodes = self._call_api('podcasts/%s/episodes', podcast_id) + for episode in (episodes.get('collection') or []): + info = self._parse_episode(episode) + info['series'] = podcast_title + yield info + + return self.playlist_result(entries(), podcast_id, podcast_title) diff --git a/youtube_dl/extractor/southpark.py b/youtube_dl/extractor/southpark.py index da75a43a7..0774da06e 100644 --- a/youtube_dl/extractor/southpark.py +++ b/youtube_dl/extractor/southpark.py @@ -6,9 +6,9 @@ from .mtv import MTVServicesInfoExtractor class SouthParkIE(MTVServicesInfoExtractor): IE_NAME = 'southpark.cc.com' - _VALID_URL = r'https?://(?:www\.)?(?P<url>southpark\.cc\.com/(?:clips|(?:full-)?episodes|collections)/(?P<id>.+?)(\?|#|$))' + _VALID_URL = r'https?://(?:www\.)?(?P<url>southpark(?:\.cc|studios)\.com/(?:clips|(?:full-)?episodes|collections)/(?P<id>.+?)(\?|#|$))' - _FEED_URL = 'http://www.southparkstudios.com/feeds/video-player/mrss' + _FEED_URL = 'http://feeds.mtvnservices.com/od/feed/intl-mrss-player-feed' _TESTS = [{ 'url': 'http://southpark.cc.com/clips/104437/bat-daded#tab=featured', @@ -23,8 +23,20 @@ class SouthParkIE(MTVServicesInfoExtractor): }, { 'url': 'http://southpark.cc.com/collections/7758/fan-favorites/1', 'only_matching': True, + }, { + 'url': 'https://www.southparkstudios.com/episodes/h4o269/south-park-stunning-and-brave-season-19-ep-1', + 'only_matching': True, }] + def _get_feed_query(self, uri): + return { + 'accountOverride': 'intl.mtvi.com', + 'arcEp': 'shared.southpark.global', + 'ep': '90877963', + 'imageEp': 'shared.southpark.global', + 'mgid': uri, + } + class SouthParkEsIE(SouthParkIE): IE_NAME = 'southpark.cc.com:español' diff --git a/youtube_dl/extractor/spike.py b/youtube_dl/extractor/spike.py index 4c5e3f7c2..5805f3d44 100644 --- a/youtube_dl/extractor/spike.py +++ b/youtube_dl/extractor/spike.py @@ -20,9 +20,6 @@ class BellatorIE(MTVServicesInfoExtractor): _FEED_URL = 'http://www.bellator.com/feeds/mrss/' _GEO_COUNTRIES = ['US'] - def _extract_mgid(self, webpage): - return self._extract_triforce_mgid(webpage) - class ParamountNetworkIE(MTVServicesInfoExtractor): _VALID_URL = r'https?://(?:www\.)?paramountnetwork\.com/[^/]+/[\da-z]{6}(?:[/?#&]|$)' @@ -46,16 +43,6 @@ class ParamountNetworkIE(MTVServicesInfoExtractor): def _get_feed_query(self, uri): return { 'arcEp': 'paramountnetwork.com', + 'imageEp': 'paramountnetwork.com', 'mgid': uri, } - - def _extract_mgid(self, webpage): - root_data = self._parse_json(self._search_regex( - r'window\.__DATA__\s*=\s*({.+})', - webpage, 'data'), None) - - def find_sub_data(data, data_type): - return next(c for c in data['children'] if c.get('type') == data_type) - - c = find_sub_data(find_sub_data(root_data, 'MainContainer'), 'VideoPlayer') - return c['props']['media']['video']['config']['uri'] diff --git a/youtube_dl/extractor/sportdeutschland.py b/youtube_dl/extractor/sportdeutschland.py index 378fc7568..3e497a939 100644 --- a/youtube_dl/extractor/sportdeutschland.py +++ b/youtube_dl/extractor/sportdeutschland.py @@ -1,82 +1,105 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor +from ..compat import ( + compat_parse_qs, + compat_urllib_parse_urlparse, +) from ..utils import ( + clean_html, + float_or_none, + int_or_none, parse_iso8601, - sanitized_Request, + strip_or_none, + try_get, ) class SportDeutschlandIE(InfoExtractor): - _VALID_URL = r'https?://sportdeutschland\.tv/(?P<sport>[^/?#]+)/(?P<id>[^?#/]+)(?:$|[?#])' + _VALID_URL = r'https?://sportdeutschland\.tv/(?P<id>(?:[^/]+/)?[^?#/&]+)' _TESTS = [{ 'url': 'https://sportdeutschland.tv/badminton/re-live-deutsche-meisterschaften-2020-halbfinals?playlistId=0', 'info_dict': { - 'id': 're-live-deutsche-meisterschaften-2020-halbfinals', + 'id': '5318cac0275701382770543d7edaf0a0', 'ext': 'mp4', - 'title': 're:Re-live: Deutsche Meisterschaften 2020.*Halbfinals', - 'categories': ['Badminton-Deutschland'], - 'view_count': int, - 'thumbnail': r're:^https?://.*\.(?:jpg|png)$', - 'timestamp': int, - 'upload_date': '20200201', - 'description': 're:.*', # meaningless description for THIS video + 'title': 'Re-live: Deutsche Meisterschaften 2020 - Halbfinals - Teil 1', + 'duration': 16106.36, }, + 'params': { + 'noplaylist': True, + # m3u8 download + 'skip_download': True, + }, + }, { + 'url': 'https://sportdeutschland.tv/badminton/re-live-deutsche-meisterschaften-2020-halbfinals?playlistId=0', + 'info_dict': { + 'id': 'c6e2fdd01f63013854c47054d2ab776f', + 'title': 'Re-live: Deutsche Meisterschaften 2020 - Halbfinals', + 'description': 'md5:5263ff4c31c04bb780c9f91130b48530', + 'duration': 31397, + }, + 'playlist_count': 2, + }, { + 'url': 'https://sportdeutschland.tv/freeride-world-tour-2021-fieberbrunn-oesterreich', + 'only_matching': True, }] def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id = mobj.group('id') - sport_id = mobj.group('sport') - - api_url = 'https://proxy.vidibusdynamic.net/ssl/backend.sportdeutschland.tv/api/permalinks/%s/%s?access_token=true' % ( - sport_id, video_id) - req = sanitized_Request(api_url, headers={ - 'Accept': 'application/vnd.vidibus.v2.html+json', - 'Referer': url, - }) - data = self._download_json(req, video_id) - + display_id = self._match_id(url) + data = self._download_json( + 'https://backend.sportdeutschland.tv/api/permalinks/' + display_id, + display_id, query={'access_token': 'true'}) asset = data['asset'] - categories = [data['section']['title']] - - formats = [] - smil_url = asset['video'] - if '.smil' in smil_url: - m3u8_url = smil_url.replace('.smil', '.m3u8') - formats.extend( - self._extract_m3u8_formats(m3u8_url, video_id, ext='mp4')) + title = (asset.get('title') or asset['label']).strip() + asset_id = asset.get('id') or asset.get('uuid') + info = { + 'id': asset_id, + 'title': title, + 'description': clean_html(asset.get('body') or asset.get('description')) or asset.get('teaser'), + 'duration': int_or_none(asset.get('seconds')), + } + videos = asset.get('videos') or [] + if len(videos) > 1: + playlist_id = compat_parse_qs(compat_urllib_parse_urlparse(url).query).get('playlistId', [None])[0] + if playlist_id: + if self._downloader.params.get('noplaylist'): + videos = [videos[int(playlist_id)]] + self.to_screen('Downloading just a single video because of --no-playlist') + else: + self.to_screen('Downloading playlist %s - add --no-playlist to just download video' % asset_id) - smil_doc = self._download_xml( - smil_url, video_id, note='Downloading SMIL metadata') - base_url_el = smil_doc.find('./head/meta') - if base_url_el: - base_url = base_url_el.attrib['base'] - formats.extend([{ - 'format_id': 'rmtp', - 'url': base_url if base_url_el else n.attrib['src'], - 'play_path': n.attrib['src'], - 'ext': 'flv', - 'preference': -100, - 'format_note': 'Seems to fail at example stream', - } for n in smil_doc.findall('./body/video')]) + def entries(): + for i, video in enumerate(videos, 1): + video_id = video.get('uuid') + video_url = video.get('url') + if not (video_id and video_url): + continue + formats = self._extract_m3u8_formats( + video_url.replace('.smil', '.m3u8'), video_id, 'mp4', fatal=False) + if not formats: + continue + yield { + 'id': video_id, + 'formats': formats, + 'title': title + ' - ' + (video.get('label') or 'Teil %d' % i), + 'duration': float_or_none(video.get('duration')), + } + info.update({ + '_type': 'multi_video', + 'entries': entries(), + }) else: - formats.append({'url': smil_url}) - - self._sort_formats(formats) - - return { - 'id': video_id, - 'formats': formats, - 'title': asset['title'], - 'thumbnail': asset.get('image'), - 'description': asset.get('teaser'), - 'duration': asset.get('duration'), - 'categories': categories, - 'view_count': asset.get('views'), - 'rtmp_live': asset.get('live'), - 'timestamp': parse_iso8601(asset.get('date')), - } + formats = self._extract_m3u8_formats( + videos[0]['url'].replace('.smil', '.m3u8'), asset_id, 'mp4') + section_title = strip_or_none(try_get(data, lambda x: x['section']['title'])) + info.update({ + 'formats': formats, + 'display_id': asset.get('permalink'), + 'thumbnail': try_get(asset, lambda x: x['images'][0]), + 'categories': [section_title] if section_title else None, + 'view_count': int_or_none(asset.get('views')), + 'is_live': asset.get('is_live') is True, + 'timestamp': parse_iso8601(asset.get('date') or asset.get('published_at')), + }) + return info diff --git a/youtube_dl/extractor/spotify.py b/youtube_dl/extractor/spotify.py new file mode 100644 index 000000000..826f98cff --- /dev/null +++ b/youtube_dl/extractor/spotify.py @@ -0,0 +1,156 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import json +import re + +from .common import InfoExtractor +from ..utils import ( + clean_podcast_url, + float_or_none, + int_or_none, + strip_or_none, + try_get, + unified_strdate, +) + + +class SpotifyBaseIE(InfoExtractor): + _ACCESS_TOKEN = None + _OPERATION_HASHES = { + 'Episode': '8276d4423d709ae9b68ec1b74cc047ba0f7479059a37820be730f125189ac2bf', + 'MinimalShow': '13ee079672fad3f858ea45a55eb109553b4fb0969ed793185b2e34cbb6ee7cc0', + 'ShowEpisodes': 'e0e5ce27bd7748d2c59b4d44ba245a8992a05be75d6fabc3b20753fc8857444d', + } + _VALID_URL_TEMPL = r'https?://open\.spotify\.com/%s/(?P<id>[^/?&#]+)' + + def _real_initialize(self): + self._ACCESS_TOKEN = self._download_json( + 'https://open.spotify.com/get_access_token', None)['accessToken'] + + def _call_api(self, operation, video_id, variables): + return self._download_json( + 'https://api-partner.spotify.com/pathfinder/v1/query', video_id, query={ + 'operationName': 'query' + operation, + 'variables': json.dumps(variables), + 'extensions': json.dumps({ + 'persistedQuery': { + 'sha256Hash': self._OPERATION_HASHES[operation], + }, + }) + }, headers={'authorization': 'Bearer ' + self._ACCESS_TOKEN})['data'] + + def _extract_episode(self, episode, series): + episode_id = episode['id'] + title = episode['name'].strip() + + formats = [] + audio_preview = episode.get('audioPreview') or {} + audio_preview_url = audio_preview.get('url') + if audio_preview_url: + f = { + 'url': audio_preview_url.replace('://p.scdn.co/mp3-preview/', '://anon-podcast.scdn.co/'), + 'vcodec': 'none', + } + audio_preview_format = audio_preview.get('format') + if audio_preview_format: + f['format_id'] = audio_preview_format + mobj = re.match(r'([0-9A-Z]{3})_(?:[A-Z]+_)?(\d+)', audio_preview_format) + if mobj: + f.update({ + 'abr': int(mobj.group(2)), + 'ext': mobj.group(1).lower(), + }) + formats.append(f) + + for item in (try_get(episode, lambda x: x['audio']['items']) or []): + item_url = item.get('url') + if not (item_url and item.get('externallyHosted')): + continue + formats.append({ + 'url': clean_podcast_url(item_url), + 'vcodec': 'none', + }) + + thumbnails = [] + for source in (try_get(episode, lambda x: x['coverArt']['sources']) or []): + source_url = source.get('url') + if not source_url: + continue + thumbnails.append({ + 'url': source_url, + 'width': int_or_none(source.get('width')), + 'height': int_or_none(source.get('height')), + }) + + return { + 'id': episode_id, + 'title': title, + 'formats': formats, + 'thumbnails': thumbnails, + 'description': strip_or_none(episode.get('description')), + 'duration': float_or_none(try_get( + episode, lambda x: x['duration']['totalMilliseconds']), 1000), + 'release_date': unified_strdate(try_get( + episode, lambda x: x['releaseDate']['isoString'])), + 'series': series, + } + + +class SpotifyIE(SpotifyBaseIE): + IE_NAME = 'spotify' + _VALID_URL = SpotifyBaseIE._VALID_URL_TEMPL % 'episode' + _TEST = { + 'url': 'https://open.spotify.com/episode/4Z7GAJ50bgctf6uclHlWKo', + 'md5': '74010a1e3fa4d9e1ab3aa7ad14e42d3b', + 'info_dict': { + 'id': '4Z7GAJ50bgctf6uclHlWKo', + 'ext': 'mp3', + 'title': 'From the archive: Why time management is ruining our lives', + 'description': 'md5:b120d9c4ff4135b42aa9b6d9cde86935', + 'duration': 2083.605, + 'release_date': '20201217', + 'series': "The Guardian's Audio Long Reads", + } + } + + def _real_extract(self, url): + episode_id = self._match_id(url) + episode = self._call_api('Episode', episode_id, { + 'uri': 'spotify:episode:' + episode_id + })['episode'] + return self._extract_episode( + episode, try_get(episode, lambda x: x['podcast']['name'])) + + +class SpotifyShowIE(SpotifyBaseIE): + IE_NAME = 'spotify:show' + _VALID_URL = SpotifyBaseIE._VALID_URL_TEMPL % 'show' + _TEST = { + 'url': 'https://open.spotify.com/show/4PM9Ke6l66IRNpottHKV9M', + 'info_dict': { + 'id': '4PM9Ke6l66IRNpottHKV9M', + 'title': 'The Story from the Guardian', + 'description': 'The Story podcast is dedicated to our finest audio documentaries, investigations and long form stories', + }, + 'playlist_mincount': 36, + } + + def _real_extract(self, url): + show_id = self._match_id(url) + podcast = self._call_api('ShowEpisodes', show_id, { + 'limit': 1000000000, + 'offset': 0, + 'uri': 'spotify:show:' + show_id, + })['podcast'] + podcast_name = podcast.get('name') + + entries = [] + for item in (try_get(podcast, lambda x: x['episodes']['items']) or []): + episode = item.get('episode') + if not episode: + continue + entries.append(self._extract_episode(episode, podcast_name)) + + return self.playlist_result( + entries, show_id, podcast_name, podcast.get('description')) diff --git a/youtube_dl/extractor/srgssr.py b/youtube_dl/extractor/srgssr.py index f63a1359a..ac018e740 100644 --- a/youtube_dl/extractor/srgssr.py +++ b/youtube_dl/extractor/srgssr.py @@ -4,16 +4,32 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_urllib_parse_urlparse from ..utils import ( ExtractorError, + float_or_none, + int_or_none, parse_iso8601, qualities, + try_get, ) class SRGSSRIE(InfoExtractor): - _VALID_URL = r'(?:https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=urn|srgssr):(?P<bu>srf|rts|rsi|rtr|swi):(?:[^:]+:)?(?P<type>video|audio):(?P<id>[0-9a-f\-]{36}|\d+)' + _VALID_URL = r'''(?x) + (?: + https?://tp\.srgssr\.ch/p(?:/[^/]+)+\?urn=urn| + srgssr + ): + (?P<bu> + srf|rts|rsi|rtr|swi + ):(?:[^:]+:)? + (?P<type> + video|audio + ): + (?P<id> + [0-9a-f\-]{36}|\d+ + ) + ''' _GEO_BYPASS = False _GEO_COUNTRIES = ['CH'] @@ -25,25 +41,39 @@ class SRGSSRIE(InfoExtractor): 'LEGAL': 'The video cannot be transmitted for legal reasons.', 'STARTDATE': 'This video is not yet available. Please try again later.', } + _DEFAULT_LANGUAGE_CODES = { + 'srf': 'de', + 'rts': 'fr', + 'rsi': 'it', + 'rtr': 'rm', + 'swi': 'en', + } def _get_tokenized_src(self, url, video_id, format_id): - sp = compat_urllib_parse_urlparse(url).path.split('/') token = self._download_json( - 'http://tp.srgssr.ch/akahd/token?acl=/%s/%s/*' % (sp[1], sp[2]), + 'http://tp.srgssr.ch/akahd/token?acl=*', video_id, 'Downloading %s token' % format_id, fatal=False) or {} - auth_params = token.get('token', {}).get('authparams') + auth_params = try_get(token, lambda x: x['token']['authparams']) if auth_params: - url += '?' + auth_params + url += ('?' if '?' not in url else '&') + auth_params return url - def get_media_data(self, bu, media_type, media_id): - media_data = self._download_json( - 'http://il.srgssr.ch/integrationlayer/1.0/ue/%s/%s/play/%s.json' % (bu, media_type, media_id), - media_id)[media_type.capitalize()] - - if media_data.get('block') and media_data['block'] in self._ERRORS: - message = self._ERRORS[media_data['block']] - if media_data['block'] == 'GEOBLOCK': + def _get_media_data(self, bu, media_type, media_id): + query = {'onlyChapters': True} if media_type == 'video' else {} + full_media_data = self._download_json( + 'https://il.srgssr.ch/integrationlayer/2.0/%s/mediaComposition/%s/%s.json' + % (bu, media_type, media_id), + media_id, query=query)['chapterList'] + try: + media_data = next( + x for x in full_media_data if x.get('id') == media_id) + except StopIteration: + raise ExtractorError('No media information found') + + block_reason = media_data.get('blockReason') + if block_reason and block_reason in self._ERRORS: + message = self._ERRORS[block_reason] + if block_reason == 'GEOBLOCK': self.raise_geo_restricted( msg=message, countries=self._GEO_COUNTRIES) raise ExtractorError( @@ -53,53 +83,75 @@ class SRGSSRIE(InfoExtractor): def _real_extract(self, url): bu, media_type, media_id = re.match(self._VALID_URL, url).groups() + media_data = self._get_media_data(bu, media_type, media_id) + title = media_data['title'] - media_data = self.get_media_data(bu, media_type, media_id) - - metadata = media_data['AssetMetadatas']['AssetMetadata'][0] - title = metadata['title'] - description = metadata.get('description') - created_date = media_data.get('createdDate') or metadata.get('createdDate') - timestamp = parse_iso8601(created_date) - - thumbnails = [{ - 'id': image.get('id'), - 'url': image['url'], - } for image in media_data.get('Image', {}).get('ImageRepresentations', {}).get('ImageRepresentation', [])] - - preference = qualities(['LQ', 'MQ', 'SD', 'HQ', 'HD']) formats = [] - for source in media_data.get('Playlists', {}).get('Playlist', []) + media_data.get('Downloads', {}).get('Download', []): - protocol = source.get('@protocol') - for asset in source['url']: - asset_url = asset['text'] - quality = asset['@quality'] - format_id = '%s-%s' % (protocol, quality) - if protocol.startswith('HTTP-HDS') or protocol.startswith('HTTP-HLS'): - asset_url = self._get_tokenized_src(asset_url, media_id, format_id) - if protocol.startswith('HTTP-HDS'): - formats.extend(self._extract_f4m_formats( - asset_url + ('?' if '?' not in asset_url else '&') + 'hdcore=3.4.0', - media_id, f4m_id=format_id, fatal=False)) - elif protocol.startswith('HTTP-HLS'): - formats.extend(self._extract_m3u8_formats( - asset_url, media_id, 'mp4', 'm3u8_native', - m3u8_id=format_id, fatal=False)) - else: - formats.append({ - 'format_id': format_id, - 'url': asset_url, - 'preference': preference(quality), - 'ext': 'flv' if protocol == 'RTMP' else None, - }) + q = qualities(['SD', 'HD']) + for source in (media_data.get('resourceList') or []): + format_url = source.get('url') + if not format_url: + continue + protocol = source.get('protocol') + quality = source.get('quality') + format_id = [] + for e in (protocol, source.get('encoding'), quality): + if e: + format_id.append(e) + format_id = '-'.join(format_id) + + if protocol in ('HDS', 'HLS'): + if source.get('tokenType') == 'AKAMAI': + format_url = self._get_tokenized_src( + format_url, media_id, format_id) + formats.extend(self._extract_akamai_formats( + format_url, media_id)) + elif protocol == 'HLS': + formats.extend(self._extract_m3u8_formats( + format_url, media_id, 'mp4', 'm3u8_native', + m3u8_id=format_id, fatal=False)) + elif protocol in ('HTTP', 'HTTPS'): + formats.append({ + 'format_id': format_id, + 'url': format_url, + 'quality': q(quality), + }) + + # This is needed because for audio medias the podcast url is usually + # always included, even if is only an audio segment and not the + # whole episode. + if int_or_none(media_data.get('position')) == 0: + for p in ('S', 'H'): + podcast_url = media_data.get('podcast%sdUrl' % p) + if not podcast_url: + continue + quality = p + 'D' + formats.append({ + 'format_id': 'PODCAST-' + quality, + 'url': podcast_url, + 'quality': q(quality), + }) self._sort_formats(formats) + subtitles = {} + if media_type == 'video': + for sub in (media_data.get('subtitleList') or []): + sub_url = sub.get('url') + if not sub_url: + continue + lang = sub.get('locale') or self._DEFAULT_LANGUAGE_CODES[bu] + subtitles.setdefault(lang, []).append({ + 'url': sub_url, + }) + return { 'id': media_id, 'title': title, - 'description': description, - 'timestamp': timestamp, - 'thumbnails': thumbnails, + 'description': media_data.get('description'), + 'timestamp': parse_iso8601(media_data.get('date')), + 'thumbnail': media_data.get('imageUrl'), + 'duration': float_or_none(media_data.get('duration'), 1000), + 'subtitles': subtitles, 'formats': formats, } @@ -119,26 +171,17 @@ class SRGSSRPlayIE(InfoExtractor): _TESTS = [{ 'url': 'http://www.srf.ch/play/tv/10vor10/video/snowden-beantragt-asyl-in-russland?id=28e1a57d-5b76-4399-8ab3-9097f071e6c5', - 'md5': 'da6b5b3ac9fa4761a942331cef20fcb3', + 'md5': '6db2226ba97f62ad42ce09783680046c', 'info_dict': { 'id': '28e1a57d-5b76-4399-8ab3-9097f071e6c5', 'ext': 'mp4', 'upload_date': '20130701', 'title': 'Snowden beantragt Asyl in Russland', - 'timestamp': 1372713995, - } - }, { - # No Speichern (Save) button - 'url': 'http://www.srf.ch/play/tv/top-gear/video/jaguar-xk120-shadow-und-tornado-dampflokomotive?id=677f5829-e473-4823-ac83-a1087fe97faa', - 'md5': '0a274ce38fda48c53c01890651985bc6', - 'info_dict': { - 'id': '677f5829-e473-4823-ac83-a1087fe97faa', - 'ext': 'flv', - 'upload_date': '20130710', - 'title': 'Jaguar XK120, Shadow und Tornado-Dampflokomotive', - 'description': 'md5:88604432b60d5a38787f152dec89cd56', - 'timestamp': 1373493600, + 'timestamp': 1372708215, + 'duration': 113.827, + 'thumbnail': r're:^https?://.*1383719781\.png$', }, + 'expected_warnings': ['Unable to download f4m manifest'], }, { 'url': 'http://www.rtr.ch/play/radio/actualitad/audio/saira-tujetsch-tuttina-cuntinuar-cun-sedrun-muster-turissem?id=63cb0778-27f8-49af-9284-8c7a8c6d15fc', 'info_dict': { @@ -146,7 +189,8 @@ class SRGSSRPlayIE(InfoExtractor): 'ext': 'mp3', 'upload_date': '20151013', 'title': 'Saira: Tujetsch - tuttina cuntinuar cun Sedrun Mustér Turissem', - 'timestamp': 1444750398, + 'timestamp': 1444709160, + 'duration': 336.816, }, 'params': { # rtmp download @@ -159,20 +203,33 @@ class SRGSSRPlayIE(InfoExtractor): 'id': '6348260', 'display_id': '6348260', 'ext': 'mp4', - 'duration': 1796, + 'duration': 1796.76, 'title': 'Le 19h30', - 'description': '', - 'uploader': '19h30', 'upload_date': '20141201', 'timestamp': 1417458600, 'thumbnail': r're:^https?://.*\.image', - 'view_count': int, }, 'params': { # m3u8 download 'skip_download': True, } }, { + 'url': 'http://play.swissinfo.ch/play/tv/business/video/why-people-were-against-tax-reforms?id=42960270', + 'info_dict': { + 'id': '42960270', + 'ext': 'mp4', + 'title': 'Why people were against tax reforms', + 'description': 'md5:7ac442c558e9630e947427469c4b824d', + 'duration': 94.0, + 'upload_date': '20170215', + 'timestamp': 1487173560, + 'thumbnail': r're:https?://www\.swissinfo\.ch/srgscalableimage/42961964', + 'subtitles': 'count:9', + }, + 'params': { + 'skip_download': True, + } + }, { 'url': 'https://www.srf.ch/play/tv/popupvideoplayer?id=c4dba0ca-e75b-43b2-a34f-f708a4932e01', 'only_matching': True, }, { @@ -181,6 +238,10 @@ class SRGSSRPlayIE(InfoExtractor): }, { 'url': 'https://www.rts.ch/play/tv/19h30/video/le-19h30?urn=urn:rts:video:6348260', 'only_matching': True, + }, { + # audio segment, has podcastSdUrl of the full episode + 'url': 'https://www.srf.ch/play/radio/popupaudioplayer?id=50b20dc8-f05b-4972-bf03-e438ff2833eb', + 'only_matching': True, }] def _real_extract(self, url): @@ -188,5 +249,4 @@ class SRGSSRPlayIE(InfoExtractor): bu = mobj.group('bu') media_type = mobj.group('type') or mobj.group('type_2') media_id = mobj.group('id') - # other info can be extracted from url + '&layout=json' return self.url_result('srgssr:%s:%s:%s' % (bu[:3], media_type, media_id), 'SRGSSR') diff --git a/youtube_dl/extractor/storyfire.py b/youtube_dl/extractor/storyfire.py new file mode 100644 index 000000000..9c698626f --- /dev/null +++ b/youtube_dl/extractor/storyfire.py @@ -0,0 +1,151 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import functools + +from .common import InfoExtractor +from ..utils import ( + # HEADRequest, + int_or_none, + OnDemandPagedList, + smuggle_url, +) + + +class StoryFireBaseIE(InfoExtractor): + _VALID_URL_BASE = r'https?://(?:www\.)?storyfire\.com/' + + def _call_api(self, path, video_id, resource, query=None): + return self._download_json( + 'https://storyfire.com/app/%s/%s' % (path, video_id), video_id, + 'Downloading %s JSON metadata' % resource, query=query) + + def _parse_video(self, video): + title = video['title'] + vimeo_id = self._search_regex( + r'https?://player\.vimeo\.com/external/(\d+)', + video['vimeoVideoURL'], 'vimeo id') + + # video_url = self._request_webpage( + # HEADRequest(video['vimeoVideoURL']), video_id).geturl() + # formats = [] + # for v_url, suffix in [(video_url, '_sep'), (video_url.replace('/sep/video/', '/video/'), '')]: + # formats.extend(self._extract_m3u8_formats( + # v_url, video_id, 'mp4', 'm3u8_native', + # m3u8_id='hls' + suffix, fatal=False)) + # formats.extend(self._extract_mpd_formats( + # v_url.replace('.m3u8', '.mpd'), video_id, + # mpd_id='dash' + suffix, fatal=False)) + # self._sort_formats(formats) + + uploader_id = video.get('hostID') + + return { + '_type': 'url_transparent', + 'id': vimeo_id, + 'title': title, + 'description': video.get('description'), + 'url': smuggle_url( + 'https://player.vimeo.com/video/' + vimeo_id, { + 'http_headers': { + 'Referer': 'https://storyfire.com/', + } + }), + # 'formats': formats, + 'thumbnail': video.get('storyImage'), + 'view_count': int_or_none(video.get('views')), + 'like_count': int_or_none(video.get('likesCount')), + 'comment_count': int_or_none(video.get('commentsCount')), + 'duration': int_or_none(video.get('videoDuration')), + 'timestamp': int_or_none(video.get('publishDate')), + 'uploader': video.get('username'), + 'uploader_id': uploader_id, + 'uploader_url': 'https://storyfire.com/user/%s/video' % uploader_id if uploader_id else None, + 'episode_number': int_or_none(video.get('episodeNumber') or video.get('episode_number')), + } + + +class StoryFireIE(StoryFireBaseIE): + _VALID_URL = StoryFireBaseIE._VALID_URL_BASE + r'video-details/(?P<id>[0-9a-f]{24})' + _TEST = { + 'url': 'https://storyfire.com/video-details/5df1d132b6378700117f9181', + 'md5': 'caec54b9e4621186d6079c7ec100c1eb', + 'info_dict': { + 'id': '378954662', + 'ext': 'mp4', + 'title': 'Buzzfeed Teaches You About Memes', + 'uploader_id': 'ntZAJFECERSgqHSxzonV5K2E89s1', + 'timestamp': 1576129028, + 'description': 'md5:0b4e28021548e144bed69bb7539e62ea', + 'uploader': 'whang!', + 'upload_date': '20191212', + 'duration': 418, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + }, + 'params': { + 'skip_download': True, + }, + 'expected_warnings': ['Unable to download JSON metadata'] + } + + def _real_extract(self, url): + video_id = self._match_id(url) + video = self._call_api( + 'generic/video-detail', video_id, 'video')['video'] + return self._parse_video(video) + + +class StoryFireUserIE(StoryFireBaseIE): + _VALID_URL = StoryFireBaseIE._VALID_URL_BASE + r'user/(?P<id>[^/]+)/video' + _TEST = { + 'url': 'https://storyfire.com/user/UQ986nFxmAWIgnkZQ0ftVhq4nOk2/video', + 'info_dict': { + 'id': 'UQ986nFxmAWIgnkZQ0ftVhq4nOk2', + }, + 'playlist_mincount': 151, + } + _PAGE_SIZE = 20 + + def _fetch_page(self, user_id, page): + videos = self._call_api( + 'publicVideos', user_id, 'page %d' % (page + 1), { + 'skip': page * self._PAGE_SIZE, + })['videos'] + for video in videos: + yield self._parse_video(video) + + def _real_extract(self, url): + user_id = self._match_id(url) + entries = OnDemandPagedList(functools.partial( + self._fetch_page, user_id), self._PAGE_SIZE) + return self.playlist_result(entries, user_id) + + +class StoryFireSeriesIE(StoryFireBaseIE): + _VALID_URL = StoryFireBaseIE._VALID_URL_BASE + r'write/series/stories/(?P<id>[^/?&#]+)' + _TESTS = [{ + 'url': 'https://storyfire.com/write/series/stories/-Lq6MsuIHLODO6d2dDkr/', + 'info_dict': { + 'id': '-Lq6MsuIHLODO6d2dDkr', + }, + 'playlist_mincount': 13, + }, { + 'url': 'https://storyfire.com/write/series/stories/the_mortal_one/', + 'info_dict': { + 'id': 'the_mortal_one', + }, + 'playlist_count': 0, + }] + + def _extract_videos(self, stories): + for story in stories.values(): + if story.get('hasVideo'): + yield self._parse_video(story) + + def _real_extract(self, url): + series_id = self._match_id(url) + stories = self._call_api( + 'seriesStories', series_id, 'series stories') + return self.playlist_result(self._extract_videos(stories), series_id) diff --git a/youtube_dl/extractor/stretchinternet.py b/youtube_dl/extractor/stretchinternet.py index 4dbead2ba..ec08eae55 100644 --- a/youtube_dl/extractor/stretchinternet.py +++ b/youtube_dl/extractor/stretchinternet.py @@ -1,7 +1,6 @@ from __future__ import unicode_literals from .common import InfoExtractor -from ..utils import int_or_none class StretchInternetIE(InfoExtractor): @@ -11,22 +10,28 @@ class StretchInternetIE(InfoExtractor): 'info_dict': { 'id': '573272', 'ext': 'mp4', - 'title': 'University of Mary Wrestling vs. Upper Iowa', - 'timestamp': 1575668361, - 'upload_date': '20191206', + 'title': 'UNIVERSITY OF MARY WRESTLING VS UPPER IOWA', + # 'timestamp': 1575668361, + # 'upload_date': '20191206', + 'uploader_id': '99997', } } def _real_extract(self, url): video_id = self._match_id(url) + media_url = self._download_json( + 'https://core.stretchlive.com/trinity/event/tcg/' + video_id, + video_id)[0]['media'][0]['url'] event = self._download_json( - 'https://api.stretchinternet.com/trinity/event/tcg/' + video_id, - video_id)[0] + 'https://neo-client.stretchinternet.com/portal-ws/getEvent.json', + video_id, query={'eventID': video_id, 'token': 'asdf'})['event'] return { 'id': video_id, 'title': event['title'], - 'timestamp': int_or_none(event.get('dateCreated'), 1000), - 'url': 'https://' + event['media'][0]['url'], + # TODO: parse US timezone abbreviations + # 'timestamp': event.get('dateTimeString'), + 'url': 'https://' + media_url, + 'uploader_id': event.get('ownerID'), } diff --git a/youtube_dl/extractor/svt.py b/youtube_dl/extractor/svt.py index a0b6ef4db..a5bb6daa7 100644 --- a/youtube_dl/extractor/svt.py +++ b/youtube_dl/extractor/svt.py @@ -146,18 +146,19 @@ class SVTPlayIE(SVTPlayBaseIE): ) (?P<svt_id>[^/?#&]+)| https?://(?:www\.)?(?:svtplay|oppetarkiv)\.se/(?:video|klipp|kanaler)/(?P<id>[^/?#&]+) + (?:.*?(?:modalId|id)=(?P<modal_id>[\da-zA-Z-]+))? ) ''' _TESTS = [{ - 'url': 'https://www.svtplay.se/video/26194546/det-har-ar-himlen', + 'url': 'https://www.svtplay.se/video/30479064', 'md5': '2382036fd6f8c994856c323fe51c426e', 'info_dict': { - 'id': 'jNwpV9P', + 'id': '8zVbDPA', 'ext': 'mp4', - 'title': 'Det här är himlen', - 'timestamp': 1586044800, - 'upload_date': '20200405', - 'duration': 3515, + 'title': 'Designdrömmar i Stenungsund', + 'timestamp': 1615770000, + 'upload_date': '20210315', + 'duration': 3519, 'thumbnail': r're:^https?://(?:.*[\.-]jpg|www.svtstatic.se/image/.*)$', 'age_limit': 0, 'subtitles': { @@ -174,6 +175,12 @@ class SVTPlayIE(SVTPlayBaseIE): 'skip_download': True, }, }, { + 'url': 'https://www.svtplay.se/video/30479064/husdrommar/husdrommar-sasong-8-designdrommar-i-stenungsund?modalId=8zVbDPA', + 'only_matching': True, + }, { + 'url': 'https://www.svtplay.se/video/30684086/rapport/rapport-24-apr-18-00-7?id=e72gVpa', + 'only_matching': True, + }, { # geo restricted to Sweden 'url': 'http://www.oppetarkiv.se/video/5219710/trollflojten', 'only_matching': True, @@ -219,7 +226,8 @@ class SVTPlayIE(SVTPlayBaseIE): def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) - video_id, svt_id = mobj.group('id', 'svt_id') + video_id = mobj.group('id') + svt_id = mobj.group('svt_id') or mobj.group('modal_id') if svt_id: return self._extract_by_video_id(svt_id) @@ -254,9 +262,12 @@ class SVTPlayIE(SVTPlayBaseIE): if not svt_id: svt_id = self._search_regex( (r'<video[^>]+data-video-id=["\']([\da-zA-Z-]+)', + r'<[^>]+\bdata-rt=["\']top-area-play-button["\'][^>]+\bhref=["\'][^"\']*video/%s/[^"\']*\b(?:modalId|id)=([\da-zA-Z-]+)' % re.escape(video_id), r'["\']videoSvtId["\']\s*:\s*["\']([\da-zA-Z-]+)', + r'["\']videoSvtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)', r'"content"\s*:\s*{.*?"id"\s*:\s*"([\da-zA-Z-]+)"', - r'["\']svtId["\']\s*:\s*["\']([\da-zA-Z-]+)'), + r'["\']svtId["\']\s*:\s*["\']([\da-zA-Z-]+)', + r'["\']svtId\\?["\']\s*:\s*\\?["\']([\da-zA-Z-]+)'), webpage, 'video id') info_dict = self._extract_by_video_id(svt_id, webpage) diff --git a/youtube_dl/extractor/ted.py b/youtube_dl/extractor/ted.py index 63e2455b2..f09f1a3f9 100644 --- a/youtube_dl/extractor/ted.py +++ b/youtube_dl/extractor/ted.py @@ -123,6 +123,10 @@ class TEDIE(InfoExtractor): 'params': { 'skip_download': True, }, + }, { + # with own formats and private Youtube external + 'url': 'https://www.ted.com/talks/spencer_wells_a_family_tree_for_humanity', + 'only_matching': True, }] _NATIVE_FORMATS = { @@ -210,16 +214,6 @@ class TEDIE(InfoExtractor): player_talk = talk_info['player_talks'][0] - external = player_talk.get('external') - if isinstance(external, dict): - service = external.get('service') - if isinstance(service, compat_str): - ext_url = None - if service.lower() == 'youtube': - ext_url = external.get('code') - - return self.url_result(ext_url or external['uri']) - resources_ = player_talk.get('resources') or talk_info.get('resources') http_url = None @@ -294,6 +288,16 @@ class TEDIE(InfoExtractor): 'vcodec': 'none', }) + if not formats: + external = player_talk.get('external') + if isinstance(external, dict): + service = external.get('service') + if isinstance(service, compat_str): + ext_url = None + if service.lower() == 'youtube': + ext_url = external.get('code') + return self.url_result(ext_url or external['uri']) + self._sort_formats(formats) video_id = compat_str(talk_info['id']) diff --git a/youtube_dl/extractor/tf1.py b/youtube_dl/extractor/tf1.py index 55e2a0721..23c2808a1 100644 --- a/youtube_dl/extractor/tf1.py +++ b/youtube_dl/extractor/tf1.py @@ -1,92 +1,87 @@ # coding: utf-8 from __future__ import unicode_literals +import json +import re + from .common import InfoExtractor -from ..compat import compat_str +from ..utils import ( + int_or_none, + parse_iso8601, + try_get, +) class TF1IE(InfoExtractor): - """TF1 uses the wat.tv player.""" - _VALID_URL = r'https?://(?:(?:videos|www|lci)\.tf1|(?:www\.)?(?:tfou|ushuaiatv|histoire|tvbreizh))\.fr/(?:[^/]+/)*(?P<id>[^/?#.]+)' + _VALID_URL = r'https?://(?:www\.)?tf1\.fr/[^/]+/(?P<program_slug>[^/]+)/videos/(?P<id>[^/?