Merge 8db4101fbe into 6b54cccdcb

Closes #4989
Authored by: alexhuot1, dirkf

yt_dlp/extractor/bilibili.py

@@ -93,11 +93,11 @@ def extract_formats(self, play_info):
 
         return formats
 
-    def _download_playinfo(self, video_id, cid):
+    def _download_playinfo(self, video_id, cid, headers=None):
         return self._download_json(
             'https://api.bilibili.com/x/player/playurl', video_id,
             query={'bvid': video_id, 'cid': cid, 'fnval': 4048},
-            note=f'Downloading video formats for cid {cid}')['data']
+            note=f'Downloading video formats for cid {cid}', headers=headers)['data']
 
     def json2srt(self, json_data):
         srt_data = ''
@@ -493,7 +493,8 @@ class BiliBiliIE(BilibiliBaseIE):
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage, urlh = self._download_webpage_handle(url, video_id)
+        headers = self.geo_verification_headers()
+        webpage, urlh = self._download_webpage_handle(url, video_id, headers=headers)
         if not self._match_valid_url(urlh.url):
             return self.url_result(urlh.url)
 
@@ -531,7 +532,7 @@ def _real_extract(self, url):
             self._download_json(
                 'https://api.bilibili.com/x/player/pagelist', video_id,
                 fatal=False, query={'bvid': video_id, 'jsonp': 'jsonp'},
-                note='Extracting videos in anthology'),
+                note='Extracting videos in anthology', headers=headers),
             'data', expected_type=list) or []
         is_anthology = len(page_list_json) > 1
 
@@ -552,7 +553,7 @@ def _real_extract(self, url):
 
         festival_info = {}
         if is_festival:
-            play_info = self._download_playinfo(video_id, cid)
+            play_info = self._download_playinfo(video_id, cid, headers=headers)
 
             festival_info = traverse_obj(initial_state, {
                 'uploader': ('videoInfo', 'upName'),
@@ -666,14 +667,15 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
 
     def _real_extract(self, url):
         episode_id = self._match_id(url)
-        webpage = self._download_webpage(url, episode_id)
+        headers = self.geo_verification_headers()
+        webpage = self._download_webpage(url, episode_id, headers=headers)
 
         if '您所在的地区无法观看本片' in webpage:
             raise GeoRestrictedError('This video is restricted')
         elif '正在观看预览，大会员免费看全片' in webpage:
             self.raise_login_required('This video is for premium members only')
 
-        headers = {'Referer': url, **self.geo_verification_headers()}
+        headers['Referer'] = url
         play_info = self._download_json(
             'https://api.bilibili.com/pgc/player/web/v2/playurl', episode_id,
             'Extracting episode', query={'fnval': '4048', 'ep_id': episode_id},
@@ -724,7 +726,7 @@ def _real_extract(self, url):
             'duration': float_or_none(play_info.get('timelength'), scale=1000),
             'subtitles': self.extract_subtitles(episode_id, episode_info.get('cid'), aid=aid),
             '__post_extractor': self.extract_comments(aid),
-            'http_headers': headers,
+            'http_headers': {'Referer': url},
         }
 
 
@@ -1049,9 +1051,10 @@ def fetch_page(page_idx):
                     raise ExtractorError(
                         'Request is blocked by server (412), please add cookies, wait and try later.', expected=True)
                 raise
-            if response['code'] == -401:
+            if response['code'] in (-352, -401):
                 raise ExtractorError(
-                    'Request is blocked by server (401), please add cookies, wait and try later.', expected=True)
+                    f'Request is blocked by server ({-response["code"]}), '
+                    'please add cookies, wait and try later.', expected=True)
             return response['data']
 
         def get_metadata(page_data):

yt_dlp/extractor/boosty.py

@@ -1,7 +1,11 @@
+import json
+import urllib.parse
+
 from .common import InfoExtractor
 from .youtube import YoutubeIE
 from ..utils import (
     ExtractorError,
+    bug_reports_message,
     int_or_none,
     qualities,
     str_or_none,
@@ -162,9 +166,19 @@ def _extract_formats(self, player_urls, video_id):
 
     def _real_extract(self, url):
         user, post_id = self._match_valid_url(url).group('user', 'post_id')
+
+        auth_headers = {}
+        auth_cookie = self._get_cookies('https://boosty.to/').get('auth')
+        if auth_cookie is not None:
+            try:
+                auth_data = json.loads(urllib.parse.unquote(auth_cookie.value))
+                auth_headers['Authorization'] = f'Bearer {auth_data["accessToken"]}'
+            except (json.JSONDecodeError, KeyError):
+                self.report_warning(f'Failed to extract token from auth cookie{bug_reports_message()}')
+
         post = self._download_json(
             f'https://api.boosty.to/v1/blog/{user}/post/{post_id}', post_id,
-            note='Downloading post data', errnote='Unable to download post data')
+            note='Downloading post data', errnote='Unable to download post data', headers=auth_headers)
 
         post_title = post.get('title')
         if not post_title:
@@ -202,7 +216,9 @@ def _real_extract(self, url):
                         'thumbnail': (('previewUrl', 'defaultPreview'), {url_or_none}),
                     }, get_all=False)})
 