&#]+)\.html' _TESTS = [{ - 'url': 'http://videos.tf1.fr/auto-moto/citroen-grand-c4-picasso-2013-presentation-officielle-8062060.html', - 'info_dict': { - 'id': '10635995', - 'ext': 'mp4', - 'title': 'Citroën Grand C4 Picasso 2013 : présentation officielle', - 'description': 'Vidéo officielle du nouveau Citroën Grand C4 Picasso, lancé à l\'automne 2013.', - }, - 'params': { - # Sometimes wat serves the whole file with the --test option - 'skip_download': True, - }, - 'expected_warnings': ['HTTP Error 404'], - }, { - 'url': 'http://www.tfou.fr/chuggington/videos/le-grand-mysterioso-chuggington-7085291-739.html', + 'url': 'https://www.tf1.fr/tmc/quotidien-avec-yann-barthes/videos/quotidien-premiere-partie-11-juin-2019.html', 'info_dict': { - 'id': 'le-grand-mysterioso-chuggington-7085291-739', + 'id': '13641379', 'ext': 'mp4', - 'title': 'Le grand Mystérioso - Chuggington', - 'description': 'Le grand Mystérioso - Emery rêve qu\'un article lui soit consacré dans le journal.', - 'upload_date': '20150103', + 'title': 'md5:f392bc52245dc5ad43771650c96fb620', + 'description': 'md5:a02cdb217141fb2d469d6216339b052f', + 'upload_date': '20190611', + 'timestamp': 1560273989, + 'duration': 1738, + 'series': 'Quotidien avec Yann Barthès', + 'tags': ['intégrale', 'quotidien', 'Replay'], }, 'params': { # Sometimes wat serves the whole file with the --test option 'skip_download': True, + 'format': 'bestvideo', }, - 'skip': 'HTTP Error 410: Gone', }, { 'url': 'http://www.tf1.fr/tf1/koh-lanta/videos/replay-koh-lanta-22-mai-2015.html', 'only_matching': True, }, { - 'url': 'http://lci.tf1.fr/sept-a-huit/videos/sept-a-huit-du-24-mai-2015-8611550.html', - 'only_matching': True, - }, { 'url': 'http://www.tf1.fr/hd1/documentaire/videos/mylene-farmer-d-une-icone.html', 'only_matching': True, - }, { - 'url': 'https://www.tf1.fr/tmc/quotidien-avec-yann-barthes/videos/quotidien-premiere-partie-11-juin-2019.html', - 'info_dict': { - 'id': '13641379', - 'ext': 'mp4', - 'title': 'md5:f392bc52245dc5ad43771650c96fb620', - 'description': 'md5:44bc54f0a21322f5b91d68e76a544eae', - 'upload_date': '20190611', - }, - 'params': { - # Sometimes wat serves the whole file with the --test option - 'skip_download': True, - }, }] def _real_extract(self, url): - video_id = self._match_id(url) - - webpage = self._download_webpage(url, video_id) - - wat_id = None + program_slug, slug = re.match(self._VALID_URL, url).groups() + video = self._download_json( + 'https://www.tf1.fr/graphql/web', slug, query={ + 'id': '9b80783950b85247541dd1d851f9cc7fa36574af015621f853ab111a679ce26f', + 'variables': json.dumps({ + 'programSlug': program_slug, + 'slug': slug, + }) + })['data']['videoBySlug'] + wat_id = video['streamId'] - data = self._parse_json( - self._search_regex( - r'__APOLLO_STATE__\s*=\s*({.+?})\s*(?:;|</script>)', webpage, - 'data', default='{}'), video_id, fatal=False) + tags = [] + for tag in (video.get('tags') or []): + label = tag.get('label') + if not label: + continue + tags.append(label) - if data: - try: - wat_id = next( - video.get('streamId') - for key, video in data.items() - if isinstance(video, dict) - and video.get('slug') == video_id) - if not isinstance(wat_id, compat_str) or not wat_id.isdigit(): - wat_id = None - except StopIteration: - pass + decoration = video.get('decoration') or {} - if not wat_id: - wat_id = self._html_search_regex( - (r'(["\'])(?:https?:)?//www\.wat\.tv/embedframe/.*?(?P<id>\d{8})\1', - r'(["\']?)streamId\1\s*:\s*(["\']?)(?P<id>\d+)\2'), - webpage, 'wat id', group='id') + thumbnails = [] + for source in (try_get(decoration, lambda x: x['image']['sources'], list) or []): + source_url = source.get('url') + if not source_url: + continue + thumbnails.append({ + 'url': source_url, + 'width': int_or_none(source.get('width')), + }) - return self.url_result('wat:%s' % wat_id, 'Wat') + return { + '_type': 'url_transparent', + 'id': wat_id, + 'url': 'wat:' + wat_id, + 'title': video.get('title'), + 'thumbnails': thumbnails, + 'description': decoration.get('description'), + 'timestamp': parse_iso8601(video.get('date')), + 'duration': int_or_none(try_get(video, lambda x: x['publicPlayingInfos']['duration'])), + 'tags': tags, + 'series': decoration.get('programLabel'), + 'season_number': int_or_none(video.get('season')), + 'episode_number': int_or_none(video.get('episode')), + } diff --git a/youtube_dl/extractor/threeqsdn.py b/youtube_dl/extractor/threeqsdn.py index f26937da1..f6d37bb9e 100644 --- a/youtube_dl/extractor/threeqsdn.py +++ b/youtube_dl/extractor/threeqsdn.py @@ -3,10 +3,13 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from ..compat import compat_HTTPError from ..utils import ( determine_ext, - js_to_json, - mimetype2ext, + ExtractorError, + float_or_none, + int_or_none, + parse_iso8601, ) @@ -15,29 +18,35 @@ class ThreeQSDNIE(InfoExtractor): IE_DESC = '3Q SDN' _VALID_URL = r'https?://playout\.3qsdn\.com/(?P<id>[\da-f]{8}-[\da-f]{4}-[\da-f]{4}-[\da-f]{4}-[\da-f]{12})' _TESTS = [{ - # ondemand from http://www.philharmonie.tv/veranstaltung/26/ - 'url': 'http://playout.3qsdn.com/0280d6b9-1215-11e6-b427-0cc47a188158?protocol=http', - 'md5': 'ab040e37bcfa2e0c079f92cb1dd7f6cd', + # https://player.3qsdn.com/demo.html + 'url': 'https://playout.3qsdn.com/7201c779-6b3c-11e7-a40e-002590c750be', + 'md5': '64a57396b16fa011b15e0ea60edce918', 'info_dict': { - 'id': '0280d6b9-1215-11e6-b427-0cc47a188158', + 'id': '7201c779-6b3c-11e7-a40e-002590c750be', 'ext': 'mp4', - 'title': '0280d6b9-1215-11e6-b427-0cc47a188158', + 'title': 'Video Ads', 'is_live': False, + 'description': 'Video Ads Demo', + 'timestamp': 1500334803, + 'upload_date': '20170717', + 'duration': 888.032, + 'subtitles': { + 'eng': 'count:1', + }, }, - 'expected_warnings': ['Failed to download MPD manifest', 'Failed to parse JSON'], + 'expected_warnings': ['Unknown MIME type application/mp4 in DASH manifest'], }, { # live video stream - 'url': 'https://playout.3qsdn.com/d755d94b-4ab9-11e3-9162-0025907ad44f?js=true', + 'url': 'https://playout.3qsdn.com/66e68995-11ca-11e8-9273-002590c750be', 'info_dict': { - 'id': 'd755d94b-4ab9-11e3-9162-0025907ad44f', + 'id': '66e68995-11ca-11e8-9273-002590c750be', 'ext': 'mp4', - 'title': 're:^d755d94b-4ab9-11e3-9162-0025907ad44f [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', + 'title': 're:^66e68995-11ca-11e8-9273-002590c750be [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$', 'is_live': True, }, 'params': { 'skip_download': True, # m3u8 downloads }, - 'expected_warnings': ['Failed to download MPD manifest'], }, { # live audio stream 'url': 'http://playout.3qsdn.com/9edf36e0-6bf2-11e2-a16a-9acf09e2db48', @@ -58,6 +67,14 @@ class ThreeQSDNIE(InfoExtractor): # live video with rtmp link 'url': 'https://playout.3qsdn.com/6092bb9e-8f72-11e4-a173-002590c750be', 'only_matching': True, + }, { + # ondemand from http://www.philharmonie.tv/veranstaltung/26/ + 'url': 'http://playout.3qsdn.com/0280d6b9-1215-11e6-b427-0cc47a188158?protocol=http', + 'only_matching': True, + }, { + # live video stream + 'url': 'https://playout.3qsdn.com/d755d94b-4ab9-11e3-9162-0025907ad44f?js=true', + 'only_matching': True, }] @staticmethod @@ -70,73 +87,78 @@ class ThreeQSDNIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) - js = self._download_webpage( - 'http://playout.3qsdn.com/%s' % video_id, video_id, - query={'js': 'true'}) - - if any(p in js for p in ( - '>This content is not available in your country', - 'playout.3qsdn.com/forbidden')): - self.raise_geo_restricted() - - stream_content = self._search_regex( - r'streamContent\s*:\s*(["\'])(?P<content>.+?)\1', js, - 'stream content', default='demand', group='content') + try: + config = self._download_json( + url.replace('://playout.3qsdn.com/', '://playout.3qsdn.com/config/'), video_id) + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 401: + self.raise_geo_restricted() + raise - live = stream_content == 'live' - - stream_type = self._search_regex( - r'streamType\s*:\s*(["\'])(?P<type>audio|video)\1', js, - 'stream type', default='video', group='type') + live = config.get('streamContent') == 'live' + aspect = float_or_none(config.get('aspect')) formats = [] - urls = set() - - def extract_formats(item_url, item={}): - if not item_url or item_url in urls: - return - urls.add(item_url) - ext = mimetype2ext(item.get('type')) or determine_ext(item_url, default_ext=None) - if ext == 'mpd': + for source_type, source in (config.get('sources') or {}).items(): + if not source: + continue + if source_type == 'dash': formats.extend(self._extract_mpd_formats( - item_url, video_id, mpd_id='mpd', fatal=False)) - elif ext == 'm3u8': + source, video_id, mpd_id='mpd', fatal=False)) + elif source_type == 'hls': formats.extend(self._extract_m3u8_formats( - item_url, video_id, 'mp4', - entry_protocol='m3u8' if live else 'm3u8_native', + source, video_id, 'mp4', 'm3u8' if live else 'm3u8_native', m3u8_id='hls', fatal=False)) - elif ext == 'f4m': - formats.extend(self._extract_f4m_formats( - item_url, video_id, f4m_id='hds', fatal=False)) - else: - if not self._is_valid_url(item_url, video_id): - return - formats.append({ - 'url': item_url, - 'format_id': item.get('quality'), - 'ext': 'mp4' if item_url.startswith('rtsp') else ext, - 'vcodec': 'none' if stream_type == 'audio' else None, - }) - - for item_js in re.findall(r'({[^{]*?\b(?:src|source)\s*:\s*["\'].+?})', js): - f = self._parse_json( - item_js, video_id, transform_source=js_to_json, fatal=False) - if not f: + elif source_type == 'progressive': + for s in source: + src = s.get('src') + if not (src and self._is_valid_url(src, video_id)): + continue + width = None + format_id = ['http'] + ext = determine_ext(src) + if ext: + format_id.append(ext) + height = int_or_none(s.get('height')) + if height: + format_id.append('%dp' % height) + if aspect: + width = int(height * aspect) + formats.append({ + 'ext': ext, + 'format_id': '-'.join(format_id), + 'height': height, + 'source_preference': 0, + 'url': src, + 'vcodec': 'none' if height == 0 else None, + 'width': width, + }) + for f in formats: + if f.get('acodec') == 'none': + f['preference'] = -40 + elif f.get('vcodec') == 'none': + f['preference'] = -50 + self._sort_formats(formats, ('preference', 'width', 'height', 'source_preference', 'tbr', 'vbr', 'abr', 'ext', 'format_id')) + + subtitles = {} + for subtitle in (config.get('subtitles') or []): + src = subtitle.get('src') + if not src: continue - extract_formats(f.get('src'), f) - - # More relaxed version to collect additional URLs and acting - # as a future-proof fallback - for _, src in re.findall(r'\b(?:src|source)\s*:\s*(["\'])((?:https?|rtsp)://.+?)\1', js): - extract_formats(src) - - self._sort_formats(formats) + subtitles.setdefault(subtitle.get('label') or 'eng', []).append({ + 'url': src, + }) - title = self._live_title(video_id) if live else video_id + title = config.get('title') or video_id return { 'id': video_id, - 'title': title, + 'title': self._live_title(title) if live else title, + 'thumbnail': config.get('poster') or None, + 'description': config.get('description') or None, + 'timestamp': parse_iso8601(config.get('upload_date')), + 'duration': float_or_none(config.get('vlength')) or None, 'is_live': live, 'formats': formats, + 'subtitles': subtitles, } diff --git a/youtube_dl/extractor/tiktok.py b/youtube_dl/extractor/tiktok.py index ea1beb8af..4faa6de54 100644 --- a/youtube_dl/extractor/tiktok.py +++ b/youtube_dl/extractor/tiktok.py @@ -107,9 +107,12 @@ class TikTokIE(TikTokBaseIE): def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) - data = self._parse_json(self._search_regex( + page_props = self._parse_json(self._search_regex( r'<script[^>]+\bid=["\']__NEXT_DATA__[^>]+>\s*({.+?})\s*</script', - webpage, 'data'), video_id)['props']['pageProps']['itemInfo']['itemStruct'] + webpage, 'data'), video_id)['props']['pageProps'] + data = try_get(page_props, lambda x: x['itemInfo']['itemStruct'], dict) + if not data and page_props.get('statusCode') == 10216: + raise ExtractorError('This video is private', expected=True) return self._extract_video(data, video_id) diff --git a/youtube_dl/extractor/tmz.py b/youtube_dl/extractor/tmz.py index 419f9d92e..3d1bf75ff 100644 --- a/youtube_dl/extractor/tmz.py +++ b/youtube_dl/extractor/tmz.py @@ -2,55 +2,110 @@ from __future__ import unicode_literals from .common import InfoExtractor +from .jwplatform import JWPlatformIE +from .kaltura import KalturaIE +from ..utils import ( + int_or_none, + unified_timestamp, +) class TMZIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tmz\.com/videos/(?P<id>[^/?#]+)' + _VALID_URL = r'https?://(?:www\.)?tmz\.com/videos/(?P<id>[^/?#&]+)' _TESTS = [{ - 'url': 'http://www.tmz.com/videos/0_okj015ty/', - 'md5': '4d22a51ef205b6c06395d8394f72d560', + 'url': 'http://www.tmz.com/videos/0-cegprt2p/', + 'md5': '31f9223e20eef55954973359afa61a20', 'info_dict': { - 'id': '0_okj015ty', + 'id': 'P6YjLBLk', 'ext': 'mp4', - 'title': 'Kim Kardashian\'s Boobs Unlock a Mystery!', - 'description': 'Did Kim Kardasain try to one-up Khloe by one-upping Kylie??? Or is she just showing off her amazing boobs?', - 'timestamp': 1394747163, - 'uploader_id': 'batchUser', - 'upload_date': '20140313', - } + 'title': "No Charges Against Hillary Clinton? Harvey Says It Ain't Over Yet", + 'description': 'md5:b714359fc18607715ebccbd2da8ff488', + 'timestamp': 1467831837, + 'upload_date': '20160706', + }, + 'add_ie': [JWPlatformIE.ie_key()], }, { - 'url': 'http://www.tmz.com/videos/0-cegprt2p/', + 'url': 'http://www.tmz.com/videos/0_okj015ty/', + 'only_matching': True, + }, { + 'url': 'https://www.tmz.com/videos/071119-chris-morgan-women-4590005-0-zcsejvcr/', + 'only_matching': True, + }, { + 'url': 'https://www.tmz.com/videos/2021-02-19-021921-floyd-mayweather-1043872/', 'only_matching': True, }] def _real_extract(self, url): video_id = self._match_id(url).replace('-', '_') - return self.url_result('kaltura:591531:%s' % video_id, 'Kaltura', video_id) + + webpage = self._download_webpage(url, video_id, fatal=False) + if webpage: + tmz_video_id = self._search_regex( + r'nodeRef\s*:\s*["\']tmz:video:([\da-fA-F]{8}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{4}-[\da-fA-F]{12})', + webpage, 'video id', default=None) + video = self._download_json( + 'https://www.tmz.com/_/video/%s' % tmz_video_id, video_id, + fatal=False) + if video: + message = video['message'] + info = { + '_type': 'url_transparent', + 'title': message.get('title'), + 'description': message.get('description'), + 'timestamp': unified_timestamp(message.get('published_at')), + 'duration': int_or_none(message.get('duration')), + } + jwplatform_id = message.get('jwplayer_media_id') + if jwplatform_id: + info.update({ + 'url': 'jwplatform:%s' % jwplatform_id, + 'ie_key': JWPlatformIE.ie_key(), + }) + else: + kaltura_entry_id = message.get('kaltura_entry_id') or video_id + kaltura_partner_id = message.get('kaltura_partner_id') or '591531' + info.update({ + 'url': 'kaltura:%s:%s' % (kaltura_partner_id, kaltura_entry_id), + 'ie_key': KalturaIE.ie_key(), + }) + return info + + return self.url_result( + 'kaltura:591531:%s' % video_id, KalturaIE.ie_key(), video_id) class TMZArticleIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?tmz\.com/\d{4}/\d{2}/\d{2}/(?P<id>[^/]+)/?' + _VALID_URL = r'https?://(?:www\.)?tmz\.com/\d{4}/\d{2}/\d{2}/(?P<id>[^/?#&]+)' _TEST = { 'url': 'http://www.tmz.com/2015/04/19/bobby-brown-bobbi-kristina-awake-video-concert', - 'md5': '3316ff838ae5bb7f642537825e1e90d2', 'info_dict': { - 'id': '0_6snoelag', - 'ext': 'mov', + 'id': 'PAKZa97W', + 'ext': 'mp4', 'title': 'Bobby Brown Tells Crowd ... Bobbi Kristina is Awake', 'description': 'Bobby Brown stunned his audience during a concert Saturday night, when he told the crowd, "Bobbi is awake. She\'s watching me."', - 'timestamp': 1429467813, + 'timestamp': 1429466400, 'upload_date': '20150419', - 'uploader_id': 'batchUser', - } + }, + 'params': { + 'skip_download': True, + }, + 'add_ie': [JWPlatformIE.ie_key()], } def _real_extract(self, url): video_id = self._match_id(url) webpage = self._download_webpage(url, video_id) + + tmz_url = self._search_regex( + r'clickLink\s*\(\s*["\'](?P<url>%s)' % TMZIE._VALID_URL, webpage, + 'video id', default=None, group='url') + if tmz_url: + return self.url_result(tmz_url, ie=TMZIE.ie_key()) + embedded_video_info = self._parse_json(self._html_search_regex( r'tmzVideoEmbed\(({.+?})\);', webpage, 'embedded video info'), video_id) - return self.url_result( - 'http://www.tmz.com/videos/%s/' % embedded_video_info['id']) + 'http://www.tmz.com/videos/%s/' % embedded_video_info['id'], + ie=TMZIE.ie_key()) diff --git a/youtube_dl/extractor/trovo.py b/youtube_dl/extractor/trovo.py new file mode 100644 index 000000000..de0107aa9 --- /dev/null +++ b/youtube_dl/extractor/trovo.py @@ -0,0 +1,194 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import json + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + int_or_none, + str_or_none, + try_get, +) + + +class TrovoBaseIE(InfoExtractor): + _VALID_URL_BASE = r'https?://(?:www\.)?trovo\.live/' + + def _extract_streamer_info(self, data): + streamer_info = data.get('streamerInfo') or {} + username = streamer_info.get('userName') + return { + 'uploader': streamer_info.get('nickName'), + 'uploader_id': str_or_none(streamer_info.get('uid')), + 'uploader_url': 'https://trovo.live/' + username if username else None, + } + + +class TrovoIE(TrovoBaseIE): + _VALID_URL = TrovoBaseIE._VALID_URL_BASE + r'(?!(?:clip|video)/)(?P<id>[^/?&#]+)' + + def _real_extract(self, url): + username = self._match_id(url) + live_info = self._download_json( + 'https://gql.trovo.live/', username, query={ + 'query': '''{ + getLiveInfo(params: {userName: "%s"}) { + isLive + programInfo { + coverUrl + id + streamInfo { + desc + playUrl + } + title + } + streamerInfo { + nickName + uid + userName + } + } +}''' % username, + })['data']['getLiveInfo'] + if live_info.get('isLive') == 0: + raise ExtractorError('%s is offline' % username, expected=True) + program_info = live_info['programInfo'] + program_id = program_info['id'] + title = self._live_title(program_info['title']) + + formats = [] + for stream_info in (program_info.get('streamInfo') or []): + play_url = stream_info.get('playUrl') + if not play_url: + continue + format_id = stream_info.get('desc') + formats.append({ + 'format_id': format_id, + 'height': int_or_none(format_id[:-1]) if format_id else None, + 'url': play_url, + }) + self._sort_formats(formats) + + info = { + 'id': program_id, + 'title': title, + 'formats': formats, + 'thumbnail': program_info.get('coverUrl'), + 'is_live': True, + } + info.update(self._extract_streamer_info(live_info)) + return info + + +class TrovoVodIE(TrovoBaseIE): + _VALID_URL = TrovoBaseIE._VALID_URL_BASE + r'(?:clip|video)/(?P<id>[^/?&#]+)' + _TESTS = [{ + 'url': 'https://trovo.live/video/ltv-100095501_100095501_1609596043', + 'info_dict': { + 'id': 'ltv-100095501_100095501_1609596043', + 'ext': 'mp4', + 'title': 'Spontaner 12 Stunden Stream! - Ok Boomer!', + 'uploader': 'Exsl', + 'timestamp': 1609640305, + 'upload_date': '20210103', + 'uploader_id': '100095501', + 'duration': 43977, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + 'comments': 'mincount:8', + 'categories': ['Grand Theft Auto V'], + }, + }, { + 'url': 'https://trovo.live/clip/lc-5285890810184026005', + 'only_matching': True, + }] + + def _real_extract(self, url): + vid = self._match_id(url) + resp = self._download_json( + 'https://gql.trovo.live/', vid, data=json.dumps([{ + 'query': '''{ + batchGetVodDetailInfo(params: {vids: ["%s"]}) { + VodDetailInfos + } +}''' % vid, + }, { + 'query': '''{ + getCommentList(params: {appInfo: {postID: "%s"}, pageSize: 1000000000, preview: {}}) { + commentList { + author { + nickName + uid + } + commentID + content + createdAt + parentID + } + } +}''' % vid, + }]).encode(), headers={ + 'Content-Type': 'application/json', + }) + vod_detail_info = resp[0]['data']['batchGetVodDetailInfo']['VodDetailInfos'][vid] + vod_info = vod_detail_info['vodInfo'] + title = vod_info['title'] + + language = vod_info.get('languageName') + formats = [] + for play_info in (vod_info.get('playInfos') or []): + play_url = play_info.get('playUrl') + if not play_url: + continue + format_id = play_info.get('desc') + formats.append({ + 'ext': 'mp4', + 'filesize': int_or_none(play_info.get('fileSize')), + 'format_id': format_id, + 'height': int_or_none(format_id[:-1]) if format_id else None, + 'language': language, + 'protocol': 'm3u8_native', + 'tbr': int_or_none(play_info.get('bitrate')), + 'url': play_url, + 'http_headers': {'Origin': 'https://trovo.live'}, + }) + self._sort_formats(formats) + + category = vod_info.get('categoryName') + get_count = lambda x: int_or_none(vod_info.get(x + 'Num')) + + comment_list = try_get(resp, lambda x: x[1]['data']['getCommentList']['commentList'], list) or [] + comments = [] + for comment in comment_list: + content = comment.get('content') + if not content: + continue + author = comment.get('author') or {} + parent = comment.get('parentID') + comments.append({ + 'author': author.get('nickName'), + 'author_id': str_or_none(author.get('uid')), + 'id': str_or_none(comment.get('commentID')), + 'text': content, + 'timestamp': int_or_none(comment.get('createdAt')), + 'parent': 'root' if parent == 0 else str_or_none(parent), + }) + + info = { + 'id': vid, + 'title': title, + 'formats': formats, + 'thumbnail': vod_info.get('coverUrl'), + 'timestamp': int_or_none(vod_info.get('publishTs')), + 'duration': int_or_none(vod_info.get('duration')), + 'view_count': get_count('watch'), + 'like_count': get_count('like'), + 'comment_count': get_count('comment'), + 'comments': comments, + 'categories': [category] if category else None, + } + info.update(self._extract_streamer_info(vod_detail_info)) + return info diff --git a/youtube_dl/extractor/tv2.py b/youtube_dl/extractor/tv2.py index 4a19b9be6..334b7d540 100644 --- a/youtube_dl/extractor/tv2.py +++ b/youtube_dl/extractor/tv2.py @@ -20,7 +20,7 @@ from ..utils import ( class TV2IE(InfoExtractor): _VALID_URL = r'https?://(?:www\.)?tv2\.no/v/(?P<id>\d+)' - _TEST = { + _TESTS = [{ 'url': 'http://www.tv2.no/v/916509/', 'info_dict': { 'id': '916509', @@ -33,7 +33,7 @@ class TV2IE(InfoExtractor): 'view_count': int, 'categories': list, }, - } + }] _API_DOMAIN = 'sumo.tv2.no' _PROTOCOLS = ('HDS', 'HLS', 'DASH') _GEO_COUNTRIES = ['NO'] @@ -42,6 +42,12 @@ class TV2IE(InfoExtractor): video_id = self._match_id(url) api_base = 'http://%s/api/web/asset/%s' % (self._API_DOMAIN, video_id) + asset = self._download_json( + api_base + '.json', video_id, + 'Downloading metadata JSON')['asset'] + title = asset.get('subtitle') or asset['title'] + is_live = asset.get('live') is True + formats = [] format_urls = [] for protocol in self._PROTOCOLS: @@ -81,7 +87,8 @@ class TV2IE(InfoExtractor): elif ext == 'm3u8': if not data.get('drmProtected'): formats.extend(self._extract_m3u8_formats( - video_url, video_id, 'mp4', entry_protocol='m3u8_native', + video_url, video_id, 'mp4', + 'm3u8' if is_live else 'm3u8_native', m3u8_id=format_id, fatal=False)) elif ext == 'mpd': formats.extend(self._extract_mpd_formats( @@ -99,11 +106,6 @@ class TV2IE(InfoExtractor): raise ExtractorError('This video is DRM protected.', expected=True) self._sort_formats(formats) - asset = self._download_json( - api_base + '.json', video_id, - 'Downloading metadata JSON')['asset'] - title = asset['title'] - thumbnails = [{ 'id': thumbnail.get('@type'), 'url': thumbnail.get('url'), @@ -112,7 +114,7 @@ class TV2IE(InfoExtractor): return { 'id': video_id, 'url': video_url, - 'title': title, + 'title': self._live_title(title) if is_live else title, 'description': strip_or_none(asset.get('description')), 'thumbnails': thumbnails, 'timestamp': parse_iso8601(asset.get('createTime')), @@ -120,6 +122,7 @@ class TV2IE(InfoExtractor): 'view_count': int_or_none(asset.get('views')), 'categories': asset.get('keywords', '').split(','), 'formats': formats, + 'is_live': is_live, } @@ -168,13 +171,13 @@ class TV2ArticleIE(InfoExtractor): class KatsomoIE(TV2IE): - _VALID_URL = r'https?://(?:www\.)?(?:katsomo|mtv)\.fi/(?:#!/)?(?:[^/]+/[0-9a-z-]+-\d+/[0-9a-z-]+-|[^/]+/\d+/[^/]+/)(?P<id>\d+)' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?(?:katsomo|mtv(uutiset)?)\.fi/(?:sarja/[0-9a-z-]+-\d+/[0-9a-z-]+-|(?:#!/)?jakso/(?:\d+/[^/]+/)?|video/prog)(?P<id>\d+)' + _TESTS = [{ 'url': 'https://www.mtv.fi/sarja/mtv-uutiset-live-33001002003/lahden-pelicans-teki-kovan-ratkaisun-ville-nieminen-pihalle-1181321', 'info_dict': { 'id': '1181321', 'ext': 'mp4', - 'title': 'MTV Uutiset Live', + 'title': 'Lahden Pelicans teki kovan ratkaisun – Ville Nieminen pihalle', 'description': 'Päätöksen teki Pelicansin hallitus.', 'timestamp': 1575116484, 'upload_date': '20191130', @@ -186,7 +189,60 @@ class KatsomoIE(TV2IE): # m3u8 download 'skip_download': True, }, - } + }, { + 'url': 'http://www.katsomo.fi/#!/jakso/33001005/studio55-fi/658521/jukka-kuoppamaki-tekee-yha-lauluja-vaikka-lentokoneessa', + 'only_matching': True, + }, { + 'url': 'https://www.mtvuutiset.fi/video/prog1311159', + 'only_matching': True, + }, { + 'url': 'https://www.katsomo.fi/#!/jakso/1311159', + 'only_matching': True, + }] _API_DOMAIN = 'api.katsomo.fi' _PROTOCOLS = ('HLS', 'MPD') _GEO_COUNTRIES = ['FI'] + + +class MTVUutisetArticleIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)mtvuutiset\.fi/artikkeli/[^/]+/(?P<id>\d+)' + _TESTS = [{ + 'url': 'https://www.mtvuutiset.fi/artikkeli/tallaisia-vaurioita-viking-amorellassa-on-useamman-osaston-alla-vetta/7931384', + 'info_dict': { + 'id': '1311159', + 'ext': 'mp4', + 'title': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla', + 'description': 'Viking Amorellan matkustajien evakuointi on alkanut – tältä operaatio näyttää laivalla', + 'timestamp': 1600608966, + 'upload_date': '20200920', + 'duration': 153.7886666, + 'view_count': int, + 'categories': list, + }, + 'params': { + # m3u8 download + 'skip_download': True, + }, + }, { + # multiple Youtube embeds + 'url': 'https://www.mtvuutiset.fi/artikkeli/50-vuotta-subarun-vastaiskua/6070962', + 'only_matching': True, + }] + + def _real_extract(self, url): + article_id = self._match_id(url) + article = self._download_json( + 'http://api.mtvuutiset.fi/mtvuutiset/api/json/' + article_id, + article_id) + + def entries(): + for video in (article.get('videos') or []): + video_type = video.get('videotype') + video_url = video.get('url') + if not (video_url and video_type in ('katsomo', 'youtube')): + continue + yield self.url_result( + video_url, video_type.capitalize(), video.get('video_id')) + + return self.playlist_result( + entries(), article_id, article.get('title'), article.get('description')) diff --git a/youtube_dl/extractor/tv2dk.py b/youtube_dl/extractor/tv2dk.py index 8bda9348d..8bd5fd640 100644 --- a/youtube_dl/extractor/tv2dk.py +++ b/youtube_dl/extractor/tv2dk.py @@ -74,6 +74,12 @@ class TV2DKIE(InfoExtractor): webpage = self._download_webpage(url, video_id) entries = [] + + def add_entry(partner_id, kaltura_id): + entries.append(self.url_result( + 'kaltura:%s:%s' % (partner_id, kaltura_id), 'Kaltura', + video_id=kaltura_id)) + for video_el in re.findall(r'(?s)<[^>]+\bdata-entryid\s*=[^>]*>', webpage): video = extract_attributes(video_el) kaltura_id = video.get('data-entryid') @@ -82,9 +88,14 @@ class TV2DKIE(InfoExtractor): partner_id = video.get('data-partnerid') if not partner_id: continue - entries.append(self.url_result( - 'kaltura:%s:%s' % (partner_id, kaltura_id), 'Kaltura', - video_id=kaltura_id)) + add_entry(partner_id, kaltura_id) + if not entries: + kaltura_id = self._search_regex( + r'entry_id\s*:\s*["\']([0-9a-z_]+)', webpage, 'kaltura id') + partner_id = self._search_regex( + (r'\\u002Fp\\u002F(\d+)\\u002F', r'/p/(\d+)/'), webpage, + 'partner id') + add_entry(partner_id, kaltura_id) return self.playlist_result(entries) diff --git a/youtube_dl/extractor/tv4.py b/youtube_dl/extractor/tv4.py index c498b0191..b73bab9a8 100644 --- a/youtube_dl/extractor/tv4.py +++ b/youtube_dl/extractor/tv4.py @@ -17,7 +17,7 @@ class TV4IE(InfoExtractor): tv4\.se/(?:[^/]+)/klipp/(?:.*)-| tv4play\.se/ (?: - (?:program|barn)/(?:[^/]+/|(?:[^\?]+)\?video_id=)| + (?:program|barn)/(?:(?:[^/]+/){1,2}|(?:[^\?]+)\?video_id=)| iframe/video/| film/| sport/| @@ -65,6 +65,10 @@ class TV4IE(InfoExtractor): { 'url': 'http://www.tv4play.se/program/farang/3922081', 'only_matching': True, + }, + { + 'url': 'https://www.tv4play.se/program/nyheterna/avsnitt/13315940', + 'only_matching': True, } ] diff --git a/youtube_dl/extractor/tver.py b/youtube_dl/extractor/tver.py index 931d4d650..a4a30b1e6 100644 --- a/youtube_dl/extractor/tver.py +++ b/youtube_dl/extractor/tver.py @@ -25,6 +25,10 @@ class TVerIE(InfoExtractor): }, { 'url': 'https://tver.jp/episode/79622438', 'only_matching': True, + }, { + # subtitle = ' ' + 'url': 'https://tver.jp/corner/f0068870', + 'only_matching': True, }] _TOKEN = None BRIGHTCOVE_URL_TEMPLATE = 'http://players.brightcove.net/%s/default_default/index.html?videoId=%s' @@ -40,28 +44,18 @@ class TVerIE(InfoExtractor): query={'token': self._TOKEN})['main'] p_id = main['publisher_id'] service = remove_start(main['service'], 'ts_') - info = { + + r_id = main['reference_id'] + if service not in ('tx', 'russia2018', 'sebare2018live', 'gorin'): + r_id = 'ref:' + r_id + bc_url = smuggle_url( + self.BRIGHTCOVE_URL_TEMPLATE % (p_id, r_id), + {'geo_countries': ['JP']}) + + return { '_type': 'url_transparent', 'description': try_get(main, lambda x: x['note'][0]['text'], compat_str), 'episode_number': int_or_none(try_get(main, lambda x: x['ext']['episode_number'])), + 'url': bc_url, + 'ie_key': 'BrightcoveNew', } - - if service == 'cx': - info.update({ - 'title': main.get('subtitle') or main['title'], - 'url': 'https://i.fod.fujitv.co.jp/plus7/web/%s/%s.html' % (p_id[:4], p_id), - 'ie_key': 'FujiTVFODPlus7', - }) - else: - r_id = main['reference_id'] - if service not in ('tx', 'russia2018', 'sebare2018live', 'gorin'): - r_id = 'ref:' + r_id - bc_url = smuggle_url( - self.BRIGHTCOVE_URL_TEMPLATE % (p_id, r_id), - {'geo_countries': ['JP']}) - info.update({ - 'url': bc_url, - 'ie_key': 'BrightcoveNew', - }) - - return info diff --git a/youtube_dl/extractor/twitch.py b/youtube_dl/extractor/twitch.py index db264e8a1..a378bd6dc 100644 --- a/youtube_dl/extractor/twitch.py +++ b/youtube_dl/extractor/twitch.py @@ -17,6 +17,7 @@ from ..compat import ( ) from ..utils import ( clean_html, + dict_get, ExtractorError, float_or_none, int_or_none, @@ -48,6 +49,7 @@ class TwitchBaseIE(InfoExtractor): 'ChannelCollectionsContent': '07e3691a1bad77a36aba590c351180439a40baefc1c275356f40fc7082419a84', 'StreamMetadata': '1c719a40e481453e5c48d9bb585d971b8b372f8ebb105b17076722264dfa5b3e', 'ComscoreStreamingQuery': 'e1edae8122517d013405f237ffcc124515dc6ded82480a88daef69c83b53ac01', + 'VideoAccessToken_Clip': '36b89d2507fce29e5ca551df756d27c1cfe079e2609642b4390aa4c35796eb11', 'VideoPreviewOverlay': '3006e77e51b128d838fa4e835723ca4dc9a05c5efd4466c1085215c6e437e65c', 'VideoMetadata': '226edb3e692509f727fd56821f5653c05740242c82b0388883e0c0e75dcbf687', } @@ -76,14 +78,14 @@ class TwitchBaseIE(InfoExtractor): headers = { 'Referer': page_url, - 'Origin': page_url, + 'Origin': 'https://www.twitch.tv', 'Content-Type': 'text/plain;charset=UTF-8', } response = self._download_json( post_url, None, note, data=json.dumps(form).encode(), headers=headers, expected_status=400) - error = response.get('error_description') or response.get('error_code') + error = dict_get(response, ('error', 'error_description', 'error_code')) if error: fail(error) @@ -137,13 +139,17 @@ class TwitchBaseIE(InfoExtractor): self._sort_formats(formats) def _download_base_gql(self, video_id, ops, note, fatal=True): + headers = { + 'Content-Type': 'text/plain;charset=UTF-8', + 'Client-ID': self._CLIENT_ID, + } + gql_auth = self._get_cookies('https://gql.twitch.tv').get('auth-token') + if gql_auth: + headers['Authorization'] = 'OAuth ' + gql_auth.value return self._download_json( 'https://gql.twitch.tv/gql', video_id, note, data=json.dumps(ops).encode(), - headers={ - 'Content-Type': 'text/plain;charset=UTF-8', - 'Client-ID': self._CLIENT_ID, - }, fatal=fatal) + headers=headers, fatal=fatal) def _download_gql(self, video_id, ops, note, fatal=True): for op in ops: @@ -888,7 +894,25 @@ class TwitchClipsIE(TwitchBaseIE): def _real_extract(self, url): video_id = self._match_id(url) - clip = self._download_base_gql( + clip = self._download_gql( + video_id, [{ + 'operationName': 'VideoAccessToken_Clip', + 'variables': { + 'slug': video_id, + }, + }], + 'Downloading clip access token GraphQL')[0]['data']['clip'] + + if not clip: + raise ExtractorError( + 'This clip is no longer available', expected=True) + + access_query = { + 'sig': clip['playbackAccessToken']['signature'], + 'token': clip['playbackAccessToken']['value'], + } + + data = self._download_base_gql( video_id, { 'query': '''{ clip(slug: "%s") { @@ -913,11 +937,10 @@ class TwitchClipsIE(TwitchBaseIE): } viewCount } -}''' % video_id}, 'Downloading clip GraphQL')['data']['clip'] +}''' % video_id}, 'Downloading clip GraphQL', fatal=False) - if not clip: - raise ExtractorError( - 'This clip is no longer available', expected=True) + if data: + clip = try_get(data, lambda x: x['data']['clip'], dict) or clip formats = [] for option in clip.get('videoQualities', []): @@ -927,7 +950,7 @@ class TwitchClipsIE(TwitchBaseIE): if not source: continue formats.append({ - 'url': source, + 'url': update_url_query(source, access_query), 'format_id': option.get('quality'), 'height': int_or_none(option.get('quality')), 'fps': int_or_none(option.get('frameRate')), diff --git a/youtube_dl/extractor/twitter.py b/youtube_dl/extractor/twitter.py index 1190d721e..cfa7a7326 100644 --- a/youtube_dl/extractor/twitter.py +++ b/youtube_dl/extractor/twitter.py @@ -19,6 +19,7 @@ from ..utils import ( strip_or_none, unified_timestamp, update_url_query, + url_or_none, xpath_text, ) @@ -52,6 +53,9 @@ class TwitterBaseIE(InfoExtractor): return [f] def _extract_formats_from_vmap_url(self, vmap_url, video_id): + vmap_url = url_or_none(vmap_url) + if not vmap_url: + return [] vmap_data = self._download_xml(vmap_url, video_id) formats = [] urls = [] @@ -374,6 +378,24 @@ class TwitterIE(TwitterBaseIE): }, 'add_ie': ['TwitterBroadcast'], }, { + # unified card + 'url': 'https://twitter.com/BrooklynNets/status/1349794411333394432?s=20', + 'info_dict': { + 'id': '1349794411333394432', + 'ext': 'mp4', + 'title': 'md5:d1c4941658e4caaa6cb579260d85dcba', + 'thumbnail': r're:^https?://.*\.jpg', + 'description': 'md5:71ead15ec44cee55071547d6447c6a3e', + 'uploader': 'Brooklyn Nets', + 'uploader_id': 'BrooklynNets', + 'duration': 324.484, + 'timestamp': 1610651040, + 'upload_date': '20210114', + }, + 'params': { + 'skip_download': True, + }, + }, { # Twitch Clip Embed 'url': 'https://twitter.com/GunB1g/status/1163218564784017422', 'only_matching': True, @@ -389,6 +411,22 @@ class TwitterIE(TwitterBaseIE): # appplayer card 'url': 'https://twitter.com/poco_dandy/status/1150646424461176832', 'only_matching': True, + }, { + # video_direct_message card + 'url': 'https://twitter.com/qarev001/status/1348948114569269251', + 'only_matching': True, + }, { + # poll2choice_video card + 'url': 'https://twitter.com/CAF_Online/status/1349365911120195585', + 'only_matching': True, + }, { + # poll3choice_video card + 'url': 'https://twitter.com/SamsungMobileSA/status/1348609186725289984', + 'only_matching': True, + }, { + # poll4choice_video card + 'url': 'https://twitter.com/SouthamptonFC/status/1347577658079641604', + 'only_matching': True, }] def _real_extract(self, url): @@ -433,8 +471,7 @@ class TwitterIE(TwitterBaseIE): 'tags': tags, } - media = try_get(status, lambda x: x['extended_entities']['media'][0]) - if media and media.get('type') != 'photo': + def extract_from_video_info(media): video_info = media.get('video_info') or {} formats = [] @@ -461,6 +498,10 @@ class TwitterIE(TwitterBaseIE): 'thumbnails': thumbnails, 'duration': float_or_none(video_info.get('duration_millis'), 1000), }) + + media = try_get(status, lambda x: x['extended_entities']['media'][0]) + if media and media.get('type') != 'photo': + extract_from_video_info(media) else: card = status.get('card') if card: @@ -493,7 +534,12 @@ class TwitterIE(TwitterBaseIE): '_type': 'url', 'url': get_binding_value('card_url'), }) - # amplify, promo_video_website, promo_video_convo, appplayer, ... + elif card_name == 'unified_card': + media_entities = self._parse_json(get_binding_value('unified_card'), twid)['media_entities'] + extract_from_video_info(next(iter(media_entities.values()))) + # amplify, promo_video_website, promo_video_convo, appplayer, + # video_direct_message, poll2choice_video, poll3choice_video, + # poll4choice_video, ... else: is_amplify = card_name == 'amplify' vmap_url = get_binding_value('amplify_url_vmap') if is_amplify else get_binding_value('player_stream_url') diff --git a/youtube_dl/extractor/umg.py b/youtube_dl/extractor/umg.py index d815cd9a6..47948b6ce 100644 --- a/youtube_dl/extractor/umg.py +++ b/youtube_dl/extractor/umg.py @@ -28,7 +28,7 @@ class UMGDeIE(InfoExtractor): def _real_extract(self, url): video_id = self._match_id(url) video_data = self._download_json( - 'https://api.universal-music.de/graphql', + 'https://graphql.universal-music.de/', video_id, query={ 'query': '''{ universalMusic(channel:16) { @@ -56,11 +56,9 @@ class UMGDeIE(InfoExtractor): formats = [] def add_m3u8_format(format_id): - m3u8_formats = self._extract_m3u8_formats( + formats.extend(self._extract_m3u8_formats( hls_url_template % format_id, video_id, 'mp4', - 'm3u8_native', m3u8_id='hls', fatal='False') - if m3u8_formats and m3u8_formats[0].get('height'): - formats.extend(m3u8_formats) + 'm3u8_native', m3u8_id='hls', fatal=False)) for f in video_data.get('formats', []): f_url = f.get('url') diff --git a/youtube_dl/extractor/urplay.py b/youtube_dl/extractor/urplay.py index 10b817760..d6c79147e 100644 --- a/youtube_dl/extractor/urplay.py +++ b/youtube_dl/extractor/urplay.py @@ -21,6 +21,11 @@ class URPlayIE(InfoExtractor): 'description': 'md5:5344508a52aa78c1ced6c1b8b9e44e9a', 'timestamp': 1513292400, 'upload_date': '20171214', + 'series': 'UR Samtiden - Livet, universum och rymdens märkliga musik', + 'duration': 2269, + 'categories': ['Kultur & historia'], + 'tags': ['Kritiskt tänkande', 'Vetenskap', 'Vetenskaplig verksamhet'], + 'episode': 'Om vetenskap, kritiskt tänkande och motstånd', }, }, { 'url': 'https://urskola.se/Produkter/190031-Tripp-Trapp-Trad-Sovkudde', @@ -31,6 +36,10 @@ class URPlayIE(InfoExtractor): 'description': 'md5:b86bffdae04a7e9379d1d7e5947df1d1', 'timestamp': 1440086400, 'upload_date': '20150820', + 'series': 'Tripp, Trapp, Träd', + 'duration': 865, + 'tags': ['Sova'], + 'episode': 'Sovkudde', }, }, { 'url': 'http://urskola.se/Produkter/155794-Smasagor-meankieli-Grodan-i-vida-varlden', @@ -41,9 +50,11 @@ class URPlayIE(InfoExtractor): video_id = self._match_id(url) url = url.replace('skola.se/Produkter', 'play.se/program') webpage = self._download_webpage(url, video_id) - urplayer_data = self._parse_json(self._html_search_regex( - r'data-react-class="components/Player/Player"[^>]+data-react-props="({.+?})"', - webpage, 'urplayer data'), video_id)['currentProduct'] + vid = int(video_id) + accessible_episodes = self._parse_json(self._html_search_regex( + r'data-react-class="routes/Product/components/ProgramContainer/ProgramContainer"[^>]+data-react-props="({.+?})"', + webpage, 'urplayer data'), video_id)['accessibleEpisodes'] + urplayer_data = next(e for e in accessible_episodes if e.get('id') == vid) episode = urplayer_data['title'] raw_streaming_info = urplayer_data['streamingInfo']['raw'] host = self._download_json( diff --git a/youtube_dl/extractor/ustream.py b/youtube_dl/extractor/ustream.py index 9e860aeb7..1e29cbe22 100644 --- a/youtube_dl/extractor/ustream.py +++ b/youtube_dl/extractor/ustream.py @@ -75,7 +75,7 @@ class UstreamIE(InfoExtractor): @staticmethod def _extract_url(webpage): mobj = re.search( - r'<iframe[^>]+?src=(["\'])(?P<url>http://(?:www\.)?(?:ustream\.tv|video\.ibm\.com)/embed/.+?)\1', webpage) + r'<iframe[^>]+?src=(["\'])(?P<url>https?://(?:www\.)?(?:ustream\.tv|video\.ibm\.com)/embed/.+?)\1', webpage) if mobj is not None: return mobj.group('url') diff --git a/youtube_dl/extractor/vgtv.py b/youtube_dl/extractor/vgtv.py index fe7a26b62..22e99e8f0 100644 --- a/youtube_dl/extractor/vgtv.py +++ b/youtube_dl/extractor/vgtv.py @@ -23,6 +23,8 @@ class VGTVIE(XstreamIE): 'fvn.no/fvntv': 'fvntv', 'aftenposten.no/webtv': 'aptv', 'ap.vgtv.no/webtv': 'aptv', + 'tv.aftonbladet.se': 'abtv', + # obsolete URL schemas, kept in order to save one HTTP redirect 'tv.aftonbladet.se/abtv': 'abtv', 'www.aftonbladet.se/tv': 'abtv', } @@ -141,6 +143,10 @@ class VGTVIE(XstreamIE): 'only_matching': True, }, { + 'url': 'https://tv.aftonbladet.se/video/36015/vulkanutbrott-i-rymden-nu-slapper-nasa-bilderna', + 'only_matching': True, + }, + { 'url': 'http://tv.aftonbladet.se/abtv/articles/36015', 'only_matching': True, }, diff --git a/youtube_dl/extractor/videopress.py b/youtube_dl/extractor/videopress.py index e5f964d39..6376ff096 100644 --- a/youtube_dl/extractor/videopress.py +++ b/youtube_dl/extractor/videopress.py @@ -4,21 +4,22 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..compat import compat_str from ..utils import ( determine_ext, float_or_none, + int_or_none, parse_age_limit, qualities, random_birthday, - try_get, unified_timestamp, urljoin, ) class VideoPressIE(InfoExtractor): - _VALID_URL = r'https?://videopress\.com/embed/(?P<id>[\da-zA-Z]+)' + _ID_REGEX = r'[\da-zA-Z]{8}' + _PATH_REGEX = r'video(?:\.word)?press\.com/embed/' + _VALID_URL = r'https?://%s(?P<id>%s)' % (_PATH_REGEX, _ID_REGEX) _TESTS = [{ 'url': 'https://videopress.com/embed/kUJmAcSf', 'md5': '706956a6c875873d51010921310e4bc6', @@ -36,35 +37,36 @@ class VideoPressIE(InfoExtractor): # 17+, requires birth_* params 'url': 'https://videopress.com/embed/iH3gstfZ', 'only_matching': True, + }, { + 'url': 'https://video.wordpress.com/embed/kUJmAcSf', + 'only_matching': True, }] @staticmethod def _extract_urls(webpage): return re.findall( - r'<iframe[^>]+src=["\']((?:https?://)?videopress\.com/embed/[\da-zA-Z]+)', + r'<iframe[^>]+src=["\']((?:https?://)?%s%s)' % (VideoPressIE._PATH_REGEX, VideoPressIE._ID_REGEX), webpage) def _real_extract(self, url): video_id = self._match_id(url) query = random_birthday('birth_year', 'birth_month', 'birth_day') + query['fields'] = 'description,duration,file_url_base,files,height,original,poster,rating,title,upload_date,width' video = self._download_json( 'https://public-api.wordpress.com/rest/v1.1/videos/%s' % video_id, video_id, query=query) title = video['title'] - def base_url(scheme): - return try_get( - video, lambda x: x['file_url_base'][scheme], compat_str) - - base_url = base_url('https') or base_url('http') + file_url_base = video.get('file_url_base') or {} + base_url = file_url_base.get('https') or file_url_base.get('http') QUALITIES = ('std', 'dvd', 'hd') quality = qualities(QUALITIES) formats = [] - for format_id, f in video['files'].items(): + for format_id, f in (video.get('files') or {}).items(): if not isinstance(f, dict): continue for ext, path in f.items(): @@ -75,12 +77,14 @@ class VideoPressIE(InfoExtractor): 'ext': determine_ext(path, ext), 'quality': quality(format_id), }) - original_url = try_get(video, lambda x: x['original'], compat_str) + original_url = video.get('original') if original_url: formats.append({ 'url': original_url, 'format_id': 'original', 'quality': len(QUALITIES), + 'width': int_or_none(video.get('width')), + 'height': int_or_none(video.get('height')), }) self._sort_formats(formats) diff --git a/youtube_dl/extractor/vidio.py b/youtube_dl/extractor/vidio.py index b48baf00b..b1243e847 100644 --- a/youtube_dl/extractor/vidio.py +++ b/youtube_dl/extractor/vidio.py @@ -4,7 +4,13 @@ from __future__ import unicode_literals import re from .common import InfoExtractor -from ..utils import int_or_none +from ..utils import ( + int_or_none, + parse_iso8601, + str_or_none, + strip_or_none, + try_get, +) class VidioIE(InfoExtractor): @@ -21,57 +27,63 @@ class VidioIE(InfoExtractor): 'thumbnail': r're:^https?://.*\.jpg$', 'duration': 149, 'like_count': int, + 'uploader': 'TWELVE Pic', + 'timestamp': 1444902800, + 'upload_date': '20151015', + 'uploader_id': 'twelvepictures', + 'channel': 'Cover Music Video', + 'channel_id': '280236', + 'view_count': int, + 'dislike_count': int, + 'comment_count': int, + 'tags': 'count:4', }, }, { 'url': 'https://www.vidio.com/watch/77949-south-korea-test-fires-missile-that-can-strike-all-of-the-north', 'only_matching': True, }] - def _real_extract(self, url): - mobj = re.match(self._VALID_URL, url) - video_id, display_id = mobj.group('id', 'display_id') - - webpage = self._download_webpage(url, display_id) - - title = self._og_search_title(webpage) + def _real_initialize(self): + self._api_key = self._download_json( + 'https://www.vidio.com/auth', None, data=b'')['api_key'] - m3u8_url, duration, thumbnail = [None] * 3 - - clips = self._parse_json( - self._html_search_regex( - r'data-json-clips\s*=\s*(["\'])(?P<data>\[.+?\])\1', - webpage, 'video data', default='[]', group='data'), - display_id, fatal=False) - if clips: - clip = clips[0] - m3u8_url = clip.get('sources', [{}])[0].get('file') - duration = clip.get('clip_duration') - thumbnail = clip.get('image') + def _real_extract(self, url): + video_id, display_id = re.match(self._VALID_URL, url).groups() + data = self._download_json( + 'https://api.vidio.com/videos/' + video_id, display_id, headers={ + 'Content-Type': 'application/vnd.api+json', + 'X-API-KEY': self._api_key, + }) + video = data['videos'][0] + title = video['title'].strip() - m3u8_url = m3u8_url or self._search_regex( - r'data(?:-vjs)?-clip-hls-url=(["\'])(?P<url>(?:(?!\1).)+)\1', - webpage, 'hls url', group='url') formats = self._extract_m3u8_formats( - m3u8_url, display_id, 'mp4', entry_protocol='m3u8_native') + data['clips'][0]['hls_url'], display_id, 'mp4', 'm3u8_native') self._sort_formats(formats) - duration = int_or_none(duration or self._search_regex( - r'data-video-duration=(["\'])(?P<duration>\d+)\1', webpage, - 'duration', fatal=False, group='duration')) - thumbnail = thumbnail or self._og_search_thumbnail(webpage) - - like_count = int_or_none(self._search_regex( - (r'<span[^>]+data-comment-vote-count=["\'](\d+)', - r'<span[^>]+class=["\'].*?\blike(?:__|-)count\b.*?["\'][^>]*>\s*(\d+)'), - webpage, 'like count', fatal=False)) + get_first = lambda x: try_get(data, lambda y: y[x + 's'][0], dict) or {} + channel = get_first('channel') + user = get_first('user') + username = user.get('username') + get_count = lambda x: int_or_none(video.get('total_' + x)) return { 'id': video_id, 'display_id': display_id, 'title': title, - 'description': self._og_search_description(webpage), - 'thumbnail': thumbnail, - 'duration': duration, - 'like_count': like_count, + 'description': strip_or_none(video.get('description')), + 'thumbnail': video.get('image_url_medium'), + 'duration': int_or_none(video.get('duration')), + 'like_count': get_count('likes'), 'formats': formats, + 'uploader': user.get('name'), + 'timestamp': parse_iso8601(video.get('created_at')), + 'uploader_id': username, + 'uploader_url': 'https://www.vidio.com/@' + username if username else None, + 'channel': channel.get('name'), + 'channel_id': str_or_none(channel.get('id')), + 'view_count': get_count('view_count'), + 'dislike_count': get_count('dislikes'), + 'comment_count': get_count('comments'), + 'tags': video.get('tag_list'), } diff --git a/youtube_dl/extractor/vidzi.py b/youtube_dl/extractor/vidzi.py deleted file mode 100644 index 42ea4952c..000000000 --- a/youtube_dl/extractor/vidzi.py +++ /dev/null @@ -1,68 +0,0 @@ -# coding: utf-8 -from __future__ import unicode_literals - -import re - -from .common import InfoExtractor -from ..utils import ( - decode_packed_codes, - js_to_json, - NO_DEFAULT, - PACKED_CODES_RE, -) - - -class VidziIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?vidzi\.(?:tv|cc|si|nu)/(?:embed-)?(?P<id>[0-9a-zA-Z]+)' - _TESTS = [{ - 'url': 'http://vidzi.tv/cghql9yq6emu.html', - 'md5': '4f16c71ca0c8c8635ab6932b5f3f1660', - 'info_dict': { - 'id': 'cghql9yq6emu', - 'ext': 'mp4', - 'title': 'youtube-dl test video 1\\\\2\'3/4<5\\\\6ä7↭', - }, - 'params': { - # m3u8 download - 'skip_download': True, - }, - }, { - 'url': 'http://vidzi.tv/embed-4z2yb0rzphe9-600x338.html', - 'only_matching': True, - }, { - 'url': 'http://vidzi.cc/cghql9yq6emu.html', - 'only_matching': True, - }, { - 'url': 'https://vidzi.si/rph9gztxj1et.html', - 'only_matching': True, - }, { - 'url': 'http://vidzi.nu/cghql9yq6emu.html', - 'only_matching': True, - }] - - def _real_extract(self, url): - video_id = self._match_id(url) - - webpage = self._download_webpage( - 'http://vidzi.tv/%s' % video_id, video_id) - title = self._html_search_regex( - r'(?s)<h2 class="video-title">(.*?)</h2>', webpage, 'title') - - codes = [webpage] - codes.extend([ - decode_packed_codes(mobj.group(0)).replace('\\\'', '\'') - for mobj in re.finditer(PACKED_CODES_RE, webpage)]) - for num, code in enumerate(codes, 1): - jwplayer_data = self._parse_json( - self._search_regex( - r'setup\(([^)]+)\)', code, 'jwplayer data', - default=NO_DEFAULT if num == len(codes) else '{}'), - video_id, transform_source=lambda s: js_to_json( - re.sub(r'\s*\+\s*window\[.+?\]', '', s))) - if jwplayer_data: - break - - info_dict = self._parse_jwplayer_data(jwplayer_data, video_id, require_title=False) - info_dict['title'] = title - - return info_dict diff --git a/youtube_dl/extractor/viki.py b/youtube_dl/extractor/viki.py index a311f21ef..2e9cbf148 100644 --- a/youtube_dl/extractor/viki.py +++ b/youtube_dl/extractor/viki.py @@ -21,6 +21,7 @@ from ..utils import ( parse_iso8601, sanitized_Request, std_headers, + try_get, ) @@ -30,7 +31,7 @@ class VikiBaseIE(InfoExtractor): _API_URL_TEMPLATE = 'https://api.viki.io%s&sig=%s' _APP = '100005a' - _APP_VERSION = '2.2.5.1428709186' + _APP_VERSION = '6.0.0' _APP_SECRET = 'MM_d*yP@`&1@]@!AVrXf_o-HVEnoTnm$O-ti4[G~$JDI/Dc-&piU&z&5.;:}95=Iad' _GEO_BYPASS = False @@ -41,7 +42,7 @@ class VikiBaseIE(InfoExtractor): _ERRORS = { 'geo': 'Sorry, this content is not available in your region.', 'upcoming': 'Sorry, this content is not yet available.', - # 'paywall': 'paywall', + 'paywall': 'Sorry, this content is only available to Viki Pass Plus subscribers', } def _prepare_call(self, path, timestamp=None, post_data=None): @@ -62,7 +63,8 @@ class VikiBaseIE(InfoExtractor): def _call_api(self, path, video_id, note, timestamp=None, post_data=None): resp = self._download_json( - self._prepare_call(path, timestamp, post_data), video_id, note) + self._prepare_call(path, timestamp, post_data), video_id, note, + headers={'x-viki-app-ver': self._APP_VERSION}) error = resp.get('error') if error: @@ -82,11 +84,13 @@ class VikiBaseIE(InfoExtractor): expected=True) def _check_errors(self, data): - for reason, status in data.get('blocking', {}).items(): + for reason, status in (data.get('blocking') or {}).items(): if status and reason in self._ERRORS: message = self._ERRORS[reason] if reason == 'geo': self.raise_geo_restricted(msg=message) + elif reason == 'paywall': + self.raise_login_required(message) raise ExtractorError('%s said: %s' % ( self.IE_NAME, message), expected=True) @@ -131,13 +135,19 @@ class VikiIE(VikiBaseIE): 'info_dict': { 'id': '1023585v', 'ext': 'mp4', - 'title': 'Heirs Episode 14', - 'uploader': 'SBS', - 'description': 'md5:c4b17b9626dd4b143dcc4d855ba3474e', + 'title': 'Heirs - Episode 14', + 'uploader': 'SBS Contents Hub', + 'timestamp': 1385047627, 'upload_date': '20131121', 'age_limit': 13, + 'duration': 3570, + 'episode_number': 14, + }, + 'params': { + 'format': 'bestvideo', }, 'skip': 'Blocked in the US', + 'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'], }, { # clip 'url': 'http://www.viki.com/videos/1067139v-the-avengers-age-of-ultron-press-conference', @@ -153,7 +163,8 @@ class VikiIE(VikiBaseIE): 'uploader': 'Arirang TV', 'like_count': int, 'age_limit': 0, - } + }, + 'skip': 'Sorry. There was an error loading this video', }, { 'url': 'http://www.viki.com/videos/1048879v-ankhon-dekhi', 'info_dict': { @@ -171,7 +182,7 @@ class VikiIE(VikiBaseIE): }, { # episode 'url': 'http://www.viki.com/videos/44699v-boys-over-flowers-episode-1', - 'md5': '94e0e34fd58f169f40c184f232356cfe', + 'md5': '0a53dc252e6e690feccd756861495a8c', 'info_dict': { 'id': '44699v', 'ext': 'mp4', @@ -183,6 +194,10 @@ class VikiIE(VikiBaseIE): 'uploader': 'group8', 'like_count': int, 'age_limit': 13, + 'episode_number': 1, + }, + 'params': { + 'format': 'bestvideo', }, 'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'], }, { @@ -209,7 +224,7 @@ class VikiIE(VikiBaseIE): }, { # non-English description 'url': 'http://www.viki.com/videos/158036v-love-in-magic', - 'md5': 'adf9e321a0ae5d0aace349efaaff7691', + 'md5': '41faaba0de90483fb4848952af7c7d0d', 'info_dict': { 'id': '158036v', 'ext': 'mp4', @@ -220,6 +235,10 @@ class VikiIE(VikiBaseIE): 'title': 'Love In Magic', 'age_limit': 13, }, + 'params': { + 'format': 'bestvideo', + }, + 'expected_warnings': ['Unknown MIME type image/jpeg in DASH manifest'], }] def _real_extract(self, url): @@ -229,36 +248,33 @@ class VikiIE(VikiBaseIE): 'https://www.viki.com/api/videos/' + video_id, video_id, 'Downloading video JSON', headers={ 'x-client-user-agent': std_headers['User-Agent'], - 'x-viki-app-ver': '4.0.57', + 'x-viki-app-ver': '3.0.0', }) video = resp['video'] self._check_errors(video) title = self.dict_selection(video.get('titles', {}), 'en', allow_fallback=False) + episode_number = int_or_none(video.get('number')) if not title: - title = 'Episode %d' % video.get('number') if video.get('type') == 'episode' else video.get('id') or video_id - container_titles = video.get('container', {}).get('titles', {}) + title = 'Episode %d' % episode_number if video.get('type') == 'episode' else video.get('id') or video_id + container_titles = try_get(video, lambda x: x['container']['titles'], dict) or {} container_title = self.dict_selection(container_titles, 'en') title = '%s - %s' % (container_title, title) description = self.dict_selection(video.get('descriptions', {}), 'en') - duration = int_or_none(video.get('duration')) - timestamp = parse_iso8601(video.get('created_at')) - uploader = video.get('author') - like_count = int_or_none(video.get('likes', {}).get('count')) - age_limit = parse_age_limit(video.get('rating')) + like_count = int_or_none(try_get(video, lambda x: x['likes']['count'])) thumbnails = [] - for thumbnail_id, thumbnail in video.get('images', {}).items(): + for thumbnail_id, thumbnail in (video.get('images') or {}).items(): thumbnails.append({ 'id': thumbnail_id, 'url': thumbnail.get('url'), }) subtitles = {} - for subtitle_lang, _ in video.get('subtitle_completions', {}).items(): + for subtitle_lang, _ in (video.get('subtitle_completions') or {}).items(): subtitles[subtitle_lang] = [{ 'ext': subtitles_format, 'url': self._prepare_call( @@ -269,13 +285,15 @@ class VikiIE(VikiBaseIE): 'id': video_id, 'title': title, 'description': description, - 'duration': duration, - 'timestamp': timestamp, - 'uploader': uploader, + 'duration': int_or_none(video.get('duration')), + 'timestamp': parse_iso8601(video.get('created_at')), + 'uploader': video.get('author'), + 'uploader_url': video.get('author_url'), 'like_count': like_count, - 'age_limit': age_limit, + 'age_limit': parse_age_limit(video.get('rating')), 'thumbnails': thumbnails, 'subtitles': subtitles, + 'episode_number': episode_number, } formats = [] @@ -360,7 +378,7 @@ class VikiChannelIE(VikiBaseIE): 'info_dict': { 'id': '50c', 'title': 'Boys Over Flowers', - 'description': 'md5:ecd3cff47967fe193cff37c0bec52790', + 'description': 'md5:804ce6e7837e1fd527ad2f25420f4d59', }, 'playlist_mincount': 71, }, { @@ -371,6 +389,7 @@ class VikiChannelIE(VikiBaseIE): 'description': 'md5:05bf5471385aa8b21c18ad450e350525', }, 'playlist_count': 127, + 'skip': 'Page not found', }, { 'url': 'http://www.viki.com/news/24569c-showbiz-korea', 'only_matching': True, diff --git a/youtube_dl/extractor/vimeo.py b/youtube_dl/extractor/vimeo.py index 15cd06268..0b386f450 100644 --- a/youtube_dl/extractor/vimeo.py +++ b/youtube_dl/extractor/vimeo.py @@ -3,7 +3,6 @@ from __future__ import unicode_literals import base64 import functools -import json import re import itertools @@ -17,14 +16,14 @@ from ..compat import ( from ..utils import ( clean_html, determine_ext, - dict_get, ExtractorError, + get_element_by_class, js_to_json, int_or_none, merge_dicts, OnDemandPagedList, parse_filesize, - RegexNotFoundError, + parse_iso8601, sanitized_Request, smuggle_url, std_headers, @@ -74,25 +73,28 @@ class VimeoBaseInfoExtractor(InfoExtractor): expected=True) raise ExtractorError('Unable to log in') - def _verify_video_password(self, url, video_id, webpage): + def _get_video_password(self): password = self._downloader.params.get('videopassword') if password is None: - raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) - token, vuid = self._extract_xsrft_and_vuid(webpage) - data = urlencode_postdata({ - 'password': password, - 'token': token, - }) + raise ExtractorError( + 'This video is protected by a password, use the --video-password option', + expected=True) + return password + + def _verify_video_password(self, url, video_id, password, token, vuid): if url.startswith('http://'): # vimeo only supports https now, but the user can give an http url url = url.replace('http://', 'https://') - password_request = sanitized_Request(url + '/password', data) - password_request.add_header('Content-Type', 'application/x-www-form-urlencoded') - password_request.add_header('Referer', url) self._set_vimeo_cookie('vuid', vuid) return self._download_webpage( - password_request, video_id, - 'Verifying the password', 'Wrong password') + url + '/password', video_id, 'Verifying the password', + 'Wrong password', data=urlencode_postdata({ + 'password': password, + 'token': token, + }), headers={ + 'Content-Type': 'application/x-www-form-urlencoded', + 'Referer': url, + }) def _extract_xsrft_and_vuid(self, webpage): xsrft = self._search_regex( @@ -123,10 +125,11 @@ class VimeoBaseInfoExtractor(InfoExtractor): video_title = video_data['title'] live_event = video_data.get('live_event') or {} is_live = live_event.get('status') == 'started' + request = config.get('request') or {} formats = [] - config_files = video_data.get('files') or config['request'].get('files', {}) - for f in config_files.get('progressive', []): + config_files = video_data.get('files') or request.get('files') or {} + for f in (config_files.get('progressive') or []): video_url = f.get('url') if not video_url: continue @@ -142,7 +145,7 @@ class VimeoBaseInfoExtractor(InfoExtractor): # TODO: fix handling of 308 status code returned for live archive manifest requests sep_pattern = r'/sep/video/' for files_type in ('hls', 'dash'): - for cdn_name, cdn_data in config_files.get(files_type, {}).get('cdns', {}).items(): + for cdn_name, cdn_data in (try_get(config_files, lambda x: x[files_type]['cdns']) or {}).items(): manifest_url = cdn_data.get('url') if not manifest_url: continue @@ -188,17 +191,15 @@ class VimeoBaseInfoExtractor(InfoExtractor): f['preference'] = -40 subtitles = {} - text_tracks = config['request'].get('text_tracks') - if text_tracks: - for tt in text_tracks: - subtitles[tt['lang']] = [{ - 'ext': 'vtt', - 'url': urljoin('https://vimeo.com', tt['url']), - }] + for tt in (request.get('text_tracks') or []): + subtitles[tt['lang']] = [{ + 'ext': 'vtt', + 'url': urljoin('https://vimeo.com', tt['url']), + }] thumbnails = [] if not is_live: - for key, thumb in video_data.get('thumbs', {}).items(): + for key, thumb in (video_data.get('thumbs') or {}).items(): thumbnails.append({ 'id': key, 'width': int_or_none(key), @@ -226,10 +227,12 @@ class VimeoBaseInfoExtractor(InfoExtractor): 'is_live': is_live, } - def _extract_original_format(self, url, video_id): + def _extract_original_format(self, url, video_id, unlisted_hash=None): + query = {'action': 'load_download_config'} + if unlisted_hash: + query['unlisted_hash'] = unlisted_hash download_data = self._download_json( - url, video_id, fatal=False, - query={'action': 'load_download_config'}, + url, video_id, fatal=False, query=query, headers={'X-Requested-With': 'XMLHttpRequest'}) if download_data: source_file = download_data.get('source_file') @@ -276,7 +279,7 @@ class VimeoIE(VimeoBaseInfoExtractor): )? (?:videos?/)? (?P<id>[0-9]+) - (?:/[\da-f]+)? + (?:/(?P<unlisted_hash>[\da-f]{10}))? /?(?:[?&].*)?(?:[#].*)?$ ''' IE_NAME = 'vimeo' @@ -316,6 +319,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'duration': 1595, 'upload_date': '20130610', 'timestamp': 1370893156, + 'license': 'by', }, 'params': { 'format': 'best[protocol=https]', @@ -329,9 +333,9 @@ class VimeoIE(VimeoBaseInfoExtractor): 'id': '54469442', 'ext': 'mp4', 'title': 'Kathy Sierra: Building the minimum Badass User, Business of Software 2012', - 'uploader': 'The BLN & Business of Software', - 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/theblnbusinessofsoftware', - 'uploader_id': 'theblnbusinessofsoftware', + 'uploader': 'Business of Software', + 'uploader_url': r're:https?://(?:www\.)?vimeo\.com/businessofsoftware', + 'uploader_id': 'businessofsoftware', 'duration': 3610, 'description': None, }, @@ -394,6 +398,12 @@ class VimeoIE(VimeoBaseInfoExtractor): 'uploader_id': 'staff', 'uploader': 'Vimeo Staff', 'duration': 62, + 'subtitles': { + 'de': [{'ext': 'vtt'}], + 'en': [{'ext': 'vtt'}], + 'es': [{'ext': 'vtt'}], + 'fr': [{'ext': 'vtt'}], + }, } }, { @@ -466,6 +476,7 @@ class VimeoIE(VimeoBaseInfoExtractor): 'skip_download': True, }, 'expected_warnings': ['Unable to download JSON metadata'], + 'skip': 'this page is no longer available.', }, { 'url': 'http://player.vimeo.com/video/68375962', @@ -509,6 +520,11 @@ class VimeoIE(VimeoBaseInfoExtractor): { 'url': 'https://vimeo.com/160743502/abd0e13fb4', 'only_matching': True, + }, + { + # requires passing unlisted_hash(a52724358e) to load_download_config request + 'url': 'https://vimeo.com/392479337/a52724358e', + 'only_matching': True, } # https://gettingthingsdone.com/workflowmap/ # vimeo embed with check-password page protected by Referer header @@ -543,9 +559,7 @@ class VimeoIE(VimeoBaseInfoExtractor): return urls[0] if urls else None def _verify_player_video_password(self, url, video_id, headers): - password = self._downloader.params.get('videopassword') - if password is None: - raise ExtractorError('This video is protected by a password, use the --video-password option', expected=True) + password = self._get_video_password() data = urlencode_postdata({ 'password': base64.b64encode(password.encode()), }) @@ -562,6 +576,37 @@ class VimeoIE(VimeoBaseInfoExtractor): def _real_initialize(self): self._login() + def _extract_from_api(self, video_id, unlisted_hash=None): + token = self._download_json( + 'https://vimeo.com/_rv/jwt', video_id, headers={ + 'X-Requested-With': 'XMLHttpRequest' + })['token'] + api_url = 'https://api.vimeo.com/videos/' + video_id + if unlisted_hash: + api_url += ':' + unlisted_hash + video = self._download_json( + api_url, video_id, headers={ + 'Authorization': 'jwt ' + token, + }, query={ + 'fields': 'config_url,created_time,description,license,metadata.connections.comments.total,metadata.connections.likes.total,release_time,stats.plays', + }) + info = self._parse_config(self._download_json( + video['config_url'], video_id), video_id) + self._vimeo_sort_formats(info['formats']) + get_timestamp = lambda x: parse_iso8601(video.get(x + '_time')) + info.update({ + 'description': video.get('description'), + 'license': video.get('license'), + 'release_timestamp': get_timestamp('release'), + 'timestamp': get_timestamp('created'), + 'view_count': int_or_none(try_get(video, lambda x: x['stats']['plays'])), + }) + connections = try_get( + video, lambda x: x['metadata']['connections'], dict) or {} + for k in ('comment', 'like'): + info[k + '_count'] = int_or_none(try_get(connections, lambda x: x[k + 's']['total'])) + return info + def _real_extract(self, url): url, data = unsmuggle_url(url, {}) headers = std_headers.copy() @@ -570,22 +615,19 @@ class VimeoIE(VimeoBaseInfoExtractor): if 'Referer' not in headers: headers['Referer'] = url - channel_id = self._search_regex( - r'vimeo\.com/channels/([^/]+)', url, 'channel id', default=None) + mobj = re.match(self._VALID_URL, url).groupdict() + video_id, unlisted_hash = mobj['id'], mobj.get('unlisted_hash') + if unlisted_hash: + return self._extract_from_api(video_id, unlisted_hash) - # Extract ID from URL - video_id = self._match_id(url) orig_url = url is_pro = 'vimeopro.com/' in url - is_player = '://player.vimeo.com/video/' in url if is_pro: # some videos require portfolio_id to be present in player url # https://github.com/ytdl-org/youtube-dl/issues/20070 url = self._extract_url(url, self._download_webpage(url, video_id)) if not url: url = 'https://vimeo.com/' + video_id - elif is_player: - url = 'https://player.vimeo.com/video/' + video_id elif any(p in url for p in ('play_redirect_hls', 'moogaloop.swf')): url = 'https://vimeo.com/' + video_id @@ -605,14 +647,25 @@ class VimeoIE(VimeoBaseInfoExtractor): expected=True) raise - # Now we begin extracting as much information as we can from what we - # retrieved. First we extract the information common to all extractors, - # and latter we extract those that are Vimeo specific. - self.report_extraction(video_id) + if '//player.vimeo.com/video/' in url: + config = self._parse_json(self._search_regex( + r'\bconfig\s*=\s*({.+?})\s*;', webpage, 'info section'), video_id) + if config.get('view') == 4: + config = self._verify_player_video_password( + redirect_url, video_id, headers) + info = self._parse_config(config, video_id) + self._vimeo_sort_formats(info['formats']) + return info + + if re.search(r'<form[^>]+?id="pw_form"', webpage): + video_password = self._get_video_password() + token, vuid = self._extract_xsrft_and_vuid(webpage) + webpage = self._verify_video_password( + redirect_url, video_id, video_password, token, vuid) vimeo_config = self._extract_vimeo_config(webpage, video_id, default=None) if vimeo_config: - seed_status = vimeo_config.get('seed_status', {}) + seed_status = vimeo_config.get('seed_status') or {} if seed_status.get('state') == 'failed': raise ExtractorError( '%s said: %s' % (self.IE_NAME, seed_status['title']), @@ -621,66 +674,40 @@ class VimeoIE(VimeoBaseInfoExtractor): cc_license = None timestamp = None video_description = None + info_dict = {} - # Extract the config JSON - try: - try: - config_url = self._html_search_regex( - r' data-config-url="(.+?)"', webpage, - 'config URL', default=None) - if not config_url: - # Sometimes new react-based page is served instead of old one that require - # different config URL extraction approach (see - # https://github.com/ytdl-org/youtube-dl/pull/7209) - page_config = self._parse_json(self._search_regex( - r'vimeo\.(?:clip|vod_title)_page_config\s*=\s*({.+?});', - webpage, 'page config'), video_id) - config_url = page_config['player']['config_url'] - cc_license = page_config.get('cc_license') - timestamp = try_get( - page_config, lambda x: x['clip']['uploaded_on'], - compat_str) - video_description = clean_html(dict_get( - page_config, ('description', 'description_html_escaped'))) - config = self._download_json(config_url, video_id) - except RegexNotFoundError: - # For pro videos or player.vimeo.com urls - # We try to find out to which variable is assigned the config dic - m_variable_name = re.search(r'(\w)\.video\.id', webpage) - if m_variable_name is not None: - config_re = [r'%s=({[^}].+?});' % re.escape(m_variable_name.group(1))] - else: - config_re = [r' = {config:({.+?}),assets:', r'(?:[abc])=({.+?});'] - config_re.append(r'\bvar\s+r\s*=\s*({.+?})\s*;') - config_re.append(r'\bconfig\s*=\s*({.+?})\s*;') - config = self._search_regex(config_re, webpage, 'info section', - flags=re.DOTALL) - config = json.loads(config) - except Exception as e: - if re.search('The creator of this video has not given you permission to embed it on this domain.', webpage): - raise ExtractorError('The author has restricted the access to this video, try with the "--referer" option') - - if re.search(r'<form[^>]+?id="pw_form"', webpage) is not None: - if '_video_password_verified' in data: - raise ExtractorError('video password verification failed!') - self._verify_video_password(redirect_url, video_id, webpage) - return self._real_extract( - smuggle_url(redirect_url, {'_video_password_verified': 'verified'})) - else: - raise ExtractorError('Unable to extract info section', - cause=e) + channel_id = self._search_regex( + r'vimeo\.com/channels/([^/]+)', url, 'channel id', default=None) + if channel_id: + config_url = self._html_search_regex( + r'\bdata-config-url="([^"]+)"', webpage, 'config URL') + video_description = clean_html(get_element_by_class('description', webpage)) + info_dict.update({ + 'channel_id': channel_id, + 'channel_url': 'https://vimeo.com/channels/' + channel_id, + }) else: - if config.get('view') == 4: - config = self._verify_player_video_password(redirect_url, video_id, headers) - - vod = config.get('video', {}).get('vod', {}) + page_config = self._parse_json(self._search_regex( + r'vimeo\.