-        if not entries:
+        if not entries and not post.get('hasAccess'):
+            self.raise_login_required('This post requires a subscription', metadata_available=True)
+        elif not entries:
             raise ExtractorError('No videos found', expected=True)
         if len(entries) == 1:
             return entries[0]

yt_dlp/extractor/cbc.py

@@ -151,7 +151,7 @@ def _real_extract(self, url):
 
 class CBCPlayerIE(InfoExtractor):
     IE_NAME = 'cbc.ca:player'
-    _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/|i/caffeine/syndicate/\?mediaId=))(?P<id>(?:\d\.)?\d+)'
+    _VALID_URL = r'(?:cbcplayer:|https?://(?:www\.)?cbc\.ca/(?:player/play/(?:video/)?|i/caffeine/syndicate/\?mediaId=))(?P<id>(?:\d\.)?\d+)'
     _TESTS = [{
         'url': 'http://www.cbc.ca/player/play/2683190193',
         'md5': '64d25f841ddf4ddb28a235338af32e2c',
@@ -277,6 +277,28 @@ class CBCPlayerIE(InfoExtractor):
             'location': 'Canada',
             'media_type': 'Full Program',
         },
+    }, {
+        'url': 'https://www.cbc.ca/player/play/video/1.7194274',
+        'md5': '188b96cf6bdcb2540e178a6caa957128',
+        'info_dict': {
+            'id': '2334524995812',
+            'ext': 'mp4',
+            'title': '#TheMoment a rare white spirit moose was spotted in Alberta',
+            'description': 'md5:18ae269a2d0265c5b0bbe4b2e1ac61a3',
+            'timestamp': 1714788791,
+            'duration': 77.678,
+            'subtitles': {'eng': [{'ext': 'vtt', 'protocol': 'm3u8_native'}]},
+            'thumbnail': 'https://thumbnails.cbc.ca/maven_legacy/thumbnails/201/543/THE_MOMENT.jpg',
+            'uploader': 'CBCC-NEW',
+            'chapters': 'count:0',
+            'upload_date': '20240504',
+            'categories': 'count:3',
+            'series': 'The National',
+            'tags': 'count:15',
+            'creators': ['encoder'],
+            'location': 'Canada',
+            'media_type': 'Excerpt',
+        },
     }, {
         'url': 'cbcplayer:1.7159484',
         'only_matching': True,

yt_dlp/extractor/crunchyroll.py

@@ -53,15 +53,19 @@ def _set_auth_info(self, response):
         CrunchyrollBaseIE._AUTH_EXPIRY = time_seconds(seconds=traverse_obj(response, ('expires_in', {float_or_none}), default=300) - 10)
 
     def _request_token(self, headers, data, note='Requesting token', errnote='Failed to request token'):
-        try:  # TODO: Add impersonation support here
+        try:
             return self._download_json(
                 f'{self._BASE_URL}/auth/v1/token', None, note=note, errnote=errnote,
-                headers=headers, data=urlencode_postdata(data))
+                headers=headers, data=urlencode_postdata(data), impersonate=True)
         except ExtractorError as error:
             if not isinstance(error.cause, HTTPError) or error.cause.status != 403:
                 raise
+            if target := error.cause.response.extensions.get('impersonate'):
+                raise ExtractorError(f'Got HTTP Error 403 when using impersonate target "{target}"')
             raise ExtractorError(
-                'Request blocked by Cloudflare; navigate to Crunchyroll in your browser, '
+                'Request blocked by Cloudflare. '
+                'Install the required impersonation dependency if possible, '
+                'or else navigate to Crunchyroll in your browser, '
                 'then pass the fresh cookies (with --cookies-from-browser or --cookies) '
                 'and your browser\'s User-Agent (with --user-agent)', expected=True)
 

yt_dlp/extractor/mixch.py

@@ -1,6 +1,12 @@
 from .common import InfoExtractor
 from ..networking.exceptions import HTTPError
-from ..utils import ExtractorError, UserNotLive, int_or_none, url_or_none
+from ..utils import (
+    ExtractorError,
+    UserNotLive,
+    int_or_none,
+    str_or_none,
+    url_or_none,
+)
 from ..utils.traversal import traverse_obj
 
 
@@ -9,17 +15,20 @@ class MixchIE(InfoExtractor):
     _VALID_URL = r'https?://(?:www\.)?mixch\.tv/u/(?P<id>\d+)'
 