(?:clip|vod_title)_page_config\s*=\s*({.+?});', + webpage, 'page config', default='{}'), video_id, fatal=False) + if not page_config: + return self._extract_from_api(video_id) + config_url = page_config['player']['config_url'] + cc_license = page_config.get('cc_license') + clip = page_config.get('clip') or {} + timestamp = clip.get('uploaded_on') + video_description = clean_html( + clip.get('description') or page_config.get('description_html_escaped')) + config = self._download_json(config_url, video_id) + video = config.get('video') or {} + vod = video.get('vod') or {} def is_rented(): if '>You rented this title.<' in webpage: return True - if config.get('user', {}).get('purchased'): + if try_get(config, lambda x: x['user']['purchased']): return True - for purchase_option in vod.get('purchase_options', []): + for purchase_option in (vod.get('purchase_options') or []): if purchase_option.get('purchased'): return True label = purchase_option.get('label_string') @@ -695,14 +722,10 @@ class VimeoIE(VimeoBaseInfoExtractor): 'https://player.vimeo.com/player/%s' % feature_id, {'force_feature_id': True}), 'Vimeo') - # Extract video description - if not video_description: - video_description = self._html_search_regex( - r'(?s)<div\s+class="[^"]*description[^"]*"[^>]*>(.*?)</div>', - webpage, 'description', default=None) if not video_description: video_description = self._html_search_meta( - 'description', webpage, default=None) + ['description', 'og:description', 'twitter:description'], + webpage, default=None) if not video_description and is_pro: orig_webpage = self._download_webpage( orig_url, video_id, @@ -711,29 +734,18 @@ class VimeoIE(VimeoBaseInfoExtractor): if orig_webpage: video_description = self._html_search_meta( 'description', orig_webpage, default=None) - if not video_description and not is_player: + if not video_description: self._downloader.report_warning('Cannot find video description') - # Extract upload date if not timestamp: timestamp = self._search_regex( r'<time[^>]+datetime="([^"]+)"', webpage, 'timestamp', default=None) - try: - view_count = int(self._search_regex(r'UserPlays:(\d+)', webpage, 'view count')) - like_count = int(self._search_regex(r'UserLikes:(\d+)', webpage, 'like count')) - comment_count = int(self._search_regex(r'UserComments:(\d+)', webpage, 'comment count')) - except RegexNotFoundError: - # This info is only available in vimeo.com/{id} urls - view_count = None - like_count = None - comment_count = None - formats = [] source_format = self._extract_original_format( - 'https://vimeo.com/' + video_id, video_id) + 'https://vimeo.com/' + video_id, video_id, video.get('unlisted_hash')) if source_format: formats.append(source_format) @@ -748,29 +760,20 @@ class VimeoIE(VimeoBaseInfoExtractor): r'<link[^>]+rel=["\']license["\'][^>]+href=(["\'])(?P<license>(?:(?!\1).)+)\1', webpage, 'license', default=None, group='license') - channel_url = 'https://vimeo.com/channels/%s' % channel_id if channel_id else None - - info_dict = { + info_dict.update({ 'formats': formats, 'timestamp': unified_timestamp(timestamp), 'description': video_description, 'webpage_url': url, - 'view_count': view_count, - 'like_count': like_count, - 'comment_count': comment_count, 'license': cc_license, - 'channel_id': channel_id, - 'channel_url': channel_url, - } - - info_dict = merge_dicts(info_dict, info_dict_config, json_ld) + }) - return info_dict + return merge_dicts(info_dict, info_dict_config, json_ld) class VimeoOndemandIE(VimeoIE): IE_NAME = 'vimeo:ondemand' - _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/([^/]+/)?(?P<id>[^/?#&]+)' + _VALID_URL = r'https?://(?:www\.)?vimeo\.com/ondemand/(?:[^/]+/)?(?P<id>[^/?#&]+)' _TESTS = [{ # ondemand video not available via https://vimeo.com/id 'url': 'https://vimeo.com/ondemand/20704', @@ -931,11 +934,15 @@ class VimeoAlbumIE(VimeoBaseInfoExtractor): } if hashed_pass: query['_hashed_pass'] = hashed_pass - videos = self._download_json( - 'https://api.vimeo.com/albums/%s/videos' % album_id, - album_id, 'Downloading page %d' % api_page, query=query, headers={ - 'Authorization': 'jwt ' + authorization, - })['data'] + try: + videos = self._download_json( + 'https://api.vimeo.com/albums/%s/videos' % album_id, + album_id, 'Downloading page %d' % api_page, query=query, headers={ + 'Authorization': 'jwt ' + authorization, + })['data'] + except ExtractorError as e: + if isinstance(e.cause, compat_HTTPError) and e.cause.code == 400: + return for video in videos: link = video.get('link') if not link: @@ -1050,10 +1057,23 @@ class VimeoReviewIE(VimeoBaseInfoExtractor): def _real_extract(self, url): page_url, video_id = re.match(self._VALID_URL, url).groups() - clip_data = self._download_json( - page_url.replace('/review/', '/review/data/'), - video_id)['clipData'] - config_url = clip_data['configUrl'] + data = self._download_json( + page_url.replace('/review/', '/review/data/'), video_id) + if data.get('isLocked') is True: + video_password = self._get_video_password() + viewer = self._download_json( + 'https://vimeo.com/_rv/viewer', video_id) + webpage = self._verify_video_password( + 'https://vimeo.com/' + video_id, video_id, + video_password, viewer['xsrft'], viewer['vuid']) + clip_page_config = self._parse_json(self._search_regex( + r'window\.vimeo\.clip_page_config\s*=\s*({.+?});', + webpage, 'clip page config'), video_id) + config_url = clip_page_config['player']['config_url'] + clip_data = clip_page_config.get('clip') or {} + else: + clip_data = data['clipData'] + config_url = clip_data['configUrl'] config = self._download_json(config_url, video_id) info_dict = self._parse_config(config, video_id) source_format = self._extract_original_format( diff --git a/youtube_dl/extractor/vk.py b/youtube_dl/extractor/vk.py index 00ec006c4..6b3513ee0 100644 --- a/youtube_dl/extractor/vk.py +++ b/youtube_dl/extractor/vk.py @@ -300,6 +300,13 @@ class VKIE(VKBaseIE): 'only_matching': True, }] + @staticmethod + def _extract_sibnet_urls(webpage): + # https://help.sibnet.ru/?sibnet_video_embed + return [unescapeHTML(mobj.group('url')) for mobj in re.finditer( + r'<iframe\b[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//video\.sibnet\.ru/shell\.php\?.*?\bvideoid=\d+.*?)\1', + webpage)] + def _real_extract(self, url): mobj = re.match(self._VALID_URL, url) video_id = mobj.group('videoid') @@ -408,6 +415,10 @@ class VKIE(VKBaseIE): if odnoklassniki_url: return self.url_result(odnoklassniki_url, OdnoklassnikiIE.ie_key()) + sibnet_urls = self._extract_sibnet_urls(info_page) + if sibnet_urls: + return self.url_result(sibnet_urls[0]) + m_opts = re.search(r'(?s)var\s+opts\s*=\s*({.+?});', info_page) if m_opts: m_opts_url = re.search(r"url\s*:\s*'((?!/\b)[^']+)", m_opts.group(1)) diff --git a/youtube_dl/extractor/vlive.py b/youtube_dl/extractor/vlive.py index 6224e6200..42da34d44 100644 --- a/youtube_dl/extractor/vlive.py +++ b/youtube_dl/extractor/vlive.py @@ -106,7 +106,7 @@ class VLiveIE(VLiveBaseIE): raise ExtractorError('Unable to log in', expected=True) def _call_api(self, path_template, video_id, fields=None): - query = {'appId': self._APP_ID, 'gcc': 'KR'} + query = {'appId': self._APP_ID, 'gcc': 'KR', 'platformType': 'PC'} if fields: query['fields'] = fields try: @@ -116,7 +116,7 @@ class VLiveIE(VLiveBaseIE): headers={'Referer': 'https://www.vlive.tv/'}, query=query) except ExtractorError as e: if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403: - self.raise_login_required(json.loads(e.cause.read().decode())['message']) + self.raise_login_required(json.loads(e.cause.read().decode('utf-8'))['message']) raise def _real_extract(self, url): diff --git a/youtube_dl/extractor/voxmedia.py b/youtube_dl/extractor/voxmedia.py index b318e15d4..661208125 100644 --- a/youtube_dl/extractor/voxmedia.py +++ b/youtube_dl/extractor/voxmedia.py @@ -7,6 +7,8 @@ from ..compat import compat_urllib_parse_unquote from ..utils import ( ExtractorError, int_or_none, + try_get, + unified_timestamp, ) @@ -19,14 +21,17 @@ class VoxMediaVolumeIE(OnceIE): setup = self._parse_json(self._search_regex( r'setup\s*=\s*({.+});', webpage, 'setup'), video_id) - video_data = setup.get('video') or {} + player_setup = setup.get('player_setup') or setup + video_data = player_setup.get('video') or {} + formatted_metadata = video_data.get('formatted_metadata') or {} info = { 'id': video_id, - 'title': video_data.get('title_short'), + 'title': player_setup.get('title') or video_data.get('title_short'), 'description': video_data.get('description_long') or video_data.get('description_short'), - 'thumbnail': video_data.get('brightcove_thumbnail') + 'thumbnail': formatted_metadata.get('thumbnail') or video_data.get('brightcove_thumbnail'), + 'timestamp': unified_timestamp(formatted_metadata.get('video_publish_date')), } - asset = setup.get('asset') or setup.get('params') or {} + asset = try_get(setup, lambda x: x['embed_assets']['chorus'], dict) or {} formats = [] hls_url = asset.get('hls_url') @@ -47,6 +52,7 @@ class VoxMediaVolumeIE(OnceIE): if formats: self._sort_formats(formats) info['formats'] = formats + info['duration'] = int_or_none(asset.get('duration')) return info for provider_video_type in ('ooyala', 'youtube', 'brightcove'): @@ -84,7 +90,7 @@ class VoxMediaIE(InfoExtractor): }, { # Volume embed, Youtube 'url': 'http://www.theverge.com/2014/10/21/7025853/google-nexus-6-hands-on-photos-video-android-phablet', - 'md5': '4c8f4a0937752b437c3ebc0ed24802b5', + 'md5': 'fd19aa0cf3a0eea515d4fd5c8c0e9d68', 'info_dict': { 'id': 'Gy8Md3Eky38', 'ext': 'mp4', @@ -93,6 +99,7 @@ class VoxMediaIE(InfoExtractor): 'uploader_id': 'TheVerge', 'upload_date': '20141021', 'uploader': 'The Verge', + 'timestamp': 1413907200, }, 'add_ie': ['Youtube'], 'skip': 'similar to the previous test', @@ -100,13 +107,13 @@ class VoxMediaIE(InfoExtractor): # Volume embed, Youtube 'url': 'http://www.vox.com/2016/3/31/11336640/mississippi-lgbt-religious-freedom-bill', 'info_dict': { - 'id': 'YCjDnX-Xzhg', + 'id': '22986359b', 'ext': 'mp4', 'title': "Mississippi's laws are so bad that its anti-LGBTQ law isn't needed to allow discrimination", 'description': 'md5:fc1317922057de31cd74bce91eb1c66c', - 'uploader_id': 'voxdotcom', 'upload_date': '20150915', - 'uploader': 'Vox', + 'timestamp': 1442332800, + 'duration': 285, }, 'add_ie': ['Youtube'], 'skip': 'similar to the previous test', @@ -160,6 +167,9 @@ class VoxMediaIE(InfoExtractor): 'ext': 'mp4', 'title': 'Post-Post-PC CEO: The Full Code Conference Video of Microsoft\'s Satya Nadella', 'description': 'The longtime veteran was chosen earlier this year as the software giant\'s third leader in its history.', + 'timestamp': 1402938000, + 'upload_date': '20140616', + 'duration': 4114, }, 'add_ie': ['VoxMediaVolume'], }] diff --git a/youtube_dl/extractor/vtm.py b/youtube_dl/extractor/vtm.py new file mode 100644 index 000000000..093f1aa69 --- /dev/null +++ b/youtube_dl/extractor/vtm.py @@ -0,0 +1,62 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import ( + int_or_none, + parse_iso8601, + try_get, +) + + +class VTMIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?vtm\.be/([^/?&#]+)~v(?P<id>[0-9a-f]{8}(?:-[0-9a-f]{4}){3}-[0-9a-f]{12})' + _TEST = { + 'url': 'https://vtm.be/gast-vernielt-genkse-hotelkamer~ve7534523-279f-4b4d-a5c9-a33ffdbe23e1', + 'md5': '37dca85fbc3a33f2de28ceb834b071f8', + 'info_dict': { + 'id': '192445', + 'ext': 'mp4', + 'title': 'Gast vernielt Genkse hotelkamer', + 'timestamp': 1611060180, + 'upload_date': '20210119', + 'duration': 74, + # TODO: fix url _type result processing + # 'series': 'Op Interventie', + } + } + + def _real_extract(self, url): + uuid = self._match_id(url) + video = self._download_json( + 'https://omc4vm23offuhaxx6hekxtzspi.appsync-api.eu-west-1.amazonaws.com/graphql', + uuid, query={ + 'query': '''{ + getComponent(type: Video, uuid: "%s") { + ... on Video { + description + duration + myChannelsVideo + program { + title + } + publishedAt + title + } + } +}''' % uuid, + }, headers={ + 'x-api-key': 'da2-lz2cab4tfnah3mve6wiye4n77e', + })['data']['getComponent'] + + return { + '_type': 'url', + 'id': uuid, + 'title': video.get('title'), + 'url': 'http://mychannels.video/embed/%d' % video['myChannelsVideo'], + 'description': video.get('description'), + 'timestamp': parse_iso8601(video.get('publishedAt')), + 'duration': int_or_none(video.get('duration')), + 'series': try_get(video, lambda x: x['program']['title']), + 'ie_key': 'Medialaan', + } diff --git a/youtube_dl/extractor/vvvvid.py b/youtube_dl/extractor/vvvvid.py index f4cae7fe9..bc196f8a0 100644 --- a/youtube_dl/extractor/vvvvid.py +++ b/youtube_dl/extractor/vvvvid.py @@ -4,6 +4,7 @@ from __future__ import unicode_literals import re from .common import InfoExtractor +from .youtube import YoutubeIE from ..utils import ( ExtractorError, int_or_none, @@ -48,6 +49,22 @@ class VVVVIDIE(InfoExtractor): 'skip_download': True, }, }, { + # video_type == 'video/youtube' + 'url': 'https://www.vvvvid.it/show/404/one-punch-man/406/486683/trailer', + 'md5': '33e0edfba720ad73a8782157fdebc648', + 'info_dict': { + 'id': 'RzmFKUDOUgw', + 'ext': 'mp4', + 'title': 'Trailer', + 'upload_date': '20150906', + 'description': 'md5:a5e802558d35247fee285875328c0b80', + 'uploader_id': 'BandaiVisual', + 'uploader': 'BANDAI NAMCO Arts Channel', + }, + 'params': { + 'skip_download': True, + }, + }, { 'url': 'https://www.vvvvid.it/show/434/perche-dovrei-guardarlo-di-dario-moccia/437/489048', 'only_matching': True }] @@ -58,12 +75,15 @@ class VVVVIDIE(InfoExtractor): 'https://www.vvvvid.it/user/login', None, headers=self.geo_verification_headers())['data']['conn_id'] - def _download_info(self, show_id, path, video_id, fatal=True): + def _download_info(self, show_id, path, video_id, fatal=True, query=None): + q = { + 'conn_id': self._conn_id, + } + if query: + q.update(query) response = self._download_json( 'https://www.vvvvid.it/vvvvid/ondemand/%s/%s' % (show_id, path), - video_id, headers=self.geo_verification_headers(), query={ - 'conn_id': self._conn_id, - }, fatal=fatal) + video_id, headers=self.geo_verification_headers(), query=q, fatal=fatal) if not (response or fatal): return if response.get('result') == 'error': @@ -81,7 +101,8 @@ class VVVVIDIE(InfoExtractor): show_id, season_id, video_id = re.match(self._VALID_URL, url).groups() response = self._download_info( - show_id, 'season/%s' % season_id, video_id) + show_id, 'season/%s' % season_id, + video_id, query={'video_id': video_id}) vid = int(video_id) video_data = list(filter( @@ -154,37 +175,50 @@ class VVVVIDIE(InfoExtractor): if season_number: info['season_number'] = int(season_number) - for quality in ('_sd', ''): + video_type = video_data.get('video_type') + is_youtube = False + for quality in ('', '_sd'): embed_code = video_data.get('embed_info' + quality) if not embed_code: continue embed_code = ds(embed_code) - video_type = video_data.get('video_type') - if video_type in ('video/rcs', 'video/kenc'): - if video_type == 'video/kenc': - kenc = self._download_json( - 'https://www.vvvvid.it/kenc', video_id, query={ - 'action': 'kt', - 'conn_id': self._conn_id, - 'url': embed_code, - }, fatal=False) or {} - kenc_message = kenc.get('message') - if kenc_message: - embed_code += '?' + ds(kenc_message) + if video_type == 'video/kenc': + embed_code = re.sub(r'https?(://[^/]+)/z/', r'https\1/i/', embed_code).replace('/manifest.f4m', '/master.m3u8') + kenc = self._download_json( + 'https://www.vvvvid.it/kenc', video_id, query={ + 'action': 'kt', + 'conn_id': self._conn_id, + 'url': embed_code, + }, fatal=False) or {} + kenc_message = kenc.get('message') + if kenc_message: + embed_code += '?' + ds(kenc_message) + formats.extend(self._extract_m3u8_formats( + embed_code, video_id, 'mp4', m3u8_id='hls', fatal=False)) + elif video_type == 'video/rcs': formats.extend(self._extract_akamai_formats(embed_code, video_id)) + elif video_type == 'video/youtube': + info.update({ + '_type': 'url_transparent', + 'ie_key': YoutubeIE.ie_key(), + 'url': embed_code, + }) + is_youtube = True + break else: formats.extend(self._extract_wowza_formats( 'http://sb.top-ix.org/videomg/_definst_/mp4:%s/playlist.m3u8' % embed_code, video_id)) metadata_from_url(embed_code) - self._sort_formats(formats) + if not is_youtube: + self._sort_formats(formats) + info['formats'] = formats metadata_from_url(video_data.get('thumbnail')) info.update(self._extract_common_video_info(video_data)) info.update({ 'id': video_id, 'title': title, - 'formats': formats, 'duration': int_or_none(video_data.get('length')), 'series': video_data.get('show_title'), 'season_id': season_id, @@ -220,9 +254,13 @@ class VVVVIDShowIE(VVVVIDIE): show_info = self._download_info( show_id, 'info/', show_title, fatal=False) + if not show_title: + base_url += "/title" + entries = [] for season in (seasons or []): episodes = season.get('episodes') or [] + playlist_title = season.get('name') or show_info.get('title') for episode in episodes: if episode.get('playable') is False: continue @@ -232,12 +270,13 @@ class VVVVIDShowIE(VVVVIDIE): continue info = self._extract_common_video_info(episode) info.update({ - '_type': 'url', + '_type': 'url_transparent', 'ie_key': VVVVIDIE.ie_key(), 'url': '/'.join([base_url, season_id, video_id]), 'title': episode.get('title'), 'description': episode.get('description'), 'season_id': season_id, + 'playlist_title': playlist_title, }) entries.append(info) diff --git a/youtube_dl/extractor/wat.py b/youtube_dl/extractor/wat.py index 8ef3e0906..f1bccc2d6 100644 --- a/youtube_dl/extractor/wat.py +++ b/youtube_dl/extractor/wat.py @@ -1,15 +1,13 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..compat import compat_str from ..utils import ( ExtractorError, - unified_strdate, - HEADRequest, int_or_none, + try_get, + unified_strdate, ) @@ -32,6 +30,7 @@ class WatIE(InfoExtractor): 'skip_download': True, }, 'expected_warnings': ['HTTP Error 404'], + 'skip': 'This content is no longer available', }, { 'url': 'http://www.wat.tv/video/gregory-lemarchal-voix-ange-6z1v7_6ygkj_.html', @@ -43,17 +42,10 @@ class WatIE(InfoExtractor): 'upload_date': '20140816', }, 'expected_warnings': ["Ce contenu n'est pas disponible pour l'instant."], + 'skip': 'This content is no longer available', }, ] - - _FORMATS = ( - (200, 416, 234), - (400, 480, 270), - (600, 640, 360), - (1200, 640, 360), - (1800, 960, 540), - (2500, 1280, 720), - ) + _GEO_BYPASS = False def _real_extract(self, url): video_id = self._match_id(url) @@ -61,97 +53,54 @@ class WatIE(InfoExtractor): # 'contentv4' is used in the website, but it also returns the related # videos, we don't need them + # video_data = self._download_json( + # 'http://www.wat.tv/interface/contentv4s/' + video_id, video_id) video_data = self._download_json( - 'http://www.wat.tv/interface/contentv4s/' + video_id, video_id) + 'https://mediainfo.tf1.fr/mediainfocombo/' + video_id, + video_id, query={'context': 'MYTF1'}) video_info = video_data['media'] error_desc = video_info.get('error_desc') if error_desc: - self.report_warning( - '%s returned error: %s' % (self.IE_NAME, error_desc)) - - chapters = video_info['chapters'] - if chapters: - first_chapter = chapters[0] - - def video_id_for_chapter(chapter): - return chapter['tc_start'].split('-')[0] - - if video_id_for_chapter(first_chapter) != video_id: - self.to_screen('Multipart video detected') - entries = [self.url_result('wat:%s' % video_id_for_chapter(chapter)) for chapter in chapters] - return self.playlist_result(entries, video_id, video_info['title']) - # Otherwise we can continue and extract just one part, we have to use - # the video id for getting the video url - else: - first_chapter = video_info - - title = first_chapter['title'] + if video_info.get('error_code') == 'GEOBLOCKED': + self.raise_geo_restricted(error_desc, video_info.get('geoList')) + raise ExtractorError(error_desc, expected=True) - def extract_url(path_template, url_type): - req_url = 'http://www.wat.tv/get/%s' % (path_template % video_id) - head = self._request_webpage(HEADRequest(req_url), video_id, 'Extracting %s url' % url_type, fatal=False) - if head: - red_url = head.geturl() - if req_url != red_url: - return red_url - return None - - def remove_bitrate_limit(manifest_url): - return re.sub(r'(?:max|min)_bitrate=\d+&?', '', manifest_url) + title = video_info['title'] formats = [] - try: - alt_urls = lambda manifest_url: [re.sub(r'(?:wdv|ssm)?\.ism/', repl + '.ism/', manifest_url) for repl in ('', 'ssm')] - manifest_urls = self._download_json( - 'http://www.wat.tv/get/webhtml/' + video_id, video_id) - m3u8_url = manifest_urls.get('hls') - if m3u8_url: - m3u8_url = remove_bitrate_limit(m3u8_url) - for m3u8_alt_url in alt_urls(m3u8_url): + + def extract_formats(manifest_urls): + for f, f_url in manifest_urls.items(): + if not f_url: + continue + if f in ('dash', 'mpd'): + formats.extend(self._extract_mpd_formats( + f_url.replace('://das-q1.tf1.fr/', '://das-q1-ssl.tf1.fr/'), + video_id, mpd_id='dash', fatal=False)) + elif f == 'hls': formats.extend(self._extract_m3u8_formats( - m3u8_alt_url, video_id, 'mp4', + f_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls', fatal=False)) - formats.extend(self._extract_f4m_formats( - m3u8_alt_url.replace('ios', 'web').replace('.m3u8', '.f4m'), - video_id, f4m_id='hds', fatal=False)) - mpd_url = manifest_urls.get('mpd') - if mpd_url: - mpd_url = remove_bitrate_limit(mpd_url) - for mpd_alt_url in alt_urls(mpd_url): - formats.extend(self._extract_mpd_formats( - mpd_alt_url, video_id, mpd_id='dash', fatal=False)) - self._sort_formats(formats) - except ExtractorError: - abr = 64 - for vbr, width, height in self._FORMATS: - tbr = vbr + abr - format_id = 'http-%s' % tbr - fmt_url = 'http://dnl.adv.tf1.fr/2/USP-0x0/%s/%s/%s/ssm/%s-%s-64k.mp4' % (video_id[-4:-2], video_id[-2:], video_id, video_id, vbr) - if self._is_valid_url(fmt_url, video_id, format_id): - formats.append({ - 'format_id': format_id, - 'url': fmt_url, - 'vbr': vbr, - 'abr': abr, - 'width': width, - 'height': height, - }) - date_diffusion = first_chapter.get('date_diffusion') or video_data.get('configv4', {}).get('estatS4') - upload_date = unified_strdate(date_diffusion) if date_diffusion else None - duration = None - files = video_info['files'] - if files: - duration = int_or_none(files[0].get('duration')) + delivery = video_data.get('delivery') or {} + extract_formats({delivery.get('format'): delivery.get('url')}) + if not formats: + if delivery.get('drm'): + raise ExtractorError('This video is DRM protected.', expected=True) + manifest_urls = self._download_json( + 'http://www.wat.tv/get/webhtml/' + video_id, video_id, fatal=False) + if manifest_urls: + extract_formats(manifest_urls) + + self._sort_formats(formats) return { 'id': video_id, 'title': title, - 'thumbnail': first_chapter.get('preview'), - 'description': first_chapter.get('description'), - 'view_count': int_or_none(video_info.get('views')), - 'upload_date': upload_date, - 'duration': duration, + 'thumbnail': video_info.get('preview'), + 'upload_date': unified_strdate(try_get( + video_data, lambda x: x['mediametrie']['chapters'][0]['estatS4'])), + 'duration': int_or_none(video_info.get('duration')), 'formats': formats, } diff --git a/youtube_dl/extractor/xboxclips.py b/youtube_dl/extractor/xboxclips.py index d9c277bc3..25f487e1e 100644 --- a/youtube_dl/extractor/xboxclips.py +++ b/youtube_dl/extractor/xboxclips.py @@ -1,40 +1,55 @@ # coding: utf-8 from __future__ import unicode_literals +import re + from .common import InfoExtractor +from ..compat import ( + compat_parse_qs, + compat_urllib_parse_urlparse, +) from ..utils import ( int_or_none, + month_by_abbreviation, parse_filesize, - unified_strdate, ) class XboxClipsIE(InfoExtractor): - _VALID_URL = r'https?://(?:www\.)?xboxclips\.com/(?:video\.php\?.*vid=|[^/]+/)(?P<id>[\w-]{36})' - _TEST = { + _VALID_URL = r'https?://(?:www\.)?(?:xboxclips\.com|gameclips\.io)/(?:video\.php\?.*vid=|[^/]+/)(?P<id>[\da-f]{8}-(?:[\da-f]{4}-){3}[\da-f]{12})' + _TESTS = [{ 'url': 'http://xboxclips.com/video.php?uid=2533274823424419&gamertag=Iabdulelah&vid=074a69a9-5faf-46aa-b93b-9909c1720325', 'md5': 'fbe1ec805e920aeb8eced3c3e657df5d', 'info_dict': { 'id': '074a69a9-5faf-46aa-b93b-9909c1720325', 'ext': 'mp4', - 'title': 'Iabdulelah playing Titanfall', + 'title': 'iAbdulElah playing Titanfall', 'filesize_approx': 26800000, 'upload_date': '20140807', 'duration': 56, } - } + }, { + 'url': 'https://gameclips.io/iAbdulElah/074a69a9-5faf-46aa-b93b-9909c1720325', + 'only_matching': True, + }] def _real_extract(self, url): video_id = self._match_id(url) + if '/video.php' in url: + qs = compat_parse_qs(compat_urllib_parse_urlparse(url).query) + url = 'https://gameclips.io/%s/%s' % (qs['gamertag'][0], qs['vid'][0]) + webpage = self._download_webpage(url, video_id) + info = self._parse_html5_media_entries(url, webpage, video_id)[0] - video_url = self._html_search_regex( - r'>(?:Link|Download): <a[^>]+href="([^"]+)"', webpage, 'video URL') - title = self._html_search_regex( - r'<title>XboxClips \| ([^<]+)</title>', webpage, 'title') - upload_date = unified_strdate(self._html_search_regex( - r'>Recorded: ([^<]+)<', webpage, 'upload date', fatal=False)) + title = self._html_search_meta(['og:title', 'twitter:title'], webpage) + upload_date = None + mobj = re.search( + r'>Recorded: (\d{2})-(Jan|Feb|Mar|Apr|May|Ju[nl]|Aug|Sep|Oct|Nov|Dec)-(\d{4})', + webpage) + if mobj: + upload_date = '%s%.2d%s' % (mobj.group(3), month_by_abbreviation(mobj.group(2)), mobj.group(1)) filesize = parse_filesize(self._html_search_regex( r'>Size: ([^<]+)<', webpage, 'file size', fatal=False)) duration = int_or_none(self._html_search_regex( @@ -42,12 +57,12 @@ class XboxClipsIE(InfoExtractor): view_count = int_or_none(self._html_search_regex( r'>Views: (\d+)<', webpage, 'view count', fatal=False)) - return { + info.update({ 'id': video_id, - 'url': video_url, 'title': title, 'upload_date': upload_date, 'filesize_approx': filesize, 'duration': duration, 'view_count': view_count, - } + }) + return info diff --git a/youtube_dl/extractor/xfileshare.py b/youtube_dl/extractor/xfileshare.py index cbd5d1cbb..df9efa9fa 100644 --- a/youtube_dl/extractor/xfileshare.py +++ b/youtube_dl/extractor/xfileshare.py @@ -58,6 +58,7 @@ class XFileShareIE(InfoExtractor): (r'vidlocker\.xyz', 'VidLocker'), (r'vidshare\.tv', 'VidShare'), (r'vup\.to', 'VUp'), + (r'wolfstream\.tv', 'WolfStream'), (r'xvideosharing\.com', 'XVideoSharing'), ) @@ -82,6 +83,9 @@ class XFileShareIE(InfoExtractor): }, { 'url': 'https://aparat.cam/n4d6dh0wvlpr', 'only_matching': True, + }, { + 'url': 'https://wolfstream.tv/nthme29v9u2x', + 'only_matching': True, }] @staticmethod diff --git a/youtube_dl/extractor/xhamster.py b/youtube_dl/extractor/xhamster.py index 76aeaf9a4..f73b9778f 100644 --- a/youtube_dl/extractor/xhamster.py +++ b/youtube_dl/extractor/xhamster.py @@ -11,11 +11,14 @@ from ..utils import ( dict_get, extract_attributes, ExtractorError, + float_or_none, int_or_none, parse_duration, + str_or_none, try_get, unified_strdate, url_or_none, + urljoin, ) @@ -146,36 +149,89 @@ class XHamsterIE(InfoExtractor): video = initials['videoModel'] title = video['title'] formats = [] - for format_id, formats_dict in video['sources'].items(): + format_urls = set() + format_sizes = {} + sources = try_get(video, lambda x: x['sources'], dict) or {} + for format_id, formats_dict in sources.items(): if not isinstance(formats_dict, dict): continue + download_sources = try_get(sources, lambda x: x['download'], dict) or {} + for quality, format_dict in download_sources.items(): + if not isinstance(format_dict, dict): + continue + format_sizes[quality] = float_or_none(format_dict.get('size')) for quality, format_item in formats_dict.items(): if format_id == 'download': # Download link takes some time to be generated, # skipping for now continue - if not isinstance(format_item, dict): - continue - format_url = format_item.get('link') - filesize = int_or_none( - format_item.get('size'), invscale=1000000) - else: - format_url = format_item - filesize = None + format_url = format_item format_url = url_or_none(format_url) - if not format_url: + if not format_url or format_url in format_urls: continue + format_urls.add(format_url) formats.append({ 'format_id': '%s-%s' % (format_id, quality), 'url': format_url, 'ext': determine_ext(format_url, 'mp4'), 'height': get_height(quality), - 'filesize': filesize, + 'filesize': format_sizes.get(quality), 'http_headers': { 'Referer': urlh.geturl(), }, }) - self._sort_formats(formats) + xplayer_sources = try_get( + initials, lambda x: x['xplayerSettings']['sources'], dict) + if xplayer_sources: + hls_sources = xplayer_sources.get('hls') + if isinstance(hls_sources, dict): + for hls_format_key in ('url', 'fallback'): + hls_url = hls_sources.get(hls_format_key) + if not hls_url: + continue + hls_url = urljoin(url, hls_url) + if not hls_url or hls_url in format_urls: + continue + format_urls.add(hls_url) + formats.extend(self._extract_m3u8_formats( + hls_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + standard_sources = xplayer_sources.get('standard') + if isinstance(standard_sources, dict): + for format_id, formats_list in standard_sources.items(): + if not isinstance(formats_list, list): + continue + for standard_format in formats_list: + if not isinstance(standard_format, dict): + continue + for standard_format_key in ('url', 'fallback'): + standard_url = standard_format.get(standard_format_key) + if not standard_url: + continue + standard_url = urljoin(url, standard_url) + if not standard_url or standard_url in format_urls: + continue + format_urls.add(standard_url) + ext = determine_ext(standard_url, 'mp4') + if ext == 'm3u8': + formats.extend(self._extract_m3u8_formats( + standard_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + continue + quality = (str_or_none(standard_format.get('quality')) + or str_or_none(standard_format.get('label')) + or '') + formats.append({ + 'format_id': '%s-%s' % (format_id, quality), + 'url': standard_url, + 'ext': ext, + 'height': get_height(quality), + 'filesize': format_sizes.get(quality), + 'http_headers': { + 'Referer': standard_url, + }, + }) + self._sort_formats(formats, field_preference=('height', 'width', 'tbr', 'format_id')) categories_list = video.get('categories') if isinstance(categories_list, list): diff --git a/youtube_dl/extractor/xtube.py b/youtube_dl/extractor/xtube.py index 18969058f..7246409e3 100644 --- a/youtube_dl/extractor/xtube.py +++ b/youtube_dl/extractor/xtube.py @@ -11,6 +11,7 @@ from ..utils import ( parse_duration, sanitized_Request, str_to_int, + url_or_none, ) @@ -87,10 +88,10 @@ class XTubeIE(InfoExtractor): 'Cookie': 'age_verified=1; cookiesAccepted=1', }) - title, thumbnail, duration = [None] * 3 + title, thumbnail, duration, sources, media_definition = [None] * 5 config = self._parse_json(self._search_regex( - r'playerConf\s*=\s*({.+?})\s*,\s*(?:\n|loaderConf)', webpage, 'config', + r'playerConf\s*=\s*({.+?})\s*,\s*(?:\n|loaderConf|playerWrapper)', webpage, 'config', default='{}'), video_id, transform_source=js_to_json, fatal=False) if config: config = config.get('mainRoll') @@ -99,20 +100,52 @@ class XTubeIE(InfoExtractor): thumbnail = config.get('poster') duration = int_or_none(config.get('duration')) sources = config.get('sources') or config.get('format') + media_definition = config.get('mediaDefinition') - if not isinstance(sources, dict): + if not isinstance(sources, dict) and not media_definition: sources = self._parse_json(self._search_regex( r'(["\'])?sources\1?\s*:\s*(?P<sources>{.+?}),', webpage, 'sources', group='sources'), video_id, transform_source=js_to_json) formats = [] - for format_id, format_url in sources.items(): - formats.append({ - 'url': format_url, - 'format_id': format_id, - 'height': int_or_none(format_id), - }) + format_urls = set() + + if isinstance(sources, dict): + for format_id, format_url in sources.items(): + format_url = url_or_none(format_url) + if not format_url: + continue + if format_url in format_urls: + continue + format_urls.add(format_url) + formats.append({ + 'url': format_url, + 'format_id': format_id, + 'height': int_or_none(format_id), + }) + + if isinstance(media_definition, list): + for media in media_definition: + video_url = url_or_none(media.get('videoUrl')) + if not video_url: + continue + if video_url in format_urls: + continue + format_urls.add(video_url) + format_id = media.get('format') + if format_id == 'hls': + formats.extend(self._extract_m3u8_formats( + video_url, video_id, 'mp4', entry_protocol='m3u8_native', + m3u8_id='hls', fatal=False)) + elif format_id == 'mp4': + height = int_or_none(media.get('quality')) + formats.append({ + 'url': video_url, + 'format_id': '%s-%d' % (format_id, height) if height else format_id, + 'height': height, + }) + self._remove_duplicate_formats(formats) self._sort_formats(formats) diff --git a/youtube_dl/extractor/yahoo.py b/youtube_dl/extractor/yahoo.py index e4615376c..a17b10d6e 100644 --- a/youtube_dl/extractor/yahoo.py +++ b/youtube_dl/extractor/yahoo.py @@ -177,46 +177,9 @@ class YahooIE(InfoExtractor): 'only_matching': True, }] - def _real_extract(self, url): - url, country, display_id = re.match(self._VALID_URL, url).groups() - if not country: - country = 'us' - else: - country = country.split('-')[0] - api_base = 'https://%s.yahoo.com/_td/api/resource/' % country - - for i, uuid in enumerate(['url=' + url, 'ymedia-alias=' + display_id]): - content = self._download_json( - api_base + 'content;getDetailView=true;uuids=["%s"]' % uuid, - display_id, 'Downloading content JSON metadata', fatal=i == 1) - if content: - item = content['items'][0] - break - - if item.get('type') != 'video': - entries = [] - - cover = item.get('cover') or {} - if cover.get('type') == 'yvideo': - cover_url = cover.get('url') - if cover_url: - entries.append(self.url_result( - cover_url, 'Yahoo', cover.get('uuid'))) - - for e in item.get('body', []): - if e.get('type') == 'videoIframe': - iframe_url = e.get('url') - if not iframe_url: - continue - entries.append(self.url_result(iframe_url)) - - return self.playlist_result( - entries, item.get('uuid'), - item.get('title'), item.get('summary')) - - video_id = item['uuid'] + def _extract_yahoo_video(self, video_id, country): video = self._download_json( - api_base + 'VideoService.videos;view=full;video_ids=["%s"]' % video_id, + 'https://%s.yahoo.com/_td/api/resource/VideoService.videos;view=full;video_ids=["%s"]' % (country, video_id), video_id, 'Downloading video JSON metadata')[0] title = video['title'] @@ -298,7 +261,6 @@ class YahooIE(InfoExtractor): 'id': video_id, 'title': self._live_title(title) if is_live else title, 'formats': formats, - 'display_id': display_id, 'thumbnails': thumbnails, 'description': clean_html(video.get('description')), 'timestamp': parse_iso8601(video.get('publish_time')), @@ -311,6 +273,44 @@ class YahooIE(InfoExtractor): 'episode_number': int_or_none(series_info.get('episode_number')), } + def _real_extract(self, url): + url, country, display_id = re.match(self._VALID_URL, url).groups() + if not country: + country = 'us' + else: + country = country.split('-')[0] + + item = self._download_json( + 'https://%s.yahoo.com/caas/content/article' % country, display_id, + 'Downloading content JSON metadata', query={ + 'url': url + })['items'][0]['data']['partnerData'] + + if item.get('type') != 'video': + entries = [] + + cover = item.get('cover') or {} + if cover.get('type') == 'yvideo': + cover_url = cover.get('url') + if cover_url: + entries.append(self.url_result( + cover_url, 'Yahoo', cover.get('uuid'))) + + for e in (item.get('body') or []): + if e.get('type') == 'videoIframe': + iframe_url = e.get('url') + if not iframe_url: + continue + entries.append(self.url_result(iframe_url)) + + return self.playlist_result( + entries, item.get('uuid'), + item.get('title'), item.get('summary')) + + info = self._extract_yahoo_video(item['uuid'], country) + info['display_id'] = display_id + return info + class YahooSearchIE(SearchInfoExtractor): IE_DESC = 'Yahoo screen search' diff --git a/youtube_dl/extractor/yandexmusic.py b/youtube_dl/extractor/yandexmusic.py index 7893f363e..84969f8e1 100644 --- a/youtube_dl/extractor/yandexmusic.py +++ b/youtube_dl/extractor/yandexmusic.py @@ -1,8 +1,9 @@ # coding: utf-8 from __future__ import unicode_literals -import re import hashlib +import itertools +import re from .common import InfoExtractor from ..compat import compat_str @@ -209,17 +210,27 @@ class YandexMusicPlaylistBaseIE(YandexMusicBaseIE): missing_track_ids = [ track_id for track_id in track_ids if track_id not in present_track_ids] - missing_tracks = self._call_api( - 'track-entries', tld, url, item_id, - 'Downloading missing tracks JSON', { - 'entries': ','.join(missing_track_ids), - 'lang': tld, - 'external-domain': 'music.yandex.%s' % tld, - 'overembed': 'false', - 'strict': 'true', - }) - if missing_tracks: - tracks.extend(missing_tracks) + # Request missing tracks in chunks to avoid exceeding max HTTP header size, + # see https://github.com/ytdl-org/youtube-dl/issues/27355 + _TRACKS_PER_CHUNK = 250 + for chunk_num in itertools.count(0): + start = chunk_num * _TRACKS_PER_CHUNK + end = start + _TRACKS_PER_CHUNK + missing_track_ids_req = missing_track_ids[start:end] + assert missing_track_ids_req + missing_tracks = self._call_api( + 'track-entries', tld, url, item_id, + 'Downloading missing tracks JSON chunk %d' % (chunk_num + 1), { + 'entries': ','.join(missing_track_ids_req), + 'lang': tld, + 'external-domain': 'music.yandex.%s' % tld, + 'overembed': 'false', + 'strict': 'true', + }) + if missing_tracks: + tracks.extend(missing_tracks) + if end >= len(missing_track_ids): + break return tracks diff --git a/youtube_dl/extractor/youku.py b/youtube_dl/extractor/youku.py index 61d1ab209..880c89687 100644 --- a/youtube_dl/extractor/youku.py +++ b/youtube_dl/extractor/youku.py @@ -154,7 +154,7 @@ class YoukuIE(InfoExtractor): # request basic data basic_data_params = { 'vid': video_id, - 'ccode': '0590', + 'ccode': '0532', 'client_ip': '192.168.1.1', 'utid': cna, 'client_ts': time.time() / 1000, diff --git a/youtube_dl/extractor/youporn.py b/youtube_dl/extractor/youporn.py index 7b9feafeb..7084d3d12 100644 --- a/youtube_dl/extractor/youporn.py +++ b/youtube_dl/extractor/youporn.py @@ -4,13 +4,12 @@ import re from .common import InfoExtractor from ..utils import ( + extract_attributes, int_or_none, str_to_int, - unescapeHTML, unified_strdate, url_or_none, ) -from ..aes import aes_decrypt_text class YouPornIE(InfoExtractor): @@ -25,6 +24,7 @@ class YouPornIE(InfoExtractor): 'title': 'Sex Ed: Is It Safe To Masturbate Daily?', 'description': 'Love & Sex Answers: http://bit.ly/DanAndJenn -- Is It Unhealthy To Masturbate Daily?', 'thumbnail': r're:^https?://.*\.jpg$', + 'duration': 210, 'uploader': 'Ask Dan And Jennifer', 'upload_date': '20101217', 'average_rating': int, @@ -33,6 +33,7 @@ class YouPornIE(InfoExtractor): 'tags': list, 'age_limit': 18, }, + 'skip': 'This video has been disabled', }, { # Unknown uploader 'url': 'http://www.youporn.com/watch/561726/big-tits-awesome-brunette-on-amazing-webcam-show/?from=related3&al=2&from_id=561726&pos=4', @@ -54,12 +55,16 @@ class YouPornIE(InfoExtractor): 'params': { 'skip_download': True, }, + 'skip': '404', }, { 'url': 'https://www.youporn.com/embed/505835/sex-ed-is-it-safe-to-masturbate-daily/', 'only_matching': True, }, { 'url': 'http://www.youporn.com/watch/505835', 'only_matching': True, + }, { + 'url': 'https://www.youporn.com/watch/13922959/femdom-principal/', + 'only_matching': True, }] @staticmethod @@ -73,74 +78,50 @@ class YouPornIE(InfoExtractor): video_id = mobj.group('id') display_id = mobj.group('display_id') or video_id - webpage = self._download_webpage( - 'http://www.youporn.com/watch/%s' % video_id, display_id, - headers={'Cookie': 'age_verified=1'}) - - title = self._html_search_regex( - r'(?s)<div[^>]+class=["\']watchVideoTitle[^>]+>(.+?)</div>', - webpage, 'title', default=None) or self._og_search_title( - webpage, default=None) or self._html_search_meta( - 'title', webpage, fatal=True) - - links = [] - - # Main source - definitions = self._parse_json( - self._search_regex( - r'mediaDefinition\s*=\s*(\[.+?\]);', webpage, - 'media definitions', default='[]'), - video_id, fatal=False) - if definitions: - for definition in definitions: - if not isinstance(definition, dict): - continue - video_url = url_or_none(definition.get('videoUrl')) - if video_url: - links.append(video_url) - - # Fallback #1, this also contains extra low quality 180p format - for _, link in re.findall(r'<a[^>]+href=(["\'])(http.+?)\1[^>]+title=["\']Download [Vv]ideo', webpage): - links.append(link) - - # Fallback #2 (unavailable as at 22.06.2017) - sources = self._search_regex( - r'(?s)sources\s*:\s*({.+?})', webpage, 'sources', default=None) - if sources: - for _, link in re.findall(r'[^:]+\s*:\s*(["\'])(http.+?)\1', sources): - links.append(link) - - # Fallback #3 (unavailable as at 22.06.2017) - for _, link in re.findall( - r'(?:videoSrc|videoIpadUrl|html5PlayerSrc)\s*[:=]\s*(["\'])(http.+?)\1', webpage): - links.append(link) - - # Fallback #4, encrypted links (unavailable as at 22.06.2017) - for _, encrypted_link in re.findall( - r'encryptedQuality\d{3,4}URL\s*=\s*(["\'])([\da-zA-Z+/=]+)\1', webpage): - links.append(aes_decrypt_text(encrypted_link, title, 32).decode('utf-8')) + definitions = self._download_json( + 'https://www.youporn.com/api/video/media_definitions/%s/' % video_id, + display_id) formats = [] - for video_url in set(unescapeHTML(link) for link in links): + for definition in definitions: + if not isinstance(definition, dict): + continue + video_url = url_or_none(definition.get('videoUrl')) + if not video_url: + continue f = { 'url': video_url, + 'filesize': int_or_none(definition.get('videoSize')), } + height = int_or_none(definition.get('quality')) # Video URL's path looks like this: # /201012/17/505835/720p_1500k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4 # /201012/17/505835/vl_240p_240k_505835/YouPorn%20-%20Sex%20Ed%20Is%20It%20Safe%20To%20Masturbate%20Daily.mp4 + # /videos/201703/11/109285532/1080P_4000K_109285532.mp4 # We will benefit from it by extracting some metadata - mobj = re.search(r'(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+/', video_url) + mobj = re.search(r'(?P<height>\d{3,4})[pP]_(?P<bitrate>\d+)[kK]_\d+', video_url) if mobj: - height = int(mobj.group('height')) + if not height: + height = int(mobj.group('height')) bitrate = int(mobj.group('bitrate')) f.update({ 'format_id': '%dp-%dk' % (height, bitrate), - 'height': height, 'tbr': bitrate, }) + f['height'] = height formats.append(f) self._sort_formats(formats) + webpage = self._download_webpage( + 'http://www.youporn.com/watch/%s' % video_id, display_id, + headers={'Cookie': 'age_verified=1'}) + + title = self._html_search_regex( + r'(?s)<div[^>]+class=["\']watchVideoTitle[^>]+>(.+?)</div>', + webpage, 'title', default=None) or self._og_search_title( + webpage, default=None) or self._html_search_meta( + 'title', webpage, fatal=True) + description = self._html_search_regex( r'(?s)<div[^>]+\bid=["\']description["\'][^>]*>(.+?)</div>', webpage, 'description', @@ -149,6 +130,8 @@ class YouPornIE(InfoExtractor): thumbnail = self._search_regex( r'(?:imageurl\s*=|poster\s*:)\s*(["\'])(?P<thumbnail>.+?)\1', webpage, 'thumbnail', fatal=False, group='thumbnail') + duration = int_or_none(self._html_search_meta( + 'video:duration', webpage, 'duration', fatal=False)) uploader = self._html_search_regex( r'(?s)<div[^>]+class=["\']submitByLink["\'][^>]*>(.+?)</div>', @@ -161,13 +144,12 @@ class YouPornIE(InfoExtractor): age_limit = self._rta_search(webpage) - average_rating = int_or_none(self._search_regex( - r'<div[^>]+class=["\']videoRatingPercentage["\'][^>]*>(\d+)%</div>', - webpage, 'average rating', fatal=False)) - - view_count = str_to_int(self._search_regex( - r'(?s)<div[^>]+class=(["\']).*?\bvideoInfoViews\b.*?\1[^>]*>.*?(?P<count>[\d,.]+)<', - webpage, 'view count', fatal=False, group='count')) + view_count = None + views = self._search_regex( + r'(<div[^>]+\bclass=["\']js_videoInfoViews["\']>)', webpage, + 'views', default=None) + if views: + view_count = str_to_int(extract_attributes(views).get('data-value')) comment_count = str_to_int(self._search_regex( r'>All [Cc]omments? \(([\d,.]+)\)', webpage, 'comment count', default=None)) @@ -190,9 +172,9 @@ class YouPornIE(InfoExtractor): 'title': title, 'description': description, 'thumbnail': thumbnail, + 'duration': duration, 'uploader': uploader, 'upload_date': upload_date, - 'average_rating': average_rating, 'view_count': view_count, 'comment_count': comment_count, 'categories': categories, diff --git a/youtube_dl/extractor/youtube.py b/youtube_dl/extractor/youtube.py index bbd3e80d8..7cd651c67 100644 --- a/youtube_dl/extractor/youtube.py +++ b/youtube_dl/extractor/youtube.py @@ -2,41 +2,35 @@ from __future__ import unicode_literals - import itertools import json import os.path import random import re -import time import traceback from .common import InfoExtractor, SearchInfoExtractor -from ..jsinterp import JSInterpreter -from ..swfinterp import SWFInterpreter from ..compat import ( compat_chr, compat_HTTPError, compat_parse_qs, - compat_urllib_parse_unquote, + compat_str, compat_urllib_parse_unquote_plus, compat_urllib_parse_urlencode, compat_urllib_parse_urlparse, compat_urlparse, - compat_str, ) +from ..jsinterp import JSInterpreter from ..utils import ( - bool_or_none, - clean_html, - error_to_compat_str, ExtractorError, + clean_html, + dict_get, float_or_none, - get_element_by_id, int_or_none, mimetype2ext, parse_codecs, parse_duration, - remove_quotes, + qualities, remove_start, smuggle_url, str_or_none, @@ -46,13 +40,16 @@ from ..utils import ( unified_strdate, unsmuggle_url, update_url_query, - uppercase_escape, url_or_none, urlencode_postdata, urljoin, ) +def parse_qs(url): + return compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + + class YoutubeBaseInfoExtractor(InfoExtractor): """Provide base functions for Youtube extractors""" _LOGIN_URL = 'https://accounts.google.com/ServiceLogin' @@ -68,17 +65,6 @@ class YoutubeBaseInfoExtractor(InfoExtractor): _PLAYLIST_ID_RE = r'(?:(?:PL|LL|EC|UU|FL|RD|UL|TL|PU|OLAK5uy_)[0-9A-Za-z-_]{10,}|RDMM)' - def _set_language(self): - self._set_cookie( - '.youtube.com', 'PREF', 'f1=50000000&f6=8&hl=en', - # YouTube sets the expire time to about two months - expire_time=time.time() + 2 * 30 * 24 * 3600) - - def _ids_to_results(self, ids): - return [ - self.url_result(vid_id, 'Youtube', video_id=vid_id) - for vid_id in ids] - def _login(self): """ Attempt to log in to YouTube. @@ -262,10 +248,25 @@ class YoutubeBaseInfoExtractor(InfoExtractor): return True + def _initialize_consent(self): + cookies = self._get_cookies('https://www.youtube.com/') + if cookies.get('__Secure-3PSID'): + return + consent_id = None + consent = cookies.get('CONSENT') + if consent: + if 'YES' in consent.value: + return + consent_id = self._search_regex( + r'PENDING\+(\d+)', consent.value, 'consent', default=None) + if not consent_id: + consent_id = random.randint(100, 999) + self._set_cookie('.youtube.com', 'CONSENT', 'YES+cb.20210328-17-p0.en+FX+%s' % consent_id) + def _real_initialize(self): + self._initialize_consent() if self._downloader is None: return - self._set_language() if not self._login(): return @@ -282,19 +283,17 @@ class YoutubeBaseInfoExtractor(InfoExtractor): _YT_INITIAL_PLAYER_RESPONSE_RE = r'ytInitialPlayerResponse\s*=\s*({.+?})\s*;' _YT_INITIAL_BOUNDARY_RE = r'(?:var\s+meta|</script|\n)' - def _call_api(self, ep, query, video_id): + def _call_api(self, ep, query, video_id, fatal=True): data = self._DEFAULT_API_DATA.copy() data.update(query) - response = self._download_json( + return self._download_json( 'https://www.youtube.com/youtubei/v1/%s' % ep, video_id=video_id, note='Downloading API JSON', errnote='Unable to download API page', - data=json.dumps(data).encode('utf8'), + data=json.dumps(data).encode('utf8'), fatal=fatal, headers={'content-type': 'application/json'}, query={'key': 'AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8'}) - return response - def _extract_yt_initial_data(self, video_id, webpage): return self._parse_json( self._search_regex( @@ -306,7 +305,39 @@ class YoutubeBaseInfoExtractor(InfoExtractor): return self._parse_json( self._search_regex( r'ytcfg\.set\s*\(\s*({.+?})\s*\)\s*;', webpage, 'ytcfg', - default='{}'), video_id, fatal=False) + default='{}'), video_id, fatal=False) or {} + + def _extract_video(self, renderer): + video_id = renderer['videoId'] + title = try_get( + renderer, + (lambda x: x['title']['runs'][0]['text'], + lambda x: x['title']['simpleText']), compat_str) + description = try_get( + renderer, lambda x: x['descriptionSnippet']['runs'][0]['text'], + compat_str) + duration = parse_duration(try_get( + renderer, lambda x: x['lengthText']['simpleText'], compat_str)) + view_count_text = try_get( + renderer, lambda x: x['viewCountText']['simpleText'], compat_str) or '' + view_count = str_to_int(self._search_regex( + r'^([\d,]+)', re.sub(r'\s', '', view_count_text), + 'view count', default=None)) + uploader = try_get( + renderer, + (lambda x: x['ownerText']['runs'][0]['text'], + lambda x: x['shortBylineText']['runs'][0]['text']), compat_str) + return { + '_type': 'url', + 'ie_key': YoutubeIE.ie_key(), + 'id': video_id, + 'url': video_id, + 'title': title, + 'description': description, + 'duration': duration, + 'view_count': view_count, + 'uploader': uploader, + } def _search_results(self, query, params): data = { @@ -337,43 +368,26 @@ class YoutubeBaseInfoExtractor(InfoExtractor): list) if not slr_contents: break - isr_contents = try_get( - slr_contents, - lambda x: x[0]['itemSectionRenderer']['contents'], - list) - if not isr_contents: - break - for content in isr_contents: - if not isinstance(content, dict): - continue - video = content.get('videoRenderer') - if not isinstance(video, dict): - continue - video_id = video.get('videoId') - if not video_id: + for slr_content in slr_contents: + isr_contents = try_get( + slr_content, + lambda x: x['itemSectionRenderer']['contents'], + list) + if not isr_contents: continue - title = try_get(video, lambda x: x['title']['runs'][0]['text'], compat_str) - description = try_get(video, lambda x: x['descriptionSnippet']['runs'][0]['text'], compat_str) - duration = parse_duration(try_get(video, lambda x: x['lengthText']['simpleText'], compat_str)) - view_count_text = try_get(video, lambda x: x['viewCountText']['simpleText'], compat_str) or '' - view_count = int_or_none(self._search_regex( - r'^(\d+)', re.sub(r'\s', '', view_count_text), - 'view count', default=None)) - uploader = try_get(video, lambda x: x['ownerText']['runs'][0]['text'], compat_str) - yield { - '_type': 'url_transparent', - 'ie_key': YoutubeIE.ie_key(), - 'id': video_id, - 'url': video_id, - 'title': title, - 'description': description, - 'duration': duration, - 'view_count': view_count, - 'uploader': uploader, - } + for content in isr_contents: + if not isinstance(content, dict): + continue + video = content.get('videoRenderer') + if not isinstance(video, dict): + continue + video_id = video.get('videoId') + if not video_id: + continue + yield self._extract_video(video) token = try_get( slr_contents, - lambda x: x[1]['continuationItemRenderer']['continuationEndpoint']['continuationCommand']['token'], + lambda x: x[-1]['continuationItemRenderer']['continuationEndpoint']['continuationCommand']['token'], compat_str) if not token: break @@ -382,52 +396,76 @@ class YoutubeBaseInfoExtractor(InfoExtractor): class YoutubeIE(YoutubeBaseInfoExtractor): IE_DESC = 'YouTube.com' + _INVIDIOUS_SITES = ( + # invidious-redirect websites + r'(?:www\.)?redirect\.invidious\.io', + r'(?:(?:www|dev)\.)?invidio\.us', + # Invidious instances taken from https://github.com/iv-org/documentation/blob/master/Invidious-Instances.md + r'(?:(?:www|no)\.)?invidiou\.sh', + r'(?:(?:www|fi)\.)?invidious\.snopyta\.org', + r'(?:www\.)?invidious\.kabi\.tk', + r'(?:www\.)?invidious\.13ad\.de', + r'(?:www\.)?invidious\.mastodon\.host', + r'(?:www\.)?invidious\.zapashcanon\.fr', + r'(?:www\.)?(?:invidious(?:-us)?|piped)\.kavin\.rocks', + r'(?:www\.)?invidious\.tinfoil-hat\.net', + r'(?:www\.)?invidious\.himiko\.cloud', + r'(?:www\.)?invidious\.reallyancient\.tech', + r'(?:www\.)?invidious\.tube', + r'(?:www\.)?invidiou\.site', + r'(?:www\.)?invidious\.site', + r'(?:www\.)?invidious\.xyz', + r'(?:www\.)?invidious\.nixnet\.xyz', + r'(?:www\.)?invidious\.048596\.xyz', + r'(?:www\.)?invidious\.drycat\.fr', + r'(?:www\.)?inv\.skyn3t\.in', + r'(?:www\.)?tube\.poal\.co', + r'(?:www\.)?tube\.connect\.cafe', + r'(?:www\.)?vid\.wxzm\.sx', + r'(?:www\.)?vid\.mint\.lgbt', + r'(?:www\.)?vid\.puffyan\.us', + r'(?:www\.)?yewtu\.be', + r'(?:www\.)?yt\.elukerio\.org', + r'(?:www\.)?yt\.lelux\.fi', + r'(?:www\.)?invidious\.ggc-project\.de', + r'(?:www\.)?yt\.maisputain\.ovh', + r'(?:www\.)?ytprivate\.com', + r'(?:www\.)?invidious\.13ad\.de', + r'(?:www\.)?invidious\.toot\.koeln', + r'(?:www\.)?invidious\.fdn\.fr', + r'(?:www\.)?watch\.nettohikari\.com', + r'(?:www\.)?invidious\.namazso\.eu', + r'(?:www\.)?invidious\.silkky\.cloud', + r'(?:www\.)?invidious\.exonip\.de', + r'(?:www\.)?invidious\.riverside\.rocks', + r'(?:www\.)?invidious\.blamefran\.net', + r'(?:www\.)?invidious\.moomoo\.de', + r'(?:www\.)?ytb\.trom\.tf', + r'(?:www\.)?yt\.cyberhost\.uk', + r'(?:www\.)?kgg2m7yk5aybusll\.onion', + r'(?:www\.)?qklhadlycap4cnod\.onion', + r'(?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion', + r'(?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion', + r'(?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion', + r'(?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion', + r'(?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p', + r'(?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion', + r'(?:www\.)?w6ijuptxiku4xpnnaetxvnkc5vqcdu7mgns2u77qefoixi63vbvnpnqd\.onion', + r'(?:www\.)?kbjggqkzv65ivcqj6bumvp337z6264huv5kpkwuv6gu5yjiskvan7fad\.onion', + r'(?:www\.)?grwp24hodrefzvjjuccrkw3mjq4tzhaaq32amf33dzpmuxe7ilepcmad\.onion', + r'(?:www\.)?hpniueoejy4opn7bc4ftgazyqjoeqwlvh2uiku2xqku6zpoa4bf5ruid\.onion', + ) _VALID_URL = r"""(?x)^ ( (?:https?://|//) # http(s):// or protocol-independent URL - (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com/| - (?:www\.)?deturl\.com/www\.youtube\.com/| - (?:www\.)?pwnyoutube\.com/| - (?:www\.)?hooktube\.com/| - (?:www\.)?yourepeat\.com/| - tube\.majestyc\.net/| - # Invidious instances taken from https://github.com/omarroth/invidious/wiki/Invidious-Instances - (?:(?:www|dev)\.)?invidio\.us/| - (?:(?:www|no)\.)?invidiou\.sh/| - (?:(?:www|fi)\.)?invidious\.snopyta\.org/| - (?:www\.)?invidious\.kabi\.tk/| - (?:www\.)?invidious\.13ad\.de/| - (?:www\.)?invidious\.mastodon\.host/| - (?:www\.)?invidious\.zapashcanon\.fr/| - (?:www\.)?invidious\.kavin\.rocks/| - (?:www\.)?invidious\.tube/| - (?:www\.)?invidiou\.site/| - (?:www\.)?invidious\.site/| - (?:www\.)?invidious\.xyz/| - (?:www\.)?invidious\.nixnet\.xyz/| - (?:www\.)?invidious\.drycat\.fr/| - (?:www\.)?tube\.poal\.co/| - (?:www\.)?tube\.connect\.cafe/| - (?:www\.)?vid\.wxzm\.sx/| - (?:www\.)?vid\.mint\.lgbt/| - (?:www\.)?yewtu\.be/| - (?:www\.)?yt\.elukerio\.org/| - (?:www\.)?yt\.lelux\.fi/| - (?:www\.)?invidious\.ggc-project\.de/| - (?:www\.)?yt\.maisputain\.ovh/| - (?:www\.)?invidious\.13ad\.de/| - (?:www\.)?invidious\.toot\.koeln/| - (?:www\.)?invidious\.fdn\.fr/| - (?:www\.)?watch\.nettohikari\.com/| - (?:www\.)?kgg2m7yk5aybusll\.onion/| - (?:www\.)?qklhadlycap4cnod\.onion/| - (?:www\.)?axqzx4s6s54s32yentfqojs3x5i7faxza6xo3ehd4bzzsg2ii4fv2iid\.onion/| - (?:www\.)?c7hqkpkpemu6e7emz5b4vyz7idjgdvgaaa3dyimmeojqbgpea3xqjoid\.onion/| - (?:www\.)?fz253lmuao3strwbfbmx46yu7acac2jz27iwtorgmbqlkurlclmancad\.onion/| - (?:www\.)?invidious\.l4qlywnpwqsluw65ts7md3khrivpirse744un3x7mlskqauz5pyuzgqd\.onion/| - (?:www\.)?owxfohz4kjyv25fvlqilyxast7inivgiktls3th44jhk3ej3i7ya\.b32\.i2p/| - (?:www\.)?4l2dgddgsrkf2ous66i6seeyi6etzfgrue332grh2n7madpwopotugyd\.onion/| - youtube\.googleapis\.com/) # the various hostnames, with wildcard subdomains + (?:(?:(?:(?:\w+\.)?[yY][oO][uU][tT][uU][bB][eE](?:-nocookie|kids)?\.com| + (?:www\.)?deturl\.com/www\.youtube\.com| + (?:www\.)?pwnyoutube\.com| + (?:www\.)?hooktube\.com| + (?:www\.)?yourepeat\.com| + tube\.majestyc\.net| + %(invidious)s| + youtube\.googleapis\.com)/ # the various hostnames, with wildcard subdomains (?:.*?\#/)? # handle anchor (#/) redirect urls (?: # the various things that can precede the ID: (?:(?:v|embed|e)/(?!videoseries)) # v/ or embed/ or e/ @@ -442,129 +480,21 @@ class YoutubeIE(YoutubeBaseInfoExtractor): youtu\.be| # just youtu.be/xxxx vid\.plus| # or vid.plus/xxxx zwearz\.com/watch| # or zwearz.com/watch/xxxx + %(invidious)s )/ |(?:www\.)?cleanvideosearch\.com/media/action/yt/watch\?videoId= ) )? # all until now is optional -> you can pass the naked ID - (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID - (?!.*?\blist= - (?: - %(playlist_id)s| # combined list/video URLs are handled by the playlist IE - WL # WL are handled by the watch later IE - ) - ) + (?P<id>[0-9A-Za-z_-]{11}) # here is it! the YouTube video ID (?(1).+)? # if we found the ID, everything can follow - $""" % {'playlist_id': YoutubeBaseInfoExtractor._PLAYLIST_ID_RE} - _NEXT_URL_RE = r'[\?&]next_url=([^&]+)' + $""" % { + 'invidious': '|'.join(_INVIDIOUS_SITES), + } _PLAYER_INFO_RE = ( - r'/(?P<id>[a-zA-Z0-9_-]{8,})/player_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?/base\.(?P<ext>[a-z]+)$', - r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.(?P<ext>[a-z]+)$', + r'/s/player/(?P<id>[a-zA-Z0-9_-]{8,})/player', + r'/(?P<id>[a-zA-Z0-9_-]{8,})/player(?:_ias\.vflset(?:/[a-zA-Z]{2,3}_[a-zA-Z]{2,3})?|-plasma-ias-(?:phone|tablet)-[a-z]{2}_[A-Z]{2}\.vflset)/base\.js$', + r'\b(?P<id>vfl[a-zA-Z0-9_-]+)\b.*?\.js$', ) - _formats = { - '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'}, - '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'}, - '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'}, - '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, - '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, - '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, - # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well - '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'}, - '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, - '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, - '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, - '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, - '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, - '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, - '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, - '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, - - - # 3D videos - '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, - '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, - '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, - '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, - '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20}, - '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, - '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, - - # Apple HTTP Live Streaming - '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, - '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, - '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, - '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, - '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, - '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, - '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, - '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10}, - - # DASH mp4 video - '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'}, - '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/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'}, - '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60}, - '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60}, - '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'}, - - # Dash mp4 audio - '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'}, - '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'}, - '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'}, - '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'}, - '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'}, - '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'}, - '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'}, - - # Dash webm - '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, - '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'}, - '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug) - '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, - '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, - '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, - '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'}, - '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, - - # Dash webm audio - '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128}, - '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256}, - - # Dash webm audio with opus inside - '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50}, - '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70}, - '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160}, - - # 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 = ('srv1', 'srv2', 'srv3', 'ttml', 'vtt') _GEO_BYPASS = False @@ -607,7 +537,8 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': 'setindia', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/setindia', 'age_limit': 18, - } + }, + 'skip': 'Private video', }, { 'url': 'https://www.youtube.com/watch?v=BaW_jenozKc&v=yZIXLfi8CZQ', @@ -663,6 +594,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader': 'AfrojackVEVO', 'uploader_id': 'AfrojackVEVO', 'upload_date': '20131011', + 'abr': 129.495, }, 'params': { 'youtube_include_dash_manifest': True, @@ -681,7 +613,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'uploader_id': 'TheAmazingAtheist', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/TheAmazingAtheist', 'title': 'Burning Everyone\'s Koran', - 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms\n\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html', + 'description': 'SUBSCRIBE: http://www.youtube.com/saturninefilms \r\n\r\nEven Obama has taken a stand against freedom on this issue: http://www.huffingtonpost.com/2010/09/09/obama-gma-interview-quran_n_710282.html', } }, # Normal age-gate video (No vevo, embed allowed), available via embed page @@ -717,11 +649,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'upload_date': '20100430', 'uploader_id': 'deadmau5', 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/deadmau5', - 'creator': 'Dada Life, deadmau5', - 'description': 'md5:12c56784b8032162bb936a5f76d55360', + 'creator': 'deadmau5', + 'description': 'md5:6cbcd3a92ce1bc676fc4d6ab4ace2336', 'uploader': 'deadmau5', 'title': 'Deadmau5 - Some Chords (HD)', - 'alt_title': 'This Machine Kills Some Chords', + 'alt_title': 'Some Chords', }, 'expected_warnings': [ 'DASH manifest missing', @@ -816,69 +748,64 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }, { # Multifeed videos (multiple cameras), URL is for Main Camera - 'url': 'https://www.youtube.com/watch?v=jqWvoWXjCVs', + 'url': 'https://www.youtube.com/watch?v=jvGDaLqkpTg', 'info_dict': { - 'id': 'jqWvoWXjCVs', - 'title': 'teamPGP: Rocket League Noob Stream', - 'description': 'md5:dc7872fb300e143831327f1bae3af010', + 'id': 'jvGDaLqkpTg', + 'title': 'Tom Clancy Free Weekend Rainbow Whatever', + 'description': 'md5:e03b909557865076822aa169218d6a5d', }, 'playlist': [{ 'info_dict': { - 'id': 'jqWvoWXjCVs', + 'id': 'jvGDaLqkpTg', 'ext': 'mp4', - 'title': 'teamPGP: Rocket League Noob Stream (Main Camera)', - 'description': 'md5:dc7872fb300e143831327f1bae3af010', - 'duration': 7335, - 'upload_date': '20150721', - 'uploader': 'Beer Games Beer', - 'uploader_id': 'beergamesbeer', - 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer', - 'license': 'Standard YouTube License', + 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Main Camera)', + 'description': 'md5:e03b909557865076822aa169218d6a5d', + 'duration': 10643, + 'upload_date': '20161111', + 'uploader': 'Team PGP', + 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg', + 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg', }, }, { 'info_dict': { - 'id': '6h8e8xoXJzg', + 'id': '3AKt1R1aDnw', 'ext': 'mp4', - 'title': 'teamPGP: Rocket League Noob Stream (kreestuh)', - 'description': 'md5:dc7872fb300e143831327f1bae3af010', - 'duration': 7337, - 'upload_date': '20150721', - 'uploader': 'Beer Games Beer', - 'uploader_id': 'beergamesbeer', - 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer', - 'license': 'Standard YouTube License', + 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 2)', + 'description': 'md5:e03b909557865076822aa169218d6a5d', + 'duration': 10991, + 'upload_date': '20161111', + 'uploader': 'Team PGP', + 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg', + 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg', }, }, { 'info_dict': { - 'id': 'PUOgX5z9xZw', + 'id': 'RtAMM00gpVc', 'ext': 'mp4', - 'title': 'teamPGP: Rocket League Noob Stream (grizzle)', - 'description': 'md5:dc7872fb300e143831327f1bae3af010', - 'duration': 7337, - 'upload_date': '20150721', - 'uploader': 'Beer Games Beer', - 'uploader_id': 'beergamesbeer', - 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer', - 'license': 'Standard YouTube License', + 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 3)', + 'description': 'md5:e03b909557865076822aa169218d6a5d', + 'duration': 10995, + 'upload_date': '20161111', + 'uploader': 'Team PGP', + 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg', + 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg', }, }, { 'info_dict': { - 'id': 'teuwxikvS5k', + 'id': '6N2fdlP3C5U', 'ext': 'mp4', - 'title': 'teamPGP: Rocket League Noob Stream (zim)', - 'description': 'md5:dc7872fb300e143831327f1bae3af010', - 'duration': 7334, - 'upload_date': '20150721', - 'uploader': 'Beer Games Beer', - 'uploader_id': 'beergamesbeer', - 'uploader_url': r're:https?://(?:www\.)