     _TESTS = [{
-        'url': 'https://mixch.tv/u/16236849/live',
+        'url': 'https://mixch.tv/u/16943797/live',
         'skip': 'don\'t know if this live persists',
         'info_dict': {
-            'id': '16236849',
-            'title': '24配信シェア⭕️投票🙏💦',
-            'comment_count': 13145,
-            'view_count': 28348,
-            'timestamp': 1636189377,
-            'uploader': '🦥伊咲👶🏻#フレアワ',
-            'uploader_id': '16236849',
-        }
+            'id': '16943797',
+            'ext': 'mp4',
+            'title': '#EntView #カリナ #セブチ 2024-05-05 06:58',
+            'comment_count': int,
+            'view_count': int,
+            'timestamp': 1714726805,
+            'uploader': 'Ent.View K-news🎶💕',
+            'uploader_id': '16943797',
+            'live_status': 'is_live',
+            'upload_date': '20240503',
+        },
     }, {
         'url': 'https://mixch.tv/u/16137876/live',
         'only_matching': True,
@@ -48,8 +57,20 @@ def _real_extract(self, url):
                 'protocol': 'm3u8',
             }],
             'is_live': True,
+            '__post_extractor': self.extract_comments(video_id),
         }
 
+    def _get_comments(self, video_id):
+        yield from traverse_obj(self._download_json(
+            f'https://mixch.tv/api-web/lives/{video_id}/messages', video_id,
+            note='Downloading comments', errnote='Failed to download comments'), (..., {
+                'author': ('name', {str}),
+                'author_id': ('user_id', {str_or_none}),
+                'id': ('message_id', {str}, {lambda x: x or None}),
+                'text': ('body', {str}),
+                'timestamp': ('created', {int}),
+            }))
+
 
 class MixchArchiveIE(InfoExtractor):
     IE_NAME = 'mixch:archive'

yt_dlp/extractor/neteasemusic.py

@@ -22,7 +22,7 @@
 
 
 class NetEaseMusicBaseIE(InfoExtractor):
-    _FORMATS = ['bMusic', 'mMusic', 'hMusic']
+    _FORMATS = ['sqMusic', 'hrMusic', 'lMusic', 'bMusic', 'mMusic', 'hMusic']
     _API_BASE = 'http://music.163.com/api/'
     _GEO_BYPASS = False
 

yt_dlp/extractor/tva.py

@@ -1,10 +1,9 @@
+import functools
+import re
+
 from .common import InfoExtractor
-from ..utils import (
-    float_or_none,
-    int_or_none,
-    smuggle_url,
-    strip_or_none,
-)
+from ..utils import float_or_none, int_or_none, smuggle_url, strip_or_none
+from ..utils.traversal import traverse_obj
 
 
 class TVAIE(InfoExtractor):
@@ -49,11 +48,20 @@ class QubIE(InfoExtractor):
         'info_dict': {
             'id': '6084352463001',
             'ext': 'mp4',
-            'title': 'Épisode 01',
+            'title': 'Ép 01. Mon dernier jour',
             'uploader_id': '5481942443001',
             'upload_date': '20190907',
             'timestamp': 1567899756,
             'description': 'md5:9c0d7fbb90939420c651fd977df90145',
+            'thumbnail': r're:https://.+\.jpg',
+            'episode': 'Ép 01. Mon dernier jour',
+            'episode_number': 1,
+            'tags': ['alerte amber', 'alerte amber saison 1', 'surdemande'],
+            'duration': 2625.963,
+            'season': 'Season 1',
+            'season_number': 1,
+            'series': 'Alerte Amber',
+            'channel': 'TVA',
         },
     }, {
         'url': 'https://www.qub.ca/tele/video/lcn-ca-vous-regarde-rev-30s-ap369664-1009357943',
@@ -64,22 +72,24 @@ class QubIE(InfoExtractor):
 
     def _real_extract(self, url):
         entity_id = self._match_id(url)
-        entity = self._download_json(
-            'https://www.qub.ca/proxy/pfu/content-delivery-service/v1/entities',
-            entity_id, query={'id': entity_id})
+        webpage = self._download_webpage(url, entity_id)
+        entity = self._search_nextjs_data(webpage, entity_id)['props']['initialProps']['pageProps']['fallbackData']
         video_id = entity['videoId']
         episode = strip_or_none(entity.get('name'))
 