?youtube\.com/user/beergamesbeer', - 'license': 'Standard YouTube License', + 'title': 'Tom Clancy Free Weekend Rainbow Whatever (Camera 4)', + 'description': 'md5:e03b909557865076822aa169218d6a5d', + 'duration': 10990, + 'upload_date': '20161111', + 'uploader': 'Team PGP', + 'uploader_id': 'UChORY56LMMETTuGjXaJXvLg', + 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UChORY56LMMETTuGjXaJXvLg', }, }], 'params': { 'skip_download': True, }, - 'skip': 'This video is not available.', }, { # Multifeed video with comma in title (see https://github.com/ytdl-org/youtube-dl/issues/8536) @@ -947,6 +874,11 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'skip': 'This video does not exist.', }, { + # Video with incomplete 'yt:stretch=16:' + 'url': 'https://www.youtube.com/watch?v=FRhJzUSJbGI', + 'only_matching': True, + }, + { # Video licensed under Creative Commons 'url': 'https://www.youtube.com/watch?v=M4gD1WSo5mA', 'info_dict': { @@ -972,7 +904,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'id': 'eQcmzGIKrzg', 'ext': 'mp4', 'title': 'Democratic Socialism and Foreign Policy | Bernie Sanders', - 'description': 'md5:dda0d780d5a6e120758d1711d062a867', + 'description': 'md5:13a2503d7b5904ef4b223aa101628f39', 'duration': 4060, 'upload_date': '20151119', 'uploader': 'Bernie Sanders', @@ -1019,7 +951,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'id': 'iqKdEhx-dD4', 'ext': 'mp4', 'title': 'Isolation - Mind Field (Ep 1)', - 'description': 'md5:46a29be4ceffa65b92d277b93f463c0f', + 'description': 'md5:f540112edec5d09fc8cc752d3d4ba3cd', 'duration': 2085, 'upload_date': '20170118', 'uploader': 'Vsauce', @@ -1054,6 +986,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'params': { 'skip_download': True, }, + 'skip': 'This video has been removed for violating YouTube\'s policy on hate speech.', }, { # itag 212 @@ -1070,6 +1003,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'only_matching': True, }, { + 'url': 'https://redirect.invidious.io/watch?v=BaW_jenozKc', + 'only_matching': True, + }, + { + # from https://nitter.pussthecat.org/YouTube/status/1360363141947944964#m + 'url': 'https://redirect.invidious.io/Yh0AhrY9GjA', + 'only_matching': True, + }, + { # DRM protected 'url': 'https://www.youtube.com/watch?v=s7_qI6_mIXc', 'only_matching': True, @@ -1139,6 +1081,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'params': { 'skip_download': True, }, + 'skip': 'Video unavailable', }, { # empty description results in an empty string @@ -1184,27 +1127,171 @@ class YoutubeIE(YoutubeBaseInfoExtractor): 'url': 'https://www.youtube.com/watch_popup?v=63RmMXCd_bQ', 'only_matching': True, }, + { + # https://github.com/ytdl-org/youtube-dl/pull/28094 + 'url': 'OtqTfy26tG0', + 'info_dict': { + 'id': 'OtqTfy26tG0', + 'ext': 'mp4', + 'title': 'Burn Out', + 'description': 'md5:8d07b84dcbcbfb34bc12a56d968b6131', + 'upload_date': '20141120', + 'uploader': 'The Cinematic Orchestra - Topic', + 'uploader_id': 'UCIzsJBIyo8hhpFm1NK0uLgw', + 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UCIzsJBIyo8hhpFm1NK0uLgw', + 'artist': 'The Cinematic Orchestra', + 'track': 'Burn Out', + 'album': 'Every Day', + 'release_data': None, + 'release_year': None, + }, + 'params': { + 'skip_download': True, + }, + }, + { + # controversial video, only works with bpctr when authenticated with cookies + 'url': 'https://www.youtube.com/watch?v=nGC3D_FkCmg', + 'only_matching': True, + }, + { + # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685 + 'url': 'cBvYw8_A0vQ', + 'info_dict': { + 'id': 'cBvYw8_A0vQ', + 'ext': 'mp4', + 'title': '4K Ueno Okachimachi Street Scenes 上野御徒町歩き', + 'description': 'md5:ea770e474b7cd6722b4c95b833c03630', + 'upload_date': '20201120', + 'uploader': 'Walk around Japan', + 'uploader_id': 'UC3o_t8PzBmXf5S9b7GLx1Mw', + 'uploader_url': r're:https?://(?:www\.)?youtube\.com/channel/UC3o_t8PzBmXf5S9b7GLx1Mw', + }, + 'params': { + 'skip_download': True, + }, + }, ] + _formats = { + '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'}, + '17': {'ext': '3gp', 'width': 176, 'height': 144, 'acodec': 'aac', 'abr': 24, 'vcodec': 'mp4v'}, + '18': {'ext': 'mp4', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 96, 'vcodec': 'h264'}, + '22': {'ext': 'mp4', 'width': 1280, 'height': 720, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, + '34': {'ext': 'flv', 'width': 640, 'height': 360, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + '35': {'ext': 'flv', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + # itag 36 videos are either 320x180 (BaW_jenozKc) or 320x240 (__2ABJjxzNo), abr varies as well + '36': {'ext': '3gp', 'width': 320, 'acodec': 'aac', 'vcodec': 'mp4v'}, + '37': {'ext': 'mp4', 'width': 1920, 'height': 1080, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, + '38': {'ext': 'mp4', 'width': 4096, 'height': 3072, 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264'}, + '43': {'ext': 'webm', 'width': 640, 'height': 360, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, + '44': {'ext': 'webm', 'width': 854, 'height': 480, 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8'}, + '45': {'ext': 'webm', 'width': 1280, 'height': 720, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, + '46': {'ext': 'webm', 'width': 1920, 'height': 1080, 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8'}, + '59': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, + '78': {'ext': 'mp4', 'width': 854, 'height': 480, 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264'}, - def __init__(self, *args, **kwargs): - super(YoutubeIE, self).__init__(*args, **kwargs) - self._player_cache = {} - def report_video_info_webpage_download(self, video_id): - """Report attempt to download video info webpage.""" - self.to_screen('%s: Downloading video info webpage' % video_id) + # 3D videos + '82': {'ext': 'mp4', 'height': 360, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, + '83': {'ext': 'mp4', 'height': 480, 'format_note': '3D', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -20}, + '84': {'ext': 'mp4', 'height': 720, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, + '85': {'ext': 'mp4', 'height': 1080, 'format_note': '3D', 'acodec': 'aac', 'abr': 192, 'vcodec': 'h264', 'preference': -20}, + '100': {'ext': 'webm', 'height': 360, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 128, 'vcodec': 'vp8', 'preference': -20}, + '101': {'ext': 'webm', 'height': 480, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, + '102': {'ext': 'webm', 'height': 720, 'format_note': '3D', 'acodec': 'vorbis', 'abr': 192, 'vcodec': 'vp8', 'preference': -20}, + + # Apple HTTP Live Streaming + '91': {'ext': 'mp4', 'height': 144, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, + '92': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, + '93': {'ext': 'mp4', 'height': 360, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, + '94': {'ext': 'mp4', 'height': 480, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 128, 'vcodec': 'h264', 'preference': -10}, + '95': {'ext': 'mp4', 'height': 720, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, + '96': {'ext': 'mp4', 'height': 1080, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 256, 'vcodec': 'h264', 'preference': -10}, + '132': {'ext': 'mp4', 'height': 240, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 48, 'vcodec': 'h264', 'preference': -10}, + '151': {'ext': 'mp4', 'height': 72, 'format_note': 'HLS', 'acodec': 'aac', 'abr': 24, 'vcodec': 'h264', 'preference': -10}, + + # DASH mp4 video + '133': {'ext': 'mp4', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '134': {'ext': 'mp4', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'h264'}, + '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/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'}, + '298': {'ext': 'mp4', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60}, + '299': {'ext': 'mp4', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'h264', 'fps': 60}, + '266': {'ext': 'mp4', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'h264'}, + + # Dash mp4 audio + '139': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 48, 'container': 'm4a_dash'}, + '140': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 128, 'container': 'm4a_dash'}, + '141': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'abr': 256, 'container': 'm4a_dash'}, + '256': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'}, + '258': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'aac', 'container': 'm4a_dash'}, + '325': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'dtse', 'container': 'm4a_dash'}, + '328': {'ext': 'm4a', 'format_note': 'DASH audio', 'acodec': 'ec-3', 'container': 'm4a_dash'}, + + # Dash webm + '167': {'ext': 'webm', 'height': 360, 'width': 640, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '168': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '169': {'ext': 'webm', 'height': 720, 'width': 1280, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '170': {'ext': 'webm', 'height': 1080, 'width': 1920, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '218': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '219': {'ext': 'webm', 'height': 480, 'width': 854, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp8'}, + '278': {'ext': 'webm', 'height': 144, 'format_note': 'DASH video', 'container': 'webm', 'vcodec': 'vp9'}, + '242': {'ext': 'webm', 'height': 240, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '243': {'ext': 'webm', 'height': 360, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '244': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '245': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '246': {'ext': 'webm', 'height': 480, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '247': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '248': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '271': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + # itag 272 videos are either 3840x2160 (e.g. RtoitU2A-3E) or 7680x4320 (sLprVF6d7Ug) + '272': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '302': {'ext': 'webm', 'height': 720, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, + '303': {'ext': 'webm', 'height': 1080, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, + '308': {'ext': 'webm', 'height': 1440, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, + '313': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9'}, + '315': {'ext': 'webm', 'height': 2160, 'format_note': 'DASH video', 'vcodec': 'vp9', 'fps': 60}, + + # Dash webm audio + '171': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 128}, + '172': {'ext': 'webm', 'acodec': 'vorbis', 'format_note': 'DASH audio', 'abr': 256}, - def report_information_extraction(self, video_id): - """Report attempt to extract video information.""" - self.to_screen('%s: Extracting video information' % video_id) + # Dash webm audio with opus inside + '249': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 50}, + '250': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 70}, + '251': {'ext': 'webm', 'format_note': 'DASH audio', 'acodec': 'opus', 'abr': 160}, - def report_unavailable_format(self, video_id, format): - """Report extracted video URL.""" - self.to_screen('%s: Format %s not available' % (video_id, format)) + # RTMP (unnamed) + '_rtmp': {'protocol': 'rtmp'}, - def report_rtmp_download(self): - """Indicate the download will use the RTMP protocol.""" - self.to_screen('RTMP download detected') + # 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'}, + } + + @classmethod + def suitable(cls, url): + # Hack for lazy extractors until more generic solution is implemented + # (see #28780) + from .youtube import parse_qs + qs = parse_qs(url) + if qs.get('list', [None])[0]: + return False + return super(YoutubeIE, cls).suitable(url) + + def __init__(self, *args, **kwargs): + super(YoutubeIE, self).__init__(*args, **kwargs) + self._code_cache = {} + self._player_cache = {} def _signature_cache_id(self, example_sig): """ Return a string representation of a signature """ @@ -1218,40 +1305,27 @@ class YoutubeIE(YoutubeBaseInfoExtractor): break else: raise ExtractorError('Cannot identify player %r' % player_url) - return id_m.group('ext'), id_m.group('id') + return id_m.group('id') def _extract_signature_function(self, video_id, player_url, example_sig): - player_type, player_id = self._extract_player_info(player_url) + player_id = self._extract_player_info(player_url) # Read from filesystem cache - func_id = '%s_%s_%s' % ( - player_type, player_id, self._signature_cache_id(example_sig)) + func_id = 'js_%s_%s' % ( + player_id, self._signature_cache_id(example_sig)) assert os.path.basename(func_id) == func_id cache_spec = self._downloader.cache.load('youtube-sigfuncs', func_id) if cache_spec is not None: return lambda s: ''.join(s[i] for i in cache_spec) - download_note = ( - 'Downloading player %s' % player_url - if self._downloader.params.get('verbose') else - 'Downloading %s player %s' % (player_type, player_id) - ) - if player_type == 'js': - code = self._download_webpage( - player_url, video_id, - note=download_note, - errnote='Download of %s failed' % player_url) - res = self._parse_sig_js(code) - elif player_type == 'swf': - urlh = self._request_webpage( + if player_id not in self._code_cache: + self._code_cache[player_id] = self._download_webpage( player_url, video_id, - note=download_note, + note='Downloading player ' + player_id, errnote='Download of %s failed' % player_url) - code = urlh.read() - res = self._parse_sig_swf(code) - else: - assert False, 'Invalid player type %r' % player_type + code = self._code_cache[player_id] + res = self._parse_sig_js(code) test_string = ''.join(map(compat_chr, range(len(example_sig)))) cache_res = res(test_string) @@ -1303,7 +1377,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): funcname = self._search_regex( (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'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)', + r'\bm=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(h\.s\)\)', + r'\bc&&\(c=(?P<sig>[a-zA-Z0-9$]{2,})\(decodeURIComponent\(c\)\)', + r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\);[a-zA-Z0-9$]{2}\.[a-zA-Z0-9$]{2}\(a,\d+\)', + r'(?:\b|[^a-zA-Z0-9$])(?P<sig>[a-zA-Z0-9$]{2,})\s*=\s*function\(\s*a\s*\)\s*{\s*a\s*=\s*a\.split\(\s*""\s*\)', 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$]+)\(', @@ -1320,14 +1397,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): initial_function = jsi.extract_function(funcname) return lambda s: initial_function([s]) - def _parse_sig_swf(self, file_contents): - swfi = SWFInterpreter(file_contents) - TARGET_CLASSNAME = 'SignatureDecipher' - searched_class = swfi.extract_class(TARGET_CLASSNAME) - initial_function = swfi.extract_function(searched_class, 'decipher') - return lambda s: initial_function([s]) - - def _decrypt_signature(self, s, video_id, player_url, age_gate=False): + def _decrypt_signature(self, s, video_id, player_url): """Turn the encrypted s field into a working signature""" if player_url is None: @@ -1354,158 +1424,10 @@ class YoutubeIE(YoutubeBaseInfoExtractor): raise ExtractorError( 'Signature extraction failed: ' + tb, cause=e) - def _get_subtitles(self, video_id, webpage): - try: - subs_doc = self._download_xml( - 'https://video.google.com/timedtext?hl=en&type=list&v=%s' % video_id, - video_id, note=False) - except ExtractorError as err: - self._downloader.report_warning('unable to download video subtitles: %s' % error_to_compat_str(err)) - return {} - - sub_lang_list = {} - for track in subs_doc.findall('track'): - lang = track.attrib['lang_code'] - if lang in sub_lang_list: - continue - sub_formats = [] - for ext in self._SUBTITLE_FORMATS: - params = compat_urllib_parse_urlencode({ - 'lang': lang, - 'v': video_id, - 'fmt': ext, - 'name': track.attrib['name'].encode('utf-8'), - }) - sub_formats.append({ - 'url': 'https://www.youtube.com/api/timedtext?' + params, - 'ext': ext, - }) - sub_lang_list[lang] = sub_formats - if not sub_lang_list: - self._downloader.report_warning('video doesn\'t have subtitles') - return {} - return sub_lang_list - - def _get_ytplayer_config(self, video_id, webpage): - patterns = ( - # User data may contain arbitrary character sequences that may affect - # JSON extraction with regex, e.g. when '};' is contained the second - # 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/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*({.+?});', - ) - config = self._search_regex( - patterns, webpage, 'ytplayer.config', default=None) - if config: - return self._parse_json( - uppercase_escape(config), video_id, fatal=False) - - def _get_automatic_captions(self, video_id, player_response, player_config): - """We need the webpage for getting the captions url, pass it as an - argument to speed up the process.""" - self.to_screen('%s: Looking for automatic captions' % video_id) - err_msg = 'Couldn\'t find automatic captions for %s' % video_id - if not (player_response or player_config): - self._downloader.report_warning(err_msg) - return {} - try: - args = player_config.get('args') if player_config else {} - caption_url = args.get('ttsurl') - if caption_url: - timestamp = args['timestamp'] - # We get the available subtitles - list_params = compat_urllib_parse_urlencode({ - 'type': 'list', - 'tlangs': 1, - 'asrs': 1, - }) - list_url = caption_url + '&' + list_params - caption_list = self._download_xml(list_url, video_id) - original_lang_node = caption_list.find('track') - if original_lang_node is None: - self._downloader.report_warning('Video doesn\'t have automatic captions') - return {} - original_lang = original_lang_node.attrib['lang_code'] - caption_kind = original_lang_node.attrib.get('kind', '') - - sub_lang_list = {} - for lang_node in caption_list.findall('target'): - sub_lang = lang_node.attrib['lang_code'] - sub_formats = [] - for ext in self._SUBTITLE_FORMATS: - params = compat_urllib_parse_urlencode({ - 'lang': original_lang, - 'tlang': sub_lang, - 'fmt': ext, - 'ts': timestamp, - 'kind': caption_kind, - }) - sub_formats.append({ - 'url': caption_url + '&' + params, - 'ext': ext, - }) - sub_lang_list[sub_lang] = sub_formats - return sub_lang_list - - def make_captions(sub_url, sub_langs): - parsed_sub_url = compat_urllib_parse_urlparse(sub_url) - caption_qs = compat_parse_qs(parsed_sub_url.query) - captions = {} - for sub_lang in sub_langs: - sub_formats = [] - for ext in self._SUBTITLE_FORMATS: - caption_qs.update({ - 'tlang': [sub_lang], - 'fmt': [ext], - }) - sub_url = compat_urlparse.urlunparse(parsed_sub_url._replace( - query=compat_urllib_parse_urlencode(caption_qs, True))) - sub_formats.append({ - 'url': sub_url, - 'ext': ext, - }) - captions[sub_lang] = sub_formats - return captions - - # New captions format as of 22.06.2017 - if player_response: - renderer = player_response['captions']['playerCaptionsTracklistRenderer'] - base_url = renderer['captionTracks'][0]['baseUrl'] - sub_lang_list = [] - for lang in renderer['translationLanguages']: - lang_code = lang.get('languageCode') - if lang_code: - sub_lang_list.append(lang_code) - return make_captions(base_url, sub_lang_list) - - # Some videos don't provide ttsurl but rather caption_tracks and - # caption_translation_languages (e.g. 20LmZk1hakA) - # Does not used anymore as of 22.06.2017 - caption_tracks = args['caption_tracks'] - caption_translation_languages = args['caption_translation_languages'] - caption_url = compat_parse_qs(caption_tracks.split(',')[0])['u'][0] - sub_lang_list = [] - for lang in caption_translation_languages.split(','): - lang_qs = compat_parse_qs(compat_urllib_parse_unquote_plus(lang)) - sub_lang = lang_qs.get('lc', [None])[0] - if sub_lang: - sub_lang_list.append(sub_lang) - return make_captions(caption_url, sub_lang_list) - # An extractor error can be raise by the download process if there are - # no automatic captions but there are subtitles - except (KeyError, IndexError, ExtractorError): - self._downloader.report_warning(err_msg) - return {} - - def _mark_watched(self, video_id, video_info, player_response): + def _mark_watched(self, video_id, player_response): playback_url = url_or_none(try_get( player_response, - lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl']) or try_get( - video_info, lambda x: x['videostats_playback_base_url'][0])) + lambda x: x['playbackTracking']['videostatsPlaybackUrl']['baseUrl'])) if not playback_url: return parsed_playback_url = compat_urlparse.urlparse(playback_url) @@ -1572,12 +1494,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): video_id = mobj.group(2) return video_id - def _extract_chapters_from_json(self, webpage, video_id, duration): - if not webpage: - return - data = self._extract_yt_initial_data(video_id, webpage) - if not data or not isinstance(data, dict): - return + def _extract_chapters_from_json(self, data, video_id, duration): chapters_list = try_get( data, lambda x: x['playerOverlays'] @@ -1617,244 +1534,89 @@ class YoutubeIE(YoutubeBaseInfoExtractor): }) return chapters - @staticmethod - def _extract_chapters_from_description(description, duration): - if not description: - return None - chapter_lines = re.findall( - r'(?:^|<br\s*/>)([^<]*<a[^>]+onclick=["\']yt\.www\.watch\.player\.seekTo[^>]+>(\d{1,2}:\d{1,2}(?::\d{1,2})?)</a>[^>]*)(?=$|<br\s*/>)', - description) - if not chapter_lines: - return None - chapters = [] - for next_num, (chapter_line, time_point) in enumerate( - chapter_lines, start=1): - start_time = parse_duration(time_point) - if start_time is None: - continue - if start_time > duration: - break - end_time = (duration if next_num == len(chapter_lines) - else parse_duration(chapter_lines[next_num][1])) - if end_time is None: - continue - if end_time > duration: - end_time = duration - if start_time > end_time: - break - chapter_title = re.sub( - r'<a[^>]+>[^<]+</a>', '', chapter_line).strip(' \t-') - chapter_title = re.sub(r'\s+', ' ', chapter_title) - chapters.append({ - 'start_time': start_time, - 'end_time': end_time, - 'title': chapter_title, - }) - return chapters - - def _extract_chapters(self, webpage, description, video_id, duration): - return (self._extract_chapters_from_json(webpage, video_id, duration) - or self._extract_chapters_from_description(description, duration)) + def _extract_yt_initial_variable(self, webpage, regex, video_id, name): + return self._parse_json(self._search_regex( + (r'%s\s*%s' % (regex, self._YT_INITIAL_BOUNDARY_RE), + regex), webpage, name, default='{}'), video_id, fatal=False) def _real_extract(self, url): url, smuggled_data = unsmuggle_url(url, {}) + video_id = self._match_id(url) + base_url = self.http_scheme() + '//www.youtube.com/' + webpage_url = base_url + 'watch?v=' + video_id + webpage = self._download_webpage( + webpage_url + '&bpctr=9999999999&has_verified=1', video_id, fatal=False) + + player_response = None + if webpage: + player_response = self._extract_yt_initial_variable( + webpage, self._YT_INITIAL_PLAYER_RESPONSE_RE, + video_id, 'initial player response') + if not player_response: + player_response = self._call_api( + 'player', {'videoId': video_id}, video_id) + + playability_status = player_response.get('playabilityStatus') or {} + if playability_status.get('reason') == 'Sign in to confirm your age': + video_info = self._download_webpage( + base_url + 'get_video_info', video_id, + 'Refetching age-gated info webpage', + 'unable to download video info webpage', query={ + 'video_id': video_id, + 'eurl': 'https://youtube.googleapis.com/v/' + video_id, + 'html5': 1, + # See https://github.com/ytdl-org/youtube-dl/issues/29333#issuecomment-864049544 + 'c': 'TVHTML5', + 'cver': '6.20180913', + }, fatal=False) + if video_info: + pr = self._parse_json( + try_get( + compat_parse_qs(video_info), + lambda x: x['player_response'][0], compat_str) or '{}', + video_id, fatal=False) + if pr and isinstance(pr, dict): + player_response = pr + + trailer_video_id = try_get( + playability_status, + lambda x: x['errorScreen']['playerLegacyDesktopYpcTrailerRenderer']['trailerVideoId'], + compat_str) + if trailer_video_id: + return self.url_result( + trailer_video_id, self.ie_key(), trailer_video_id) - proto = ( - 'http' if self._downloader.params.get('prefer_insecure', False) - else 'https') - - start_time = None - end_time = None - parsed_url = compat_urllib_parse_urlparse(url) - for component in [parsed_url.fragment, parsed_url.query]: - query = compat_parse_qs(component) - if start_time is None and 't' in query: - start_time = parse_duration(query['t'][0]) - if start_time is None and 'start' in query: - start_time = parse_duration(query['start'][0]) - if end_time is None and 'end' in query: - end_time = parse_duration(query['end'][0]) - - # Extract original video URL from URL with redirection, like age verification, using next_url parameter - mobj = re.search(self._NEXT_URL_RE, url) - if mobj: - url = proto + '://www.youtube.com/' + compat_urllib_parse_unquote(mobj.group(1)).lstrip('/') - video_id = self.extract_id(url) - - # Get video webpage - url = proto + '://www.youtube.com/watch?v=%s&gl=US&hl=en&has_verified=1&bpctr=9999999999' % video_id - video_webpage, urlh = self._download_webpage_handle(url, video_id) - - qs = compat_parse_qs(compat_urllib_parse_urlparse(urlh.geturl()).query) - video_id = qs.get('v', [None])[0] or video_id - - # Attempt to extract SWF player URL - mobj = re.search(r'swfConfig.*?"(https?:\\/\\/.*?watch.*?-.*?\.swf)"', video_webpage) - if mobj is not None: - player_url = re.sub(r'\\(.)', r'\1', mobj.group(1)) - else: - player_url = None - - dash_mpds = [] - - def add_dash_mpd(video_info): - dash_mpd = video_info.get('dashmpd') - if dash_mpd and dash_mpd[0] not in dash_mpds: - dash_mpds.append(dash_mpd[0]) - - def add_dash_mpd_pr(pl_response): - dash_mpd = url_or_none(try_get( - pl_response, lambda x: x['streamingData']['dashManifestUrl'], - compat_str)) - if dash_mpd and dash_mpd not in dash_mpds: - dash_mpds.append(dash_mpd) - - is_live = None - view_count = None - - def extract_view_count(v_info): - return int_or_none(try_get(v_info, lambda x: x['view_count'][0])) - - def extract_player_response(player_response, video_id): - pl_response = str_or_none(player_response) - if not pl_response: + def get_text(x): + if not x: return - pl_response = self._parse_json(pl_response, video_id, fatal=False) - if isinstance(pl_response, dict): - add_dash_mpd_pr(pl_response) - return pl_response - - player_response = {} - - # Get video info - video_info = {} - embed_webpage = None - ytplayer_config = None - - if re.search(r'["\']status["\']\s*:\s*["\']LOGIN_REQUIRED', video_webpage) is not None: - age_gate = True - # We simulate the access to the video from www.youtube.com/v/{video_id} - # this can be viewed without login into Youtube - url = proto + '://www.youtube.com/embed/%s' % video_id - embed_webpage = self._download_webpage(url, video_id, 'Downloading embed webpage') - data = compat_urllib_parse_urlencode({ - 'video_id': video_id, - 'eurl': 'https://youtube.googleapis.com/v/' + video_id, - 'sts': self._search_regex( - r'"sts"\s*:\s*(\d+)', embed_webpage, 'sts', default=''), - }) - video_info_url = proto + '://www.youtube.com/get_video_info?' + data - try: - video_info_webpage = self._download_webpage( - video_info_url, video_id, - note='Refetching age-gated info webpage', - errnote='unable to download video info webpage') - except ExtractorError: - video_info_webpage = None - if video_info_webpage: - video_info = compat_parse_qs(video_info_webpage) - pl_response = video_info.get('player_response', [None])[0] - player_response = extract_player_response(pl_response, video_id) - add_dash_mpd(video_info) - view_count = extract_view_count(video_info) - else: - age_gate = False - # Try looking directly into the video webpage - ytplayer_config = self._get_ytplayer_config(video_id, video_webpage) - if ytplayer_config: - args = ytplayer_config['args'] - if args.get('url_encoded_fmt_stream_map') or args.get('hlsvp'): - # Convert to the same format returned by compat_parse_qs - video_info = dict((k, [v]) for k, v in args.items()) - 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/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']) - if args.get('livestream') == '1' or args.get('live_playback') == 1: - is_live = True - if not player_response: - player_response = extract_player_response(args.get('player_response'), video_id) - if not video_info or self._downloader.params.get('youtube_include_dash_manifest', True): - add_dash_mpd_pr(player_response) - - if not video_info and not player_response: - player_response = extract_player_response( - self._search_regex( - (r'%s\s*%s' % (self._YT_INITIAL_PLAYER_RESPONSE_RE, self._YT_INITIAL_BOUNDARY_RE), - self._YT_INITIAL_PLAYER_RESPONSE_RE), video_webpage, - 'initial player response', default='{}'), - video_id) - - def extract_unavailable_message(): - messages = [] - for tag, kind in (('h1', 'message'), ('div', 'submessage')): - msg = self._html_search_regex( - r'(?s)<{tag}[^>]+id=["\']unavailable-{kind}["\'][^>]*>(.+?)</{tag}>'.format(tag=tag, kind=kind), - video_webpage, 'unavailable %s' % kind, default=None) - if msg: - messages.append(msg) - if messages: - return '\n'.join(messages) - - if not video_info and not player_response: - 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) - - if not isinstance(video_info, dict): - video_info = {} + text = x.get('simpleText') + if text and isinstance(text, compat_str): + return text + runs = x.get('runs') + if not isinstance(runs, list): + return + return ''.join([r['text'] for r in runs if isinstance(r.get('text'), compat_str)]) - video_details = try_get( - player_response, lambda x: x['videoDetails'], dict) or {} + search_meta = ( + lambda x: self._html_search_meta(x, webpage, default=None)) \ + if webpage else lambda x: None + video_details = player_response.get('videoDetails') or {} microformat = try_get( - player_response, lambda x: x['microformat']['playerMicroformatRenderer'], dict) or {} - - 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_original = video_description = get_element_by_id("eow-description", video_webpage) - if video_description: - - def replace_url(m): - redir_url = compat_urlparse.urljoin(url, m.group(1)) - parsed_redir_url = compat_urllib_parse_urlparse(redir_url) - if re.search(r'^(?:www\.)?(?:youtube(?:-nocookie)?\.com|youtu\.be)$', parsed_redir_url.netloc) and parsed_redir_url.path == '/redirect': - qs = compat_parse_qs(parsed_redir_url.query) - q = qs.get('q') - if q and q[0]: - return q[0] - return redir_url - - description_original = video_description = re.sub(r'''(?x) - <a\s+ - (?:[a-zA-Z-]+="[^"]*"\s+)*? - (?:title|href)="([^"]+)"\s+ - (?:[a-zA-Z-]+="[^"]*"\s+)*? - class="[^"]*"[^>]*> - [^<]+\.{3}\s* - </a> - ''', replace_url, video_description) - video_description = clean_html(video_description) - else: - video_description = video_details.get('shortDescription') - if video_description is None: - video_description = self._html_search_meta('description', video_webpage) + player_response, + lambda x: x['microformat']['playerMicroformatRenderer'], + dict) or {} + video_title = video_details.get('title') \ + or get_text(microformat.get('title')) \ + or search_meta(['og:title', 'twitter:title', 'title']) + video_description = video_details.get('shortDescription') if not smuggled_data.get('force_singlefeed', False): if not self._downloader.params.get('noplaylist'): multifeed_metadata_list = try_get( player_response, lambda x: x['multicamera']['playerLegacyMulticameraRenderer']['metadataList'], - compat_str) or try_get( - video_info, lambda x: x['multifeed_metadata_list'][0], compat_str) + compat_str) if multifeed_metadata_list: entries = [] feed_ids = [] @@ -1862,10 +1624,12 @@ class YoutubeIE(YoutubeBaseInfoExtractor): # Unquote should take place before split on comma (,) since textual # fields may contain comma as well (see # https://github.com/ytdl-org/youtube-dl/issues/8536) - feed_data = compat_parse_qs(compat_urllib_parse_unquote_plus(feed)) + feed_data = compat_parse_qs( + compat_urllib_parse_unquote_plus(feed)) def feed_entry(name): - return try_get(feed_data, lambda x: x[name][0], compat_str) + return try_get( + feed_data, lambda x: x[name][0], compat_str) feed_id = feed_entry('id') if not feed_id: @@ -1878,7 +1642,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor): '_type': 'url_transparent', 'ie_key': 'Youtube', 'url': smuggle_url( - '%s://www.youtube.com/watch?v=%s' % (proto, feed_data['id'][0]), + base_url + 'watch?v=' + feed_data['id'][0], {'force_singlefeed': True}), 'title': title, }) @@ -1886,631 +1650,416 @@ class YoutubeIE(YoutubeBaseInfoExtractor): self.to_screen( 'Downloading multifeed video (%s) - add --no-playlist to just download video %s' % (', '.join(feed_ids), video_id)) - return self.playlist_result(entries, video_id, video_title, video_description) + return self.playlist_result( + entries, video_id, video_title, video_description) else: self.to_screen('Downloading just video %s because of --no-playlist' % video_id) - if view_count is None: - view_count = extract_view_count(video_info) - if view_count is None and video_details: - view_count = int_or_none(video_details.get('viewCount')) - if view_count is None and microformat: - view_count = int_or_none(microformat.get('viewCount')) - - if is_live is None: - is_live = bool_or_none(video_details.get('isLive')) - - # 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/ytdl-org/youtube-dl/issues/359 for more information.', expected=True) - - def _extract_filesize(media_url): - return int_or_none(self._search_regex( - r'\bclen[=/](\d+)', media_url, 'filesize', default=None)) - - streaming_formats = try_get(player_response, lambda x: x['streamingData']['formats'], list) or [] - streaming_formats.extend(try_get(player_response, lambda x: x['streamingData']['adaptiveFormats'], list) or []) - - if 'conn' in video_info and video_info['conn'][0].startswith('rtmp'): - self.report_rtmp_download() - formats = [{ - 'format_id': '_rtmp', - 'protocol': 'rtmp', - 'url': video_info['conn'][0], - 'player_url': player_url, - }] - elif not is_live and (streaming_formats or 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/ytdl-org/youtube-dl/issues/343 for more information.', expected=True) - formats = [] - formats_spec = {} - fmt_list = video_info.get('fmt_list', [''])[0] - if fmt_list: - for fmt in fmt_list.split(','): - spec = fmt.split('/') - if len(spec) > 1: - width_height = spec[1].split('x') - if len(width_height) == 2: - formats_spec[spec[0]] = { - 'resolution': spec[1], - 'width': int_or_none(width_height[0]), - 'height': int_or_none(width_height[1]), - } - for fmt in streaming_formats: - itag = str_or_none(fmt.get('itag')) - if not itag: - continue - quality = fmt.get('quality') - quality_label = fmt.get('qualityLabel') or quality - formats_spec[itag] = { - 'asr': int_or_none(fmt.get('audioSampleRate')), - 'filesize': int_or_none(fmt.get('contentLength')), - 'format_note': quality_label, - 'fps': int_or_none(fmt.get('fps')), - 'height': int_or_none(fmt.get('height')), - # bitrate for itag 43 is always 2147483647 - 'tbr': float_or_none(fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) if itag != '43' else None, - 'width': int_or_none(fmt.get('width')), - } + formats = [] + itags = [] + itag_qualities = {} + player_url = None + q = qualities(['tiny', 'small', 'medium', 'large', 'hd720', 'hd1080', 'hd1440', 'hd2160', 'hd2880', 'highres']) + streaming_data = player_response.get('streamingData') or {} + streaming_formats = streaming_data.get('formats') or [] + streaming_formats.extend(streaming_data.get('adaptiveFormats') or []) + for fmt in streaming_formats: + if fmt.get('targetDurationSec') or fmt.get('drmFamilies'): + continue - for fmt in streaming_formats: - if fmt.get('drmFamilies') or fmt.get('drm_families'): - continue - url = url_or_none(fmt.get('url')) + itag = str_or_none(fmt.get('itag')) + quality = fmt.get('quality') + if itag and quality: + itag_qualities[itag] = quality + # FORMAT_STREAM_TYPE_OTF(otf=1) requires downloading the init fragment + # (adding `&sq=0` to the URL) and parsing emsg box to determine the + # number of fragment that would subsequently requested with (`&sq=N`) + if fmt.get('type') == 'FORMAT_STREAM_TYPE_OTF': + continue - if not url: - cipher = fmt.get('cipher') or fmt.get('signatureCipher') - if not cipher: + fmt_url = fmt.get('url') + if not fmt_url: + sc = compat_parse_qs(fmt.get('signatureCipher')) + fmt_url = url_or_none(try_get(sc, lambda x: x['url'][0])) + encrypted_sig = try_get(sc, lambda x: x['s'][0]) + if not (sc and fmt_url and encrypted_sig): + continue + if not player_url: + if not webpage: continue - url_data = compat_parse_qs(cipher) - url = url_or_none(try_get(url_data, lambda x: x['url'][0], compat_str)) - if not url: + player_url = self._search_regex( + r'"(?:PLAYER_JS_URL|jsUrl)"\s*:\s*"([^"]+)"', + webpage, 'player URL', fatal=False) + if not player_url: + continue + signature = self._decrypt_signature(sc['s'][0], video_id, player_url) + sp = try_get(sc, lambda x: x['sp'][0]) or 'signature' + fmt_url += '&' + sp + '=' + signature + + if itag: + itags.append(itag) + tbr = float_or_none( + fmt.get('averageBitrate') or fmt.get('bitrate'), 1000) + dct = { + 'asr': int_or_none(fmt.get('audioSampleRate')), + 'filesize': int_or_none(fmt.get('contentLength')), + 'format_id': itag, + 'format_note': fmt.get('qualityLabel') or quality, + 'fps': int_or_none(fmt.get('fps')), + 'height': int_or_none(fmt.get('height')), + 'quality': q(quality), + 'tbr': tbr, + 'url': fmt_url, + 'width': fmt.get('width'), + } + mimetype = fmt.get('mimeType') + if mimetype: + mobj = re.match( + r'((?:[^/]+)/(?:[^;]+))(?:;\s*codecs="([^"]+)")?', mimetype) + if mobj: + dct['ext'] = mimetype2ext(mobj.group(1)) + dct.update(parse_codecs(mobj.group(2))) + no_audio = dct.get('acodec') == 'none' + no_video = dct.get('vcodec') == 'none' + if no_audio: + dct['vbr'] = tbr + if no_video: + dct['abr'] = tbr + if no_audio or no_video: + dct['downloader_options'] = { + # Youtube throttles chunks >~10M + 'http_chunk_size': 10485760, + } + if dct.get('ext'): + dct['container'] = dct['ext'] + '_dash' + formats.append(dct) + + hls_manifest_url = streaming_data.get('hlsManifestUrl') + if hls_manifest_url: + for f in self._extract_m3u8_formats( + hls_manifest_url, video_id, 'mp4', fatal=False): + itag = self._search_regex( + r'/itag/(\d+)', f['url'], 'itag', default=None) + if itag: + f['format_id'] = itag + formats.append(f) + + if self._downloader.params.get('youtube_include_dash_manifest', True): + dash_manifest_url = streaming_data.get('dashManifestUrl') + if dash_manifest_url: + for f in self._extract_mpd_formats( + dash_manifest_url, video_id, fatal=False): + itag = f['format_id'] + if itag in itags: continue - else: - cipher = None - url_data = compat_parse_qs(compat_urllib_parse_urlparse(url).query) + if itag in itag_qualities: + f['quality'] = q(itag_qualities[itag]) + filesize = int_or_none(self._search_regex( + r'/clen/(\d+)', f.get('fragment_base_url') + or f['url'], 'file size', default=None)) + if filesize: + f['filesize'] = filesize + formats.append(f) - stream_type = int_or_none(try_get(url_data, lambda x: x['stream_type'][0])) - # Unsupported FORMAT_STREAM_TYPE_OTF - if stream_type == 3: - continue + if not formats: + if streaming_data.get('licenseInfos'): + raise ExtractorError( + 'This video is DRM protected.', expected=True) + pemr = try_get( + playability_status, + lambda x: x['errorScreen']['playerErrorMessageRenderer'], + dict) or {} + reason = get_text(pemr.get('reason')) or playability_status.get('reason') + subreason = pemr.get('subreason') + if subreason: + subreason = clean_html(get_text(subreason)) + if subreason == 'The uploader has not made this video available in your country.': + countries = microformat.get('availableCountries') + if not countries: + regions_allowed = search_meta('regionsAllowed') + countries = regions_allowed.split(',') if regions_allowed else None + self.raise_geo_restricted( + subreason, countries) + reason += '\n' + subreason + if reason: + raise ExtractorError(reason, expected=True) - format_id = fmt.get('itag') or url_data['itag'][0] - if not format_id: - continue - format_id = compat_str(format_id) - - if cipher: - if 's' in url_data or self._downloader.params.get('youtube_include_dash_manifest', True): - ASSETS_RE = ( - r'<script[^>]+\bsrc=("[^"]+")[^>]+\bname=["\']player_ias/base', - r'"jsUrl"\s*:\s*("[^"]+")', - r'"assets":.+?"js":\s*("[^"]+")') - jsplayer_url_json = self._search_regex( - ASSETS_RE, - embed_webpage if age_gate else video_webpage, - 'JS player URL (1)', default=None) - if not jsplayer_url_json and not age_gate: - # We need the embed website after all - if embed_webpage is None: - embed_url = proto + '://www.youtube.com/embed/%s' % video_id - embed_webpage = self._download_webpage( - embed_url, video_id, 'Downloading embed webpage') - jsplayer_url_json = self._search_regex( - ASSETS_RE, embed_webpage, 'JS player URL') - - player_url = json.loads(jsplayer_url_json) - if player_url is None: - player_url_json = self._search_regex( - r'ytplayer\.config.*?"url"\s*:\s*("[^"]+")', - video_webpage, 'age gate player URL') - player_url = json.loads(player_url_json) - - if 'sig' in url_data: - url += '&signature=' + url_data['sig'][0] - elif 's' in url_data: - encrypted_sig = url_data['s'][0] - - if self._downloader.params.get('verbose'): - if player_url is None: - player_desc = 'unknown' - else: - player_type, player_version = self._extract_player_info(player_url) - player_desc = '%s player %s' % ('flash' if player_type == 'swf' else 'html5', player_version) - parts_sizes = self._signature_cache_id(encrypted_sig) - self.to_screen('{%s} signature length %s, %s' % - (format_id, parts_sizes, player_desc)) - - signature = self._decrypt_signature( - encrypted_sig, video_id, player_url, age_gate) - 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' - - dct = { - 'format_id': format_id, - 'url': url, - 'player_url': player_url, - } - if format_id in self._formats: - dct.update(self._formats[format_id]) - if format_id in formats_spec: - dct.update(formats_spec[format_id]) - - # Some itags are not included in DASH manifest thus corresponding formats will - # 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+)$', url_data.get('size', [''])[0]) - width, height = (int(mobj.group('width')), int(mobj.group('height'))) if mobj else (None, None) - - if width is None: - width = int_or_none(fmt.get('width')) - if height is None: - height = int_or_none(fmt.get('height')) - - filesize = int_or_none(url_data.get( - 'clen', [None])[0]) or _extract_filesize(url) - - quality = url_data.get('quality', [None])[0] or fmt.get('quality') - quality_label = url_data.get('quality_label', [None])[0] or fmt.get('qualityLabel') - - tbr = (float_or_none(url_data.get('bitrate', [None])[0], 1000) - or float_or_none(fmt.get('bitrate'), 1000)) if format_id != '43' else None - fps = int_or_none(url_data.get('fps', [None])[0]) or int_or_none(fmt.get('fps')) - - more_fields = { - 'filesize': filesize, - 'tbr': tbr, - 'width': width, - 'height': height, - 'fps': fps, - 'format_note': quality_label or quality, - } - for key, value in more_fields.items(): - if value: - dct[key] = value - type_ = url_data.get('type', [None])[0] or fmt.get('mimeType') - if type_: - type_split = type_.split(';') - kind_ext = type_split[0].split('/') - if len(kind_ext) == 2: - kind, _ = kind_ext - dct['ext'] = mimetype2ext(type_split[0]) - if kind in ('audio', 'video'): - codecs = None - for mobj in re.finditer( - r'(?P<key>[a-zA-Z_-]+)=(?P<quote>["\']?)(?P<val>.+?)(?P=quote)(?:;|$)', type_): - if mobj.group('key') == 'codecs': - codecs = mobj.group('val') - break - if codecs: - dct.update(parse_codecs(codecs)) - if dct.get('acodec') == 'none' or dct.get('vcodec') == 'none': - dct['downloader_options'] = { - # Youtube throttles chunks >~10M - 'http_chunk_size': 10485760, - } - formats.append(dct) - else: - manifest_url = ( - url_or_none(try_get( - player_response, - lambda x: x['streamingData']['hlsManifestUrl'], - compat_str)) - or url_or_none(try_get( - video_info, lambda x: x['hlsvp'][0], compat_str))) - if manifest_url: - formats = [] - m3u8_formats = self._extract_m3u8_formats( - manifest_url, video_id, 'mp4', fatal=False) - for a_format in m3u8_formats: - itag = self._search_regex( - r'/itag/(\d+)/', a_format['url'], 'itag', default=None) - if itag: - a_format['format_id'] = itag - if itag in self._formats: - dct = self._formats[itag].copy() - dct.update(a_format) - a_format = dct - a_format['player_url'] = player_url - # Accept-Encoding header causes failures in live streams on Youtube and Youtube Gaming - a_format.setdefault('http_headers', {})['Youtubedl-no-compression'] = 'True' - formats.append(a_format) - else: - error_message = extract_unavailable_message() - if not error_message: - reason_list = try_get( - player_response, - lambda x: x['playabilityStatus']['errorScreen']['playerErrorMessageRenderer']['subreason']['runs'], - list) or [] - for reason in reason_list: - if not isinstance(reason, dict): - continue - reason_text = try_get(reason, lambda x: x['text'], compat_str) - if reason_text: - if not error_message: - error_message = '' - error_message += reason_text - if error_message: - error_message = clean_html(error_message) - if not error_message: - error_message = clean_html(try_get( - player_response, lambda x: x['playabilityStatus']['reason'], - compat_str)) - if not error_message: - error_message = clean_html( - try_get(video_info, lambda x: x['reason'][0], compat_str)) - if error_message: - raise ExtractorError(error_message, expected=True) - raise ExtractorError('no conn, hlsvp, hlsManifestUrl or url_encoded_fmt_stream_map information found in video info') - - # uploader - video_uploader = try_get( - video_info, lambda x: x['author'][0], - compat_str) or str_or_none(video_details.get('author')) - if video_uploader: - video_uploader = compat_urllib_parse_unquote_plus(video_uploader) - else: - self._downloader.report_warning('unable to extract uploader name') - - # uploader_id - video_uploader_id = None - video_uploader_url = None - mobj = re.search( - r'<link itemprop="url" href="(?P<uploader_url>https?://www\.youtube\.com/(?:user|channel)/(?P<uploader_id>[^"]+))">', - video_webpage) - if mobj is not None: - video_uploader_id = mobj.group('uploader_id') - video_uploader_url = mobj.group('uploader_url') - else: - owner_profile_url = url_or_none(microformat.get('ownerProfileUrl')) - if owner_profile_url: - video_uploader_id = self._search_regex( - r'(?:user|channel)/([^/]+)', owner_profile_url, 'uploader id', - default=None) - video_uploader_url = owner_profile_url - - 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 + self._sort_formats(formats) - thumbnails = [] - thumbnails_list = try_get( - video_details, lambda x: x['thumbnail']['thumbnails'], list) or [] - for t in thumbnails_list: - if not isinstance(t, dict): - continue - thumbnail_url = url_or_none(t.get('url')) - if not thumbnail_url: - continue - thumbnails.append({ - 'url': thumbnail_url, - 'width': int_or_none(t.get('width')), - 'height': int_or_none(t.get('height')), - }) + keywords = video_details.get('keywords') or [] + if not keywords and webpage: + keywords = [ + unescapeHTML(m.group('content')) + for m in re.finditer(self._meta_regex('og:video:tag'), webpage)] + for keyword in keywords: + if keyword.startswith('yt:stretch='): + mobj = re.search(r'(\d+)\s*:\s*(\d+)', keyword) + if mobj: + # NB: float is intentional for forcing float division + w, h = (float(v) for v in mobj.groups()) + if w > 0 and h > 0: + ratio = w / h + for f in formats: + if f.get('vcodec') != 'none': + f['stretched_ratio'] = ratio + break - if not thumbnails: - video_thumbnail = None - # We try first to get a high quality image: - m_thumb = re.search(r'<span itemprop="thumbnail".*?href="(.*?)">', - video_webpage, re.DOTALL) - if m_thumb is not None: - video_thumbnail = m_thumb.group(1) - thumbnail_url = try_get(video_info, lambda x: x['thumbnail_url'][0], compat_str) - if thumbnail_url: - video_thumbnail = compat_urllib_parse_unquote_plus(thumbnail_url) - if video_thumbnail: - thumbnails.append({'url': video_thumbnail}) - - # upload date - upload_date = self._html_search_meta( - 'datePublished', video_webpage, 'upload date', default=None) - if not upload_date: - upload_date = self._search_regex( - [r'(?s)id="eow-date.*?>(.*?)</span>', - r'(?:id="watch-uploader-info".*?>.*?|["\']simpleText["\']\s*:\s*["\'])(?:Published|Uploaded|Streamed live|Started) on (.+?)[<"\']'], - video_webpage, 'upload date', default=None) - if not upload_date: - upload_date = microformat.get('publishDate') or microformat.get('uploadDate') - upload_date = unified_strdate(upload_date) - - video_license = self._html_search_regex( - r'<h4[^>]+class="title"[^>]*>\s*License\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li', - video_webpage, 'license', default=None) - - m_music = re.search( - r'''(?x) - <h4[^>]+class="title"[^>]*>\s*Music\s*</h4>\s* - <ul[^>]*>\s* - <li>(?P<title>.+?) - by (?P<creator>.+?) - (?: - \(.+?\)| - <a[^>]* - (?: - \bhref=["\']/red[^>]*>| # drop possible - >\s*Listen ad-free with YouTube Red # YouTube Red ad - ) - .*? - )?</li - ''', - video_webpage) - if m_music: - video_alt_title = remove_quotes(unescapeHTML(m_music.group('title'))) - video_creator = clean_html(m_music.group('creator')) + thumbnails = [] + for container in (video_details, microformat): + for thumbnail in (try_get( + container, + lambda x: x['thumbnail']['thumbnails'], list) or []): + thumbnail_url = thumbnail.get('url') + if not thumbnail_url: + continue + thumbnails.append({ + 'height': int_or_none(thumbnail.get('height')), + 'url': thumbnail_url, + 'width': int_or_none(thumbnail.get('width')), + }) + if thumbnails: + break else: - video_alt_title = video_creator = None + thumbnail = search_meta(['og:image', 'twitter:image']) + if thumbnail: + thumbnails = [{'url': thumbnail}] + + category = microformat.get('category') or search_meta('genre') + channel_id = video_details.get('channelId') \ + or microformat.get('externalChannelId') \ + or search_meta('channelId') + duration = int_or_none( + video_details.get('lengthSeconds') + or microformat.get('lengthSeconds')) \ + or parse_duration(search_meta('duration')) + is_live = video_details.get('isLive') + owner_profile_url = microformat.get('ownerProfileUrl') + + info = { + 'id': video_id, + 'title': self._live_title(video_title) if is_live else video_title, + 'formats': formats, + 'thumbnails': thumbnails, + 'description': video_description, + 'upload_date': unified_strdate( + microformat.get('uploadDate') + or search_meta('uploadDate')), + 'uploader': video_details['author'], + 'uploader_id': self._search_regex(r'/(?:channel|user)/([^/?&#]+)', owner_profile_url, 'uploader id') if owner_profile_url else None, + 'uploader_url': owner_profile_url, + 'channel_id': channel_id, + 'channel_url': 'https://www.youtube.com/channel/' + channel_id if channel_id else None, + 'duration': duration, + 'view_count': int_or_none( + video_details.get('viewCount') + or microformat.get('viewCount') + or search_meta('interactionCount')), + 'average_rating': float_or_none(video_details.get('averageRating')), + 'age_limit': 18 if ( + microformat.get('isFamilySafe') is False + or search_meta('isFamilyFriendly') == 'false' + or search_meta('og:restrictions:age') == '18+') else 0, + 'webpage_url': webpage_url, + 'categories': [category] if category else None, + 'tags': keywords, + 'is_live': is_live, + } - def extract_meta(field): - return self._html_search_regex( - r'<h4[^>]+class="title"[^>]*>\s*%s\s*</h4>\s*<ul[^>]*>\s*<li>(.+?)</li>\s*' % field, - video_webpage, field, default=None) + pctr = try_get( + player_response, + lambda x: x['captions']['playerCaptionsTracklistRenderer'], dict) + if pctr: + def process_language(container, base_url, lang_code, query): + lang_subs = [] + for fmt in self._SUBTITLE_FORMATS: + query.update({ + 'fmt': fmt, + }) + lang_subs.append({ + 'ext': fmt, + 'url': update_url_query(base_url, query), + }) + container[lang_code] = lang_subs + + subtitles = {} + for caption_track in (pctr.get('captionTracks') or []): + base_url = caption_track.get('baseUrl') + if not base_url: + continue + if caption_track.get('kind') != 'asr': + lang_code = caption_track.get('languageCode') + if not lang_code: + continue + process_language( + subtitles, base_url, lang_code, {}) + continue + automatic_captions = {} + for translation_language in (pctr.get('translationLanguages') or []): + translation_language_code = translation_language.get('languageCode') + if not translation_language_code: + continue + process_language( + automatic_captions, base_url, translation_language_code, + {'tlang': translation_language_code}) + info['automatic_captions'] = automatic_captions + info['subtitles'] = subtitles - track = extract_meta('Song') - artist = extract_meta('Artist') - album = extract_meta('Album') + parsed_url = compat_urllib_parse_urlparse(url) + for component in [parsed_url.fragment, parsed_url.query]: + query = compat_parse_qs(component) + for k, v in query.items(): + for d_k, s_ks in [('start', ('start', 't')), ('end', ('end',))]: + d_k += '_time' + if d_k not in info and k in s_ks: + info[d_k] = parse_duration(query[k][0]) - # Youtube Music Auto-generated description - release_date = release_year = None if video_description: mobj = re.search(r'(?s)(?P<track>[^·\n]+)·(?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]+))?.+\nAuto-generated by YouTube\.\s*$', 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) + release_year = release_date[:4] + info.update({ + 'album': mobj.group('album'.strip()), + 'artist': mobj.group('clean_artist') or ', '.join(a.strip() for a in mobj.group('artist').split('·')), + 'track': mobj.group('track').strip(), + 'release_date': release_date, + 'release_year': int_or_none(release_year), + }) - yt_initial_data = self._extract_yt_initial_data(video_id, video_webpage) - contents = try_get(yt_initial_data, lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], list) or [] - for content in contents: - rows = try_get(content, lambda x: x['videoSecondaryInfoRenderer']['metadataRowContainer']['metadataRowContainerRenderer']['rows'], list) or [] - multiple_songs = False - for row in rows: - if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True: - multiple_songs = True - break - for row in rows: - mrr = row.get('metadataRowRenderer') or {} - mrr_title = try_get( - mrr, lambda x: x['title']['simpleText'], compat_str) - mrr_contents = try_get( - mrr, lambda x: x['contents'][0], dict) or {} - mrr_contents_text = try_get(mrr_contents, [lambda x: x['simpleText'], lambda x: x['runs'][0]['text']], compat_str) - if not (mrr_title and mrr_contents_text): - continue - if mrr_title == 'License': - video_license = mrr_contents_text - elif not multiple_songs: - if mrr_title == 'Album': - album = mrr_contents_text - elif mrr_title == 'Artist': - artist = mrr_contents_text - elif mrr_title == 'Song': - track = mrr_contents_text - - 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>', - video_webpage) - if m_episode: - series = unescapeHTML(m_episode.group('series')) - season_number = int(m_episode.group('season')) - episode_number = int(m_episode.group('episode')) - else: - series = season_number = episode_number = None - - m_cat_container = self._search_regex( - r'(?s)<h4[^>]*>\s*Category\s*</h4>\s*<ul[^>]*>(.*?)</ul>', - video_webpage, 'categories', default=None) - category = None - if m_cat_container: - category = self._html_search_regex( - r'(?s)<a[^<]+>(.*?)</a>', m_cat_container, 'category', - default=None) - if not category: - category = try_get( - microformat, lambda x: x['category'], compat_str) - video_categories = None if category is None else [category] - - video_tags = [ - unescapeHTML(m.group('content')) - for m in re.finditer(self._meta_regex('og:video:tag'), video_webpage)] - if not video_tags: - video_tags = try_get(video_details, lambda x: x['keywords'], list) - - def _extract_count(count_name): - return str_to_int(self._search_regex( - (r'-%s-button[^>]+><span[^>]+class="yt-uix-button-content"[^>]*>([\d,]+)</span>' % re.escape(count_name), - r'["\']label["\']\s*:\s*["\']([\d,.]+)\s+%ss["\']' % re.escape(count_name)), - video_webpage, count_name, default=None)) - - like_count = _extract_count('like') - dislike_count = _extract_count('dislike') - - if view_count is None: - view_count = str_to_int(self._search_regex( - 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, player_response, ytplayer_config) - - video_duration = try_get( - video_info, lambda x: int_or_none(x['length_seconds'][0])) - if not video_duration: - video_duration = int_or_none(video_details.get('lengthSeconds')) - if not video_duration: - video_duration = parse_duration(self._html_search_meta( - 'duration', video_webpage, 'video duration')) - - # annotations - video_annotations = None - if self._downloader.params.get('writeannotations', False): - xsrf_token = None - ytcfg = self._extract_ytcfg(video_id, video_webpage) - if ytcfg: - xsrf_token = try_get(ytcfg, lambda x: x['XSRF_TOKEN'], compat_str) - if not xsrf_token: - xsrf_token = self._search_regex( - r'([\'"])XSRF_TOKEN\1\s*:\s*([\'"])(?P<xsrf_token>(?:(?!\2).)+)\2', - video_webpage, 'xsrf token', group='xsrf_token', fatal=False) - invideo_url = try_get( - player_response, lambda x: x['annotations'][0]['playerAnnotationsUrlsRenderer']['invideoUrl'], compat_str) - if xsrf_token and invideo_url: - xsrf_field_name = None - if ytcfg: - xsrf_field_name = try_get(ytcfg, lambda x: x['XSRF_FIELD_NAME'], compat_str) - if not xsrf_field_name: - xsrf_field_name = self._search_regex( - r'([\'"])XSRF_FIELD_NAME\1\s*:\s*([\'"])(?P<xsrf_field_name>\w+)\2', - video_webpage, 'xsrf field name', - group='xsrf_field_name', default='session_token') - video_annotations = self._download_webpage( - self._proto_relative_url(invideo_url), - video_id, note='Downloading annotations', - errnote='Unable to download video annotations', fatal=False, - data=urlencode_postdata({xsrf_field_name: xsrf_token})) - - chapters = self._extract_chapters(video_webpage, description_original, video_id, video_duration) - - # Look for the DASH manifest - if self._downloader.params.get('youtube_include_dash_manifest', True): - dash_mpd_fatal = True - for mpd_url in dash_mpds: - dash_formats = {} - try: - def decrypt_sig(mobj): - s = mobj.group(1) - dec_s = self._decrypt_signature(s, video_id, player_url, age_gate) - return '/signature/%s' % dec_s - - mpd_url = re.sub(r'/s/([a-fA-F0-9\.]+)', decrypt_sig, mpd_url) - - for df in self._extract_mpd_formats( - mpd_url, video_id, fatal=dash_mpd_fatal, - formats_dict=self._formats): - if not df.get('filesize'): - df['filesize'] = _extract_filesize(df['url']) - # Do not overwrite DASH format found in some previous DASH manifest - if df['format_id'] not in dash_formats: - dash_formats[df['format_id']] = df - # Additional DASH manifests may end up in HTTP Error 403 therefore - # allow them to fail without bug report message if we already have - # some DASH manifest succeeded. This is temporary workaround to reduce - # burst of bug reports until we figure out the reason and whether it - # can be fixed at all. - dash_mpd_fatal = False - except (ExtractorError, KeyError) as e: - self.report_warning( - 'Skipping DASH manifest: %r' % e, video_id) - if dash_formats: - # 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/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()) - - # Check for malformed aspect ratio - stretched_m = re.search( - r'<meta\s+property="og:video:tag".*?content="yt:stretch=(?P<w>[0-9]+):(?P<h>[0-9]+)">', - video_webpage) - if stretched_m: - w = float(stretched_m.group('w')) - h = float(stretched_m.group('h')) - # yt:stretch may hold invalid ratio data (e.g. for Q39EVAstoRM ratio is 17:0). - # We will only process correct ratios. - if w > 0 and h > 0: - ratio = w / h - for f in formats: - if f.get('vcodec') != 'none': - f['stretched_ratio'] = ratio + initial_data = None + if webpage: + initial_data = self._extract_yt_initial_variable( + webpage, self._YT_INITIAL_DATA_RE, video_id, + 'yt initial data') + if not initial_data: + initial_data = self._call_api( + 'next', {'videoId': video_id}, video_id, fatal=False) + + if initial_data: + chapters = self._extract_chapters_from_json( + initial_data, video_id, duration) + if not chapters: + for engagment_pannel in (initial_data.get('engagementPanels') or []): + contents = try_get( + engagment_pannel, lambda x: x['engagementPanelSectionListRenderer']['content']['macroMarkersListRenderer']['contents'], + list) + if not contents: + continue - if not formats: - 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) - if video_info.get('license_info') or try_get(player_response, lambda x: x['streamingData']['licenseInfos']): - raise ExtractorError('This video is DRM protected.', expected=True) + def chapter_time(mmlir): + return parse_duration( + get_text(mmlir.get('timeDescription'))) + + chapters = [] + for next_num, content in enumerate(contents, start=1): + mmlir = content.get('macroMarkersListItemRenderer') or {} + start_time = chapter_time(mmlir) + end_time = chapter_time(try_get( + contents, lambda x: x[next_num]['macroMarkersListItemRenderer'])) \ + if next_num < len(contents) else duration + if start_time is None or end_time is None: + continue + chapters.append({ + 'start_time': start_time, + 'end_time': end_time, + 'title': get_text(mmlir.get('title')), + }) + if chapters: + break + if chapters: + info['chapters'] = chapters + + contents = try_get( + initial_data, + lambda x: x['contents']['twoColumnWatchNextResults']['results']['results']['contents'], + list) or [] + for content in contents: + vpir = content.get('videoPrimaryInfoRenderer') + if vpir: + stl = vpir.get('superTitleLink') + if stl: + stl = get_text(stl) + if try_get( + vpir, + lambda x: x['superTitleIcon']['iconType']) == 'LOCATION_PIN': + info['location'] = stl + else: + mobj = re.search(r'(.+?)\s*S(\d+)\s*•\s*E(\d+)', stl) + if mobj: + info.update({ + 'series': mobj.group(1), + 'season_number': int(mobj.group(2)), + 'episode_number': int(mobj.group(3)), + }) + for tlb in (try_get( + vpir, + lambda x: x['videoActions']['menuRenderer']['topLevelButtons'], + list) or []): + tbr = tlb.get('toggleButtonRenderer') or {} + for getter, regex in [( + lambda x: x['defaultText']['accessibility']['accessibilityData'], + r'(?P<count>[\d,]+)\s*(?P<type>(?:dis)?like)'), ([ + lambda x: x['accessibility'], + lambda x: x['accessibilityData']['accessibilityData'], + ], r'(?P<type>(?:dis)?like) this video along with (?P<count>[\d,]+) other people')]: + label = (try_get(tbr, getter, dict) or {}).get('label') + if label: + mobj = re.match(regex, label) + if mobj: + info[mobj.group('type') + '_count'] = str_to_int(mobj.group('count')) + break + sbr_tooltip = try_get( + vpir, lambda x: x['sentimentBar']['sentimentBarRenderer']['tooltip']) + if sbr_tooltip: + like_count, dislike_count = sbr_tooltip.split(' / ') + info.update({ + 'like_count': str_to_int(like_count), + 'dislike_count': str_to_int(dislike_count), + }) + vsir = content.get('videoSecondaryInfoRenderer') + if vsir: + info['channel'] = get_text(try_get( + vsir, + lambda x: x['owner']['videoOwnerRenderer']['title'], + dict)) + rows = try_get( + vsir, + lambda x: x['metadataRowContainer']['metadataRowContainerRenderer']['rows'], + list) or [] + multiple_songs = False + for row in rows: + if try_get(row, lambda x: x['metadataRowRenderer']['hasDividerLine']) is True: + multiple_songs = True + break + for row in rows: + mrr = row.get('metadataRowRenderer') or {} + mrr_title = mrr.get('title') + if not mrr_title: + continue + mrr_title = get_text(mrr['title']) + mrr_contents_text = get_text(mrr['contents'][0]) + if mrr_title == 'License': + info['license'] = mrr_contents_text + elif not multiple_songs: + if mrr_title == 'Album': + info['album'] = mrr_contents_text + elif mrr_title == 'Artist': + info['artist'] = mrr_contents_text + elif mrr_title == 'Song': + info['track'] = mrr_contents_text - self._sort_formats(formats) + for s_k, d_k in [('artist', 'creator'), ('track', 'alt_title')]: + v = info.get(s_k) + if v: + info[d_k] = v - self.mark_watched(video_id, video_info, player_response) + self.mark_watched(video_id, player_response) - return { - 'id': video_id, - 'uploader': video_uploader, - 'uploader_id': video_uploader_id, - 'uploader_url': video_uploader_url, - 'channel_id': channel_id, - 'channel_url': channel_url, - 'upload_date': upload_date, - 'license': video_license, - 'creator': video_creator or artist, - 'title': video_title, - 'alt_title': video_alt_title or track, - 'thumbnails': thumbnails, - 'description': video_description, - 'categories': video_categories, - 'tags': video_tags, - 'subtitles': video_subtitles, - 'automatic_captions': automatic_captions, - 'duration': video_duration, - 'age_limit': 18 if age_gate else 0, - 'annotations': video_annotations, - 'chapters': chapters, - 'webpage_url': proto + '://www.youtube.com/watch?v=%s' % video_id, - 'view_count': view_count, - 'like_count': like_count, - 'dislike_count': dislike_count, - 'average_rating': average_rating, - 'formats': formats, - 'is_live': is_live, - 'start_time': start_time, - 'end_time': end_time, - 'series': series, - 'season_number': season_number, - 'episode_number': episode_number, - 'track': track, - 'artist': artist, - 'album': album, - 'release_date': release_date, - 'release_year': release_year, - } + return info class YoutubeTabIE(YoutubeBaseInfoExtractor): @@ -2523,7 +2072,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): invidio\.us )/ (?: - (?:channel|c|user|feed)/| + (?:channel|c|user|feed|hashtag)/| (?:playlist|watch)\?.*?\blist=| (?!