         return {
             '_type': 'url_transparent',
+            'url': f'https://videos.tva.ca/details/_{video_id}',
+            'ie_key': TVAIE.ie_key(),
             'id': video_id,
             'title': episode,
-            # 'url': self.BRIGHTCOVE_URL_TEMPLATE % entity['referenceId'],
-            'url': 'https://videos.tva.ca/details/_' + video_id,
-            'description': entity.get('longDescription'),
-            'duration': float_or_none(entity.get('durationMillis'), 1000),
             'episode': episode,
-            'episode_number': int_or_none(entity.get('episodeNumber')),
-            # 'ie_key': 'BrightcoveNew',
-            'ie_key': TVAIE.ie_key(),
+            **traverse_obj(entity, {
+                'description': ('longDescription', {str}),
+                'duration': ('durationMillis', {functools.partial(float_or_none, scale=1000)}),
+                'channel': ('knownEntities', 'channel', 'name', {str}),
+                'series': ('knownEntities', 'videoShow', 'name', {str}),
+                'season_number': ('slug', {lambda x: re.search(r'/s(?:ai|ea)son-(\d+)/', x)}, 1, {int_or_none}),
+                'episode_number': ('episodeNumber', {int_or_none}),
+            }),
         }

yt_dlp/extractor/vk.py

@@ -451,6 +451,7 @@ def _real_extract(self, url):
             info_page, 'view count', default=None))
 
         formats = []
+        subtitles = {}
         for format_id, format_url in data.items():
             format_url = url_or_none(format_url)
             if not format_url or not format_url.startswith(('http', '//', 'rtmp')):
@@ -462,12 +463,21 @@ def _real_extract(self, url):
                 formats.append({
                     'format_id': format_id,
                     'url': format_url,
+                    'ext': 'mp4',
+                    'source_preference': 1,
                     'height': height,
                 })
             elif format_id == 'hls':
-                formats.extend(self._extract_m3u8_formats(
+                fmts, subs = self._extract_m3u8_formats_and_subtitles(
                     format_url, video_id, 'mp4', 'm3u8_native',
-                    m3u8_id=format_id, fatal=False, live=is_live))
+                    m3u8_id=format_id, fatal=False, live=is_live)
+                formats.extend(fmts)
+                self._merge_subtitles(subs, target=subtitles)
+            elif format_id.startswith('dash_'):
+                fmts, subs = self._extract_mpd_formats_and_subtitles(
+                    format_url, video_id, mpd_id=format_id, fatal=False)
+                formats.extend(fmts)
+                self._merge_subtitles(subs, target=subtitles)
             elif format_id == 'rtmp':
                 formats.append({
                     'format_id': format_id,
@@ -475,7 +485,6 @@ def _real_extract(self, url):
                     'ext': 'flv',
                 })
 
-        subtitles = {}
         for sub in data.get('subs') or {}:
             subtitles.setdefault(sub.get('lang', 'en'), []).append({
                 'ext': sub.get('title', '.srt').split('.')[-1],
@@ -496,6 +505,7 @@ def _real_extract(self, url):
             'comment_count': int_or_none(mv_data.get('commcount')),
             'is_live': is_live,
             'subtitles': subtitles,
+            '_format_sort_fields': ('res', 'source'),
         }
 
 

yt_dlp/extractor/yandexvideo.py

@@ -259,15 +259,15 @@ def _real_extract(self, url):
             webpage = self._download_webpage(redirect, video_id, note='Redirecting')
         data_json = self._search_json(
             r'("data"\s*:|data\s*=)', webpage, 'metadata', video_id, contains_pattern=r'{["\']_*serverState_*video.+}')
-        serverstate = self._search_regex(r'(_+serverState_+video-site_[^_]+_+)',
-                                         webpage, 'server state').replace('State', 'Settings')
+        serverstate = self._search_regex(r'(_+serverState_+video-site_[^_]+_+)', webpage, 'server state')
         uploader = self._search_regex(r'(<a\s*class=["\']card-channel-link[^"\']+["\'][^>]+>)',
                                       webpage, 'uploader', default='<a>')
         uploader_name = extract_attributes(uploader).get('aria-label')
-        video_json = try_get(data_json, lambda x: x[serverstate]['exportData']['video'], dict)
-        stream_urls = try_get(video_json, lambda x: x['video']['streams'])
+        item_id = traverse_obj(data_json, (serverstate, 'videoViewer', 'openedItemId', {str}))
+        video_json = traverse_obj(data_json, (serverstate, 'videoViewer', 'items', item_id, {dict})) or {}
+
         formats, subtitles = [], {}
-        for s_url in stream_urls:
+        for s_url in traverse_obj(video_json, ('video', 'streams', ..., {url_or_none})):
             ext = determine_ext(s_url)
             if ext == 'mpd':
                 fmts, subs = self._extract_mpd_formats_and_subtitles(s_url, video_id, mpd_id='dash')

yt_dlp/extractor/youporn.py

@@ -72,15 +72,15 @@ class YouPornIE(InfoExtractor):
             'id': '16290308',
             'age_limit': 18,
             'categories': [],
-            'description': 'md5:00ea70f642f431c379763c17c2f396bc',
+            'description': str,  # TODO: detect/remove SEO spam description in ytdl backport
             'display_id': 'tinderspecial-trailer1',
             'duration': 298.