(?:watch|embed|v|e|results)\b) ) @@ -2550,6 +2099,15 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 'description': 'md5:be97ee0f14ee314f1f002cf187166ee2', }, }, { + # playlists, series + 'url': 'https://www.youtube.com/c/3blue1brown/playlists?view=50&sort=dd&shelf_id=3', + 'playlist_mincount': 5, + 'info_dict': { + 'id': 'UCYO_jab_esuFRV4b17AJtAw', + 'title': '3Blue1Brown - Playlists', + 'description': 'md5:e1384e8a133307dd10edee76e875d62f', + }, + }, { # playlists, singlepage 'url': 'https://www.youtube.com/user/ThirstForScience/playlists', 'playlist_mincount': 4, @@ -2809,6 +2367,16 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): }, { 'url': 'https://www.youtube.com/TheYoungTurks/live', 'only_matching': True, + }, { + 'url': 'https://www.youtube.com/hashtag/cctv9', + 'info_dict': { + 'id': 'cctv9', + 'title': '#cctv9', + }, + 'playlist_mincount': 350, + }, { + 'url': 'https://www.youtube.com/watch?list=PLW4dVinRY435CBE_JD3t-0SRXKfnZHS1P&feature=youtu.be&v=M9cJMXmQ_ZU', + 'only_matching': True, }] @classmethod @@ -2831,40 +2399,13 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): @staticmethod def _extract_grid_item_renderer(item): - for item_kind in ('Playlist', 'Video', 'Channel'): - renderer = item.get('grid%sRenderer' % item_kind) - if renderer: - return renderer - - def _extract_video(self, renderer): - video_id = renderer.get('videoId') - title = try_get( - renderer, - (lambda x: x['title']['runs'][0]['text'], - lambda x: x['title']['simpleText']), compat_str) - description = try_get( - renderer, lambda x: x['descriptionSnippet']['runs'][0]['text'], - compat_str) - duration = parse_duration(try_get( - renderer, lambda x: x['lengthText']['simpleText'], compat_str)) - view_count_text = try_get( - renderer, lambda x: x['viewCountText']['simpleText'], compat_str) or '' - view_count = str_to_int(self._search_regex( - r'^([\d,]+)', re.sub(r'\s', '', view_count_text), - 'view count', default=None)) - uploader = try_get( - renderer, lambda x: x['ownerText']['runs'][0]['text'], compat_str) - return { - '_type': 'url_transparent', - 'ie_key': YoutubeIE.ie_key(), - 'id': video_id, - 'url': video_id, - 'title': title, - 'description': description, - 'duration': duration, - 'view_count': view_count, - 'uploader': uploader, - } + assert isinstance(item, dict) + for key, renderer in item.items(): + if not key.startswith('grid') or not key.endswith('Renderer'): + continue + if not isinstance(renderer, dict): + continue + return renderer def _grid_entries(self, grid_renderer): for item in grid_renderer['items']: @@ -2874,7 +2415,8 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): if not isinstance(renderer, dict): continue title = try_get( - renderer, lambda x: x['title']['runs'][0]['text'], compat_str) + renderer, (lambda x: x['title']['runs'][0]['text'], + lambda x: x['title']['simpleText']), compat_str) # playlist playlist_id = renderer.get('playlistId') if playlist_id: @@ -2882,10 +2424,12 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): 'https://www.youtube.com/playlist?list=%s' % playlist_id, ie=YoutubeTabIE.ie_key(), video_id=playlist_id, video_title=title) + continue # video video_id = renderer.get('videoId') if video_id: yield self._extract_video(renderer) + continue # channel channel_id = renderer.get('channelId') if channel_id: @@ -2894,6 +2438,17 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): yield self.url_result( 'https://www.youtube.com/channel/%s' % channel_id, ie=YoutubeTabIE.ie_key(), video_title=title) + continue + # generic endpoint URL support + ep_url = urljoin('https://www.youtube.com/', try_get( + renderer, lambda x: x['navigationEndpoint']['commandMetadata']['webCommandMetadata']['url'], + compat_str)) + if ep_url: + for ie in (YoutubeTabIE, YoutubePlaylistIE, YoutubeIE): + if ie.suitable(ep_url): + yield self.url_result( + ep_url, ie=ie.ie_key(), video_id=ie._match_id(ep_url), video_title=title) + break def _shelf_entries_from_content(self, shelf_renderer): content = shelf_renderer.get('content') @@ -2986,6 +2541,14 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): for entry in self._post_thread_entries(renderer): yield entry + def _rich_grid_entries(self, contents): + for content in contents: + video_renderer = try_get(content, lambda x: x['richItemRenderer']['content']['videoRenderer'], dict) + if video_renderer: + entry = self._video_entry(video_renderer) + if entry: + yield entry + @staticmethod def _build_continuation_query(continuation, ctp=None): query = { @@ -3013,9 +2576,9 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): next_continuation = cls._extract_next_continuation_data(renderer) if next_continuation: return next_continuation - contents = renderer.get('contents') - if not isinstance(contents, list): - return + contents = [] + for key in ('contents', 'items'): + contents.extend(try_get(renderer, lambda x: x[key], list) or []) for content in contents: if not isinstance(content, dict): continue @@ -3031,82 +2594,111 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): ctp = continuation_ep.get('clickTrackingParams') return YoutubeTabIE._build_continuation_query(continuation, ctp) - def _entries(self, tab, identity_token): + def _entries(self, tab, item_id, webpage): tab_content = try_get(tab, lambda x: x['content'], dict) if not tab_content: return slr_renderer = try_get(tab_content, lambda x: x['sectionListRenderer'], dict) - if not slr_renderer: - return - is_channels_tab = tab.get('title') == 'Channels' - continuation = None - slr_contents = try_get(slr_renderer, lambda x: x['contents'], list) or [] - for slr_content in slr_contents: - if not isinstance(slr_content, dict): - continue - is_renderer = try_get(slr_content, lambda x: x['itemSectionRenderer'], dict) - if not is_renderer: - continue - isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or [] - for isr_content in isr_contents: - if not isinstance(isr_content, dict): - continue - renderer = isr_content.get('playlistVideoListRenderer') - if renderer: - for entry in self._playlist_entries(renderer): - yield entry - continuation = self._extract_continuation(renderer) - continue - renderer = isr_content.get('gridRenderer') - if renderer: - for entry in self._grid_entries(renderer): - yield entry - continuation = self._extract_continuation(renderer) + if slr_renderer: + is_channels_tab = tab.get('title') == 'Channels' + continuation = None + slr_contents = try_get(slr_renderer, lambda x: x['contents'], list) or [] + for slr_content in slr_contents: + if not isinstance(slr_content, dict): continue - renderer = isr_content.get('shelfRenderer') - if renderer: - for entry in self._shelf_entries(renderer, not is_channels_tab): - yield entry + is_renderer = try_get(slr_content, lambda x: x['itemSectionRenderer'], dict) + if not is_renderer: continue - renderer = isr_content.get('backstagePostThreadRenderer') - if renderer: - for entry in self._post_thread_entries(renderer): - yield entry - continuation = self._extract_continuation(renderer) - continue - renderer = isr_content.get('videoRenderer') - if renderer: - entry = self._video_entry(renderer) - if entry: - yield entry - + isr_contents = try_get(is_renderer, lambda x: x['contents'], list) or [] + for isr_content in isr_contents: + if not isinstance(isr_content, dict): + continue + renderer = isr_content.get('playlistVideoListRenderer') + if renderer: + for entry in self._playlist_entries(renderer): + yield entry + continuation = self._extract_continuation(renderer) + continue + renderer = isr_content.get('gridRenderer') + if renderer: + for entry in self._grid_entries(renderer): + yield entry + continuation = self._extract_continuation(renderer) + continue + renderer = isr_content.get('shelfRenderer') + if renderer: + for entry in self._shelf_entries(renderer, not is_channels_tab): + yield entry + continue + renderer = isr_content.get('backstagePostThreadRenderer') + if renderer: + for entry in self._post_thread_entries(renderer): + yield entry + continuation = self._extract_continuation(renderer) + continue + renderer = isr_content.get('videoRenderer') + if renderer: + entry = self._video_entry(renderer) + if entry: + yield entry + + if not continuation: + continuation = self._extract_continuation(is_renderer) if not continuation: - continuation = self._extract_continuation(is_renderer) + continuation = self._extract_continuation(slr_renderer) + else: + rich_grid_renderer = tab_content.get('richGridRenderer') + if not rich_grid_renderer: + return + for entry in self._rich_grid_entries(rich_grid_renderer.get('contents') or []): + yield entry + continuation = self._extract_continuation(rich_grid_renderer) - if not continuation: - continuation = self._extract_continuation(slr_renderer) + ytcfg = self._extract_ytcfg(item_id, webpage) + client_version = try_get( + ytcfg, lambda x: x['INNERTUBE_CLIENT_VERSION'], compat_str) or '2.20210407.08.00' headers = { 'x-youtube-client-name': '1', - 'x-youtube-client-version': '2.20201112.04.01', + 'x-youtube-client-version': client_version, + 'content-type': 'application/json', } + + context = try_get(ytcfg, lambda x: x['INNERTUBE_CONTEXT'], dict) or { + 'client': { + 'clientName': 'WEB', + 'clientVersion': client_version, + } + } + visitor_data = try_get(context, lambda x: x['client']['visitorData'], compat_str) + + identity_token = self._extract_identity_token(ytcfg, webpage) if identity_token: headers['x-youtube-identity-token'] = identity_token + data = { + 'context': context, + } + for page_num in itertools.count(1): if not continuation: break + if visitor_data: + headers['x-goog-visitor-id'] = visitor_data + data['continuation'] = continuation['continuation'] + data['clickTracking'] = { + 'clickTrackingParams': continuation['itct'] + } 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 - browse = self._download_json( - 'https://www.youtube.com/browse_ajax', None, - 'Downloading page %d%s' - % (page_num, ' (retry #%d)' % count if count else ''), - headers=headers, query=continuation) + response = self._download_json( + 'https://www.youtube.com/youtubei/v1/browse?key=AIzaSyAO_FJ2SlqU8Q4STEHLGCilw_Y9_11qcW8', + None, 'Downloading page %d%s' % (page_num, ' (retry #%d)' % count if count else ''), + headers=headers, data=json.dumps(data).encode('utf8')) break except ExtractorError as e: if isinstance(e.cause, compat_HTTPError) and e.cause.code in (500, 503): @@ -3114,12 +2706,12 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): if count <= retries: continue raise - if not browse: - break - response = try_get(browse, lambda x: x[1]['response'], dict) if not response: break + visitor_data = try_get( + response, lambda x: x['responseContext']['visitorData'], compat_str) or visitor_data + continuation_contents = try_get( response, lambda x: x['continuationContents'], dict) if continuation_contents: @@ -3142,12 +2734,20 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): continuation = self._extract_continuation(continuation_renderer) continue + on_response_received = dict_get(response, ('onResponseReceivedActions', 'onResponseReceivedEndpoints')) continuation_items = try_get( - response, lambda x: x['onResponseReceivedActions'][0]['appendContinuationItemsAction']['continuationItems'], list) + on_response_received, lambda x: x[0]['appendContinuationItemsAction']['continuationItems'], list) if continuation_items: continuation_item = continuation_items[0] if not isinstance(continuation_item, dict): continue + renderer = self._extract_grid_item_renderer(continuation_item) + if renderer: + grid_renderer = {'items': continuation_items} + for entry in self._grid_entries(grid_renderer): + yield entry + continuation = self._extract_continuation(grid_renderer) + continue renderer = continuation_item.get('playlistVideoRenderer') or continuation_item.get('itemSectionRenderer') if renderer: video_list_renderer = {'contents': continuation_items} @@ -3155,6 +2755,19 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): yield entry continuation = self._extract_continuation(video_list_renderer) continue + renderer = continuation_item.get('backstagePostThreadRenderer') + if renderer: + continuation_renderer = {'contents': continuation_items} + for entry in self._post_thread_continuation_entries(continuation_renderer): + yield entry + continuation = self._extract_continuation(continuation_renderer) + continue + renderer = continuation_item.get('richItemRenderer') + if renderer: + for entry in self._rich_grid_entries(continuation_items): + yield entry + continuation = self._extract_continuation({'contents': continuation_items}) + continue break @@ -3207,11 +2820,12 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): alerts.append(text) return '\n'.join(alerts) - def _extract_from_tabs(self, item_id, webpage, data, tabs, identity_token): + def _extract_from_tabs(self, item_id, webpage, data, tabs): selected_tab = self._extract_selected_tab(tabs) renderer = try_get( data, lambda x: x['metadata']['channelMetadataRenderer'], dict) - playlist_id = title = description = None + playlist_id = item_id + title = description = None if renderer: channel_title = renderer.get('title') or item_id tab_title = selected_tab.get('title') @@ -3220,14 +2834,18 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): title += ' - %s' % tab_title description = renderer.get('description') playlist_id = renderer.get('externalId') - renderer = try_get( - data, lambda x: x['metadata']['playlistMetadataRenderer'], dict) - if renderer: - title = renderer.get('title') - description = None - playlist_id = item_id + else: + renderer = try_get( + data, lambda x: x['metadata']['playlistMetadataRenderer'], dict) + if renderer: + title = renderer.get('title') + else: + renderer = try_get( + data, lambda x: x['header']['hashtagHeaderRenderer'], dict) + if renderer: + title = try_get(renderer, lambda x: x['hashtag']['simpleText']) playlist = self.playlist_result( - self._entries(selected_tab, identity_token), + self._entries(selected_tab, item_id, webpage), playlist_id=playlist_id, playlist_title=title, playlist_description=description) playlist.update(self._extract_uploader(data)) @@ -3251,8 +2869,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): self._playlist_entries(playlist), playlist_id=playlist_id, playlist_title=title) - def _extract_identity_token(self, webpage, item_id): - ytcfg = self._extract_ytcfg(item_id, webpage) + def _extract_identity_token(self, ytcfg, webpage): if ytcfg: token = try_get(ytcfg, lambda x: x['ID_TOKEN'], compat_str) if token: @@ -3266,7 +2883,7 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): url = compat_urlparse.urlunparse( compat_urlparse.urlparse(url)._replace(netloc='www.youtube.com')) # Handle both video/playlist URLs - qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + qs = parse_qs(url) video_id = qs.get('v', [None])[0] playlist_id = qs.get('list', [None])[0] if video_id and playlist_id: @@ -3275,12 +2892,11 @@ class YoutubeTabIE(YoutubeBaseInfoExtractor): return self.url_result(video_id, ie=YoutubeIE.ie_key(), video_id=video_id) self.to_screen('Downloading playlist %s - add --no-playlist to just download video %s' % (playlist_id, video_id)) webpage = self._download_webpage(url, item_id) - identity_token = self._extract_identity_token(webpage, item_id) data = self._extract_yt_initial_data(item_id, webpage) tabs = try_get( data, lambda x: x['contents']['twoColumnBrowseResultsRenderer']['tabs'], list) if tabs: - return self._extract_from_tabs(item_id, webpage, data, tabs, identity_token) + return self._extract_from_tabs(item_id, webpage, data, tabs) playlist = try_get( data, lambda x: x['contents']['twoColumnWatchNextResults']['playlist']['playlist'], dict) if playlist: @@ -3363,12 +2979,19 @@ class YoutubePlaylistIE(InfoExtractor): @classmethod def suitable(cls, url): - return False if YoutubeTabIE.suitable(url) else super( - YoutubePlaylistIE, cls).suitable(url) + if YoutubeTabIE.suitable(url): + return False + # Hack for lazy extractors until more generic solution is implemented + # (see #28780) + from .youtube import parse_qs + qs = parse_qs(url) + if qs.get('v', [None])[0]: + return False + return super(YoutubePlaylistIE, cls).suitable(url) def _real_extract(self, url): playlist_id = self._match_id(url) - qs = compat_urlparse.parse_qs(compat_urlparse.urlparse(url).query) + qs = parse_qs(url) if not qs: qs = {'list': playlist_id} return self.url_result( diff --git a/youtube_dl/extractor/zdf.py b/youtube_dl/extractor/zdf.py index 5ed2946c2..4dd56f66d 100644 --- a/youtube_dl/extractor/zdf.py +++ b/youtube_dl/extractor/zdf.py @@ -7,7 +7,9 @@ from .common import InfoExtractor from ..compat import compat_str from ..utils import ( determine_ext, + float_or_none, int_or_none, + merge_dicts, NO_DEFAULT, orderedSet, parse_codecs, @@ -21,49 +23,17 @@ from ..utils import ( class ZDFBaseIE(InfoExtractor): - def _call_api(self, url, player, referrer, video_id, item): - return self._download_json( - url, video_id, 'Downloading JSON %s' % item, - headers={ - 'Referer': referrer, - 'Api-Auth': 'Bearer %s' % player['apiToken'], - }) - - def _extract_player(self, webpage, video_id, fatal=True): - return self._parse_json( - self._search_regex( - r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage, - 'player JSON', default='{}' if not fatal else NO_DEFAULT, - group='json'), - video_id) - - -class ZDFIE(ZDFBaseIE): - _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?]+)\.html' - _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd') _GEO_COUNTRIES = ['DE'] + _QUALITIES = ('auto', 'low', 'med', 'high', 'veryhigh', 'hd') - _TESTS = [{ - 'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html', - 'info_dict': { - 'id': 'die-magie-der-farben-von-koenigspurpur-und-jeansblau-100', - 'ext': 'mp4', - 'title': 'Die Magie der Farben (2/2)', - 'description': 'md5:a89da10c928c6235401066b60a6d5c1a', - 'duration': 2615, - 'timestamp': 1465021200, - 'upload_date': '20160604', - }, - }, { - 'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html', - 'only_matching': True, - }, { - 'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html', - 'only_matching': True, - }, { - 'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html', - 'only_matching': True, - }] + def _call_api(self, url, video_id, item, api_token=None, referrer=None): + headers = {} + if api_token: + headers['Api-Auth'] = 'Bearer %s' % api_token + if referrer: + headers['Referer'] = referrer + return self._download_json( + url, video_id, 'Downloading JSON %s' % item, headers=headers) @staticmethod def _extract_subtitles(src): @@ -109,20 +79,11 @@ class ZDFIE(ZDFBaseIE): }) formats.append(f) - def _extract_entry(self, url, player, content, video_id): - title = content.get('title') or content['teaserHeadline'] - - t = content['mainVideoContent']['http://zdf.de/rels/target'] - - ptmd_path = t.get('http://zdf.de/rels/streams/ptmd') - - if not ptmd_path: - ptmd_path = t[ - 'http://zdf.de/rels/streams/ptmd-template'].replace( - '{playerId}', 'ngplayer_2_4') - + def _extract_ptmd(self, ptmd_url, video_id, api_token, referrer): ptmd = self._call_api( - urljoin(url, ptmd_path), player, url, video_id, 'metadata') + ptmd_url, video_id, 'metadata', api_token, referrer) + + content_id = ptmd.get('basename') or ptmd_url.split('/')[-1] formats = [] track_uris = set() @@ -140,7 +101,7 @@ class ZDFIE(ZDFBaseIE): continue for track in tracks: self._extract_format( - video_id, formats, track_uris, { + content_id, formats, track_uris, { 'url': track.get('uri'), 'type': f.get('type'), 'mimeType': f.get('mimeType'), @@ -149,6 +110,103 @@ class ZDFIE(ZDFBaseIE): }) self._sort_formats(formats) + duration = float_or_none(try_get( + ptmd, lambda x: x['attributes']['duration']['value']), scale=1000) + + return { + 'extractor_key': ZDFIE.ie_key(), + 'id': content_id, + 'duration': duration, + 'formats': formats, + 'subtitles': self._extract_subtitles(ptmd), + } + + def _extract_player(self, webpage, video_id, fatal=True): + return self._parse_json( + self._search_regex( + r'(?s)data-zdfplayer-jsb=(["\'])(?P<json>{.+?})\1', webpage, + 'player JSON', default='{}' if not fatal else NO_DEFAULT, + group='json'), + video_id) + + +class ZDFIE(ZDFBaseIE): + _VALID_URL = r'https?://www\.zdf\.de/(?:[^/]+/)*(?P<id>[^/?#&]+)\.html' + _TESTS = [{ + # Same as https://www.phoenix.de/sendungen/ereignisse/corona-nachgehakt/wohin-fuehrt-der-protest-in-der-pandemie-a-2050630.html + 'url': 'https://www.zdf.de/politik/phoenix-sendungen/wohin-fuehrt-der-protest-in-der-pandemie-100.html', + 'md5': '34ec321e7eb34231fd88616c65c92db0', + 'info_dict': { + 'id': '210222_phx_nachgehakt_corona_protest', + 'ext': 'mp4', + 'title': 'Wohin führt der Protest in der Pandemie?', + 'description': 'md5:7d643fe7f565e53a24aac036b2122fbd', + 'duration': 1691, + 'timestamp': 1613948400, + 'upload_date': '20210221', + }, + }, { + # Same as https://www.3sat.de/film/ab-18/10-wochen-sommer-108.html + 'url': 'https://www.zdf.de/dokumentation/ab-18/10-wochen-sommer-102.html', + 'md5': '0aff3e7bc72c8813f5e0fae333316a1d', + 'info_dict': { + 'id': '141007_ab18_10wochensommer_film', + 'ext': 'mp4', + 'title': 'Ab 18! - 10 Wochen Sommer', + 'description': 'md5:8253f41dc99ce2c3ff892dac2d65fe26', + 'duration': 2660, + 'timestamp': 1608604200, + 'upload_date': '20201222', + }, + }, { + 'url': 'https://www.zdf.de/dokumentation/terra-x/die-magie-der-farben-von-koenigspurpur-und-jeansblau-100.html', + 'info_dict': { + 'id': '151025_magie_farben2_tex', + 'ext': 'mp4', + 'title': 'Die Magie der Farben (2/2)', + 'description': 'md5:a89da10c928c6235401066b60a6d5c1a', + 'duration': 2615, + 'timestamp': 1465021200, + 'upload_date': '20160604', + }, + }, { + # Same as https://www.phoenix.de/sendungen/dokumentationen/gesten-der-maechtigen-i-a-89468.html?ref=suche + 'url': 'https://www.zdf.de/politik/phoenix-sendungen/die-gesten-der-maechtigen-100.html', + 'only_matching': True, + }, { + # Same as https://www.3sat.de/film/spielfilm/der-hauptmann-100.html + 'url': 'https://www.zdf.de/filme/filme-sonstige/der-hauptmann-112.html', + 'only_matching': True, + }, { + # Same as https://www.3sat.de/wissen/nano/nano-21-mai-2019-102.html, equal media ids + 'url': 'https://www.zdf.de/wissen/nano/nano-21-mai-2019-102.html', + 'only_matching': True, + }, { + 'url': 'https://www.zdf.de/service-und-hilfe/die-neue-zdf-mediathek/zdfmediathek-trailer-100.html', + 'only_matching': True, + }, { + 'url': 'https://www.zdf.de/filme/taunuskrimi/die-lebenden-und-die-toten-1---ein-taunuskrimi-100.html', + 'only_matching': True, + }, { + 'url': 'https://www.zdf.de/dokumentation/planet-e/planet-e-uebersichtsseite-weitere-dokumentationen-von-planet-e-100.html', + 'only_matching': True, + }] + + def _extract_entry(self, url, player, content, video_id): + title = content.get('title') or content['teaserHeadline'] + + t = content['mainVideoContent']['http://zdf.de/rels/target'] + + ptmd_path = t.get('http://zdf.de/rels/streams/ptmd') + + if not ptmd_path: + ptmd_path = t[ + 'http://zdf.de/rels/streams/ptmd-template'].replace( + '{playerId}', 'ngplayer_2_4') + + info = self._extract_ptmd( + urljoin(url, ptmd_path), video_id, player['apiToken'], url) + thumbnails = [] layouts = try_get( content, lambda x: x['teaserImageRef']['layouts'], dict) @@ -169,33 +227,33 @@ class ZDFIE(ZDFBaseIE): }) thumbnails.append(thumbnail) - return { - 'id': video_id, + return merge_dicts(info, { 'title': title, 'description': content.get('leadParagraph') or content.get('teasertext'), 'duration': int_or_none(t.get('duration')), 'timestamp': unified_timestamp(content.get('editorialDate')), 'thumbnails': thumbnails, - 'subtitles': self._extract_subtitles(ptmd), - 'formats': formats, - } + }) def _extract_regular(self, url, player, video_id): content = self._call_api( - player['content'], player, url, video_id, 'content') + player['content'], video_id, 'content', player['apiToken'], url) return self._extract_entry(player['content'], player, content, video_id) def _extract_mobile(self, video_id): - document = self._download_json( + video = self._download_json( 'https://zdf-cdn.live.cellular.de/mediathekV2/document/%s' % video_id, - video_id)['document'] + video_id) + + document = video['document'] title = document['titel'] + content_id = document['basename'] formats = [] format_urls = set() for f in document['formitaeten']: - self._extract_format(video_id, formats, format_urls, f) + self._extract_format(content_id, formats, format_urls, f) self._sort_formats(formats) thumbnails = [] @@ -213,12 +271,12 @@ class ZDFIE(ZDFBaseIE): }) return { - 'id': video_id, + 'id': content_id, 'title': title, 'description': document.get('beschreibung'), 'duration': int_or_none(document.get('length')), - 'timestamp': unified_timestamp(try_get( - document, lambda x: x['meta']['editorialDate'], compat_str)), + 'timestamp': unified_timestamp(document.get('date')) or unified_timestamp( + try_get(video, lambda x: x['meta']['editorialDate'], compat_str)), 'thumbnails': thumbnails, 'subtitles': self._extract_subtitles(document), 'formats': formats, diff --git a/youtube_dl/extractor/zhihu.py b/youtube_dl/extractor/zhihu.py new file mode 100644 index 000000000..d1ed55be3 --- /dev/null +++ b/youtube_dl/extractor/zhihu.py @@ -0,0 +1,69 @@ +# coding: utf-8 +from __future__ import unicode_literals + +from .common import InfoExtractor +from ..utils import float_or_none, int_or_none + + +class ZhihuIE(InfoExtractor): + _VALID_URL = r'https?://(?:www\.)?zhihu\.com/zvideo/(?P<id>[0-9]+)' + _TEST = { + 'url': 'https://www.zhihu.com/zvideo/1342930761977176064', + 'md5': 'c8d4c9cd72dd58e6f9bc9c2c84266464', + 'info_dict': { + 'id': '1342930761977176064', + 'ext': 'mp4', + 'title': '写春联也太难了吧!', + 'thumbnail': r're:^https?://.*\.jpg', + 'uploader': '桥半舫', + 'timestamp': 1612959715, + 'upload_date': '20210210', + 'uploader_id': '244ecb13b0fd7daf92235288c8ca3365', + 'duration': 146.333, + 'view_count': int, + 'like_count': int, + 'comment_count': int, + } + } + + def _real_extract(self, url): + video_id = self._match_id(url) + zvideo = self._download_json( + 'https://www.zhihu.com/api/v4/zvideos/' + video_id, video_id) + title = zvideo['title'] + video = zvideo.get('video') or {} + + formats = [] + for format_id, q in (video.get('playlist') or {}).items(): + play_url = q.get('url') or q.get('play_url') + if not play_url: + continue + formats.append({ + 'asr': int_or_none(q.get('sample_rate')), + 'filesize': int_or_none(q.get('size')), + 'format_id': format_id, + 'fps': int_or_none(q.get('fps')), + 'height': int_or_none(q.get('height')), + 'tbr': float_or_none(q.get('bitrate')), + 'url': play_url, + 'width': int_or_none(q.get('width')), + }) + self._sort_formats(formats) + + author = zvideo.get('author') or {} + url_token = author.get('url_token') + + return { + 'id': video_id, + 'title': title, + 'formats': formats, + 'thumbnail': video.get('thumbnail') or zvideo.get('image_url'), + 'uploader': author.get('name'), + 'timestamp': int_or_none(zvideo.get('published_at')), + 'uploader_id': author.get('id'), + 'uploader_url': 'https://www.zhihu.com/people/' + url_token if url_token else None, + 'duration': float_or_none(video.get('duration')), + 'view_count': int_or_none(zvideo.get('play_count')), + 'like_count': int_or_none(zvideo.get('liked_count')), + 'comment_count': int_or_none(zvideo.get('comment_count')), + } diff --git a/youtube_dl/extractor/zingmp3.py b/youtube_dl/extractor/zingmp3.py index adfdcaabf..207c04f5e 100644 --- a/youtube_dl/extractor/zingmp3.py +++ b/youtube_dl/extractor/zingmp3.py @@ -1,93 +1,94 @@ # coding: utf-8 from __future__ import unicode_literals -import re - from .common import InfoExtractor from ..utils import ( ExtractorError, int_or_none, - update_url_query, ) -class ZingMp3BaseInfoExtractor(InfoExtractor): +class ZingMp3BaseIE(InfoExtractor): + _VALID_URL_TMPL = r'https?://(?:mp3\.zing|zingmp3)\.vn/(?:%s)/[^/]+/(?P<id>\w+)\.html' + _GEO_COUNTRIES = ['VN'] - def _extract_item(self, item, page_type, fatal=True): - error_message = item.get('msg') - if error_message: - if not fatal: - return - raise ExtractorError( - '%s returned error: %s' % (self.IE_NAME, error_message), - expected=True) + def _extract_item(self, item, fatal): + item_id = item['id'] + title = item.get('name') or item['title'] formats = [] - for quality, source_url in zip(item.get('qualities') or item.get('quality', []), item.get('source_list') or item.get('source', [])): - if not source_url or source_url == 'require vip': + for k, v in (item.get('source') or {}).items(): + if not v: continue - if not re.match(r'https?://', source_url): - source_url = '//' + source_url - source_url = self._proto_relative_url(source_url, 'http:') - quality_num = int_or_none(quality) - f = { - 'format_id': quality, - 'url': source_url, - } - if page_type == 'video': - f.update({ - 'height': quality_num, - 'ext': 'mp4', - }) + if k in ('mp4', 'hls'): + for res, video_url in v.items(): + if not video_url: + continue + if k == 'hls': + formats.extend(self._extract_m3u8_formats( + video_url, item_id, 'mp4', + 'm3u8_native', m3u8_id=k, fatal=False)) + elif k == 'mp4': + formats.append({ + 'format_id': 'mp4-' + res, + 'url': video_url, + 'height': int_or_none(self._search_regex( + r'^(\d+)p', res, 'resolution', default=None)), + }) else: - f.update({ - 'abr': quality_num, + formats.append({ 'ext': 'mp3', + 'format_id': k, + 'tbr': int_or_none(k), + 'url': self._proto_relative_url(v), + 'vcodec': 'none', }) - formats.append(f) + if not formats: + if not fatal: + return + msg = item['msg'] + if msg == 'Sorry, this content is not available in your country.': + self.raise_geo_restricted(countries=self._GEO_COUNTRIES) + raise ExtractorError(msg, expected=True) + self._sort_formats(formats) + + subtitles = None + lyric = item.get('lyric') + if lyric: + subtitles = { + 'origin': [{ + 'url': lyric, + }], + } - cover = item.get('cover') + album = item.get('album') or {} return { - 'title': (item.get('name') or item.get('title')).strip(), + 'id': item_id, + 'title': title, 'formats': formats, - 'thumbnail': 'http:/' + cover if cover else None, - 'artist': item.get('artist'), + 'thumbnail': item.get('thumbnail'), + 'subtitles': subtitles, + 'duration': int_or_none(item.get('duration')), + 'track': title, + 'artist': item.get('artists_names'), + 'album': album.get('name') or album.get('title'), + 'album_artist': album.get('artists_names'), } - def _extract_player_json(self, player_json_url, id, page_type, playlist_title=None): - player_json = self._download_json(player_json_url, id, 'Downloading Player JSON') - items = player_json['data'] - if 'item' in items: - items = items['item'] - - if len(items) == 1: - # one single song - data = self._extract_item(items[0], page_type) - data['id'] = id - - return data - else: - # playlist of songs - entries = [] - - for i, item in enumerate(items, 1): - entry = self._extract_item(item, page_type, fatal=False) - if not entry: - continue - entry['id'] = '%s-%d' % (id, i) - entries.append(entry) - - return { - '_type': 'playlist', - 'id': id, - 'title': playlist_title, - 'entries': entries, - } + def _real_extract(self, url): + page_id = self._match_id(url) + webpage = self._download_webpage( + url.replace('://zingmp3.vn/', '://mp3.zing.vn/'), + page_id, query={'play_song': 1}) + data_path = self._search_regex( + r'data-xml="([^"]+)', webpage, 'data path') + return self._process_data(self._download_json( + 'https://mp3.zing.vn/xhr' + data_path, page_id)['data']) -class ZingMp3IE(ZingMp3BaseInfoExtractor): - _VALID_URL = r'https?://mp3\.zing\.vn/(?:bai-hat|album|playlist|video-clip)/[^/]+/(?P<id>\w+)\.html' +class ZingMp3IE(ZingMp3BaseIE): + _VALID_URL = ZingMp3BaseIE._VALID_URL_TMPL % 'bai-hat|video-clip' _TESTS = [{ 'url': 'http://mp3.zing.vn/bai-hat/Xa-Mai-Xa-Bao-Thy/ZWZB9WAB.html', 'md5': 'ead7ae13693b3205cbc89536a077daed', @@ -95,49 +96,66 @@ class ZingMp3IE(ZingMp3BaseInfoExtractor): 'id': 'ZWZB9WAB', 'title': 'Xa Mãi Xa', 'ext': 'mp3', - 'thumbnail': r're:^https?://.*\.jpg$', + 'thumbnail': r're:^https?://.+\.jpg', + 'subtitles': { + 'origin': [{ + 'ext': 'lrc', + }] + }, + 'duration': 255, + 'track': 'Xa Mãi Xa', + 'artist': 'Bảo Thy', + 'album': 'Special Album', + 'album_artist': 'Bảo Thy', }, }, { - 'url': 'http://mp3.zing.vn/video-clip/Let-It-Go-Frozen-OST-Sungha-Jung/ZW6BAEA0.html', - 'md5': '870295a9cd8045c0e15663565902618d', + 'url': 'https://mp3.zing.vn/video-clip/Suong-Hoa-Dua-Loi-K-ICM-RYO/ZO8ZF7C7.html', + 'md5': 'e9c972b693aa88301ef981c8151c4343', 'info_dict': { - 'id': 'ZW6BAEA0', - 'title': 'Let It Go (Frozen OST)', + 'id': 'ZO8ZF7C7', + 'title': 'Sương Hoa Đưa Lối', 'ext': 'mp4', + 'thumbnail': r're:^https?://.+\.jpg', + 'duration': 207, + 'track': 'Sương Hoa Đưa Lối', + 'artist': 'K-ICM, RYO', }, }, { + 'url': 'https://zingmp3.vn/bai-hat/Xa-Mai-Xa-Bao-Thy/ZWZB9WAB.html', + 'only_matching': True, + }] + IE_NAME = 'zingmp3' + IE_DESC = 'mp3.zing.vn' + + def _process_data(self, data): + return self._extract_item(data, True) + + +class ZingMp3AlbumIE(ZingMp3BaseIE): + _VALID_URL = ZingMp3BaseIE._VALID_URL_TMPL % 'album|playlist' + _TESTS = [{ 'url': 'http://mp3.zing.vn/album/Lau-Dai-Tinh-Ai-Bang-Kieu-Minh-Tuyet/ZWZBWDAF.html', 'info_dict': { '_type': 'playlist', 'id': 'ZWZBWDAF', - 'title': 'Lâu Đài Tình Ái - Bằng Kiều,Minh Tuyết | Album 320 lossless', + 'title': 'Lâu Đài Tình Ái', }, 'playlist_count': 10, - 'skip': 'removed at the request of the owner', }, { 'url': 'http://mp3.zing.vn/playlist/Duong-Hong-Loan-apollobee/IWCAACCB.html', 'only_matching': True, + }, { + 'url': 'https://zingmp3.vn/album/Lau-Dai-Tinh-Ai-Bang-Kieu-Minh-Tuyet/ZWZBWDAF.html', + 'only_matching': True, }] - IE_NAME = 'zingmp3' - IE_DESC = 'mp3.zing.vn' - - def _real_extract(self, url): - page_id = self._match_id(url) - - webpage = self._download_webpage(url, page_id) - - player_json_url = self._search_regex([ - r'data-xml="([^"]+)', - r'&xmlURL=([^&]+)&' - ], webpage, 'player xml url') - - playlist_title = None - page_type = self._search_regex(r'/(?:html5)?xml/([^/-]+)', player_json_url, 'page type') - if page_type == 'video': - player_json_url = update_url_query(player_json_url, {'format': 'json'}) - else: - player_json_url = player_json_url.replace('/xml/', '/html5xml/') - if page_type == 'album': - playlist_title = self._og_search_title(webpage) - - return self._extract_player_json(player_json_url, page_id, page_type, playlist_title) + IE_NAME = 'zingmp3:album' + + def _process_data(self, data): + def entries(): + for item in (data.get('items') or []): + entry = self._extract_item(item, False) + if entry: + yield entry + info = data.get('info') or {} + return self.playlist_result( + entries(), info.get('id'), info.get('name') or info.get('title')) diff --git a/youtube_dl/extractor/zoom.py b/youtube_dl/extractor/zoom.py new file mode 100644 index 000000000..db073d91d --- /dev/null +++ b/youtube_dl/extractor/zoom.py @@ -0,0 +1,68 @@ +# coding: utf-8 +from __future__ import unicode_literals + +import re + +from .common import InfoExtractor +from ..utils import ( + ExtractorError, + int_or_none, + js_to_json, + parse_filesize, + urlencode_postdata, +) + + +class ZoomIE(InfoExtractor): + IE_NAME = 'zoom' + _VALID_URL = r'(?P<base_url>https?://(?:[^.]+\.)?zoom.us/)rec(?:ording)?/(?:play|share)/(?P<id>[A-Za-z0-9_.-]+)' + _TEST = { + 'url': 'https://economist.zoom.us/rec/play/dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5', + 'md5': 'ab445e8c911fddc4f9adc842c2c5d434', + 'info_dict': { + 'id': 'dUk_CNBETmZ5VA2BwEl-jjakPpJ3M1pcfVYAPRsoIbEByGsLjUZtaa4yCATQuOL3der8BlTwxQePl_j0.EImBkXzTIaPvdZO5', + 'ext': 'mp4', + 'title': 'China\'s "two sessions" and the new five-year plan', + } + } + + def _real_extract(self, url): + base_url, play_id = re.match(self._VALID_URL, url).groups() + webpage = self._download_webpage(url, play_id) + + try: + form = self._form_hidden_inputs('password_form', webpage) + except ExtractorError: + form = None + if form: + password = self._downloader.params.get('videopassword') + if not password: + raise ExtractorError( + 'This video is protected by a passcode, use the --video-password option', expected=True) + is_meeting = form.get('useWhichPasswd') == 'meeting' + validation = self._download_json( + base_url + 'rec/validate%s_passwd' % ('_meet' if is_meeting else ''), + play_id, 'Validating passcode', 'Wrong passcode', data=urlencode_postdata({ + 'id': form[('meet' if is_meeting else 'file') + 'Id'], + 'passwd': password, + 'action': form.get('action'), + })) + if not validation.get('status'): + raise ExtractorError(validation['errorMessage'], expected=True) + webpage = self._download_webpage(url, play_id) + + data = self._parse_json(self._search_regex( + r'(?s)window\.__data__\s*=\s*({.+?});', + webpage, 'data'), play_id, js_to_json) + + return { + 'id': play_id, + 'title': data['topic'], + 'url': data['viewMp4Url'], + 'width': int_or_none(data.get('viewResolvtionsWidth')), + 'height': int_or_none(data.get('viewResolvtionsHeight')), + 'http_headers': { + 'Referer': base_url, + }, + 'filesize_approx': parse_filesize(data.get('fileSize')), + } diff --git a/youtube_dl/extractor/zype.py b/youtube_dl/extractor/zype.py index 5288f40d8..f20f953cb 100644 --- a/youtube_dl/extractor/zype.py +++ b/youtube_dl/extractor/zype.py @@ -87,11 +87,16 @@ class ZypeIE(InfoExtractor): r'(["\'])(?P<url>(?:(?!\1).)+\.m3u8(?:(?!\1).)*)\1', body, 'm3u8 url', group='url', default=None) if not m3u8_url: - source = self._parse_json(self._search_regex( - r'(?s)sources\s*:\s*\[\s*({.+?})\s*\]', body, - 'source'), video_id, js_to_json) - if source.get('integration') == 'verizon-media': - m3u8_url = 'https://content.uplynk.com/%s.m3u8' % source['id'] + source = self._search_regex( + r'(?s)sources\s*:\s*\[\s*({.+?})\s*\]', body, 'source') + + def get_attr(key): + return self._search_regex( + r'\b%s\s*:\s*([\'"])(?P<val>(?:(?!\1).)+)\1' % key, + source, key, group='val') + + if get_attr('integration') == 'verizon-media': + m3u8_url = 'https://content.uplynk.com/%s.m3u8' % get_attr('id') formats = self._extract_m3u8_formats( m3u8_url, video_id, 'mp4', 'm3u8_native', m3u8_id='hls') text_tracks = self._search_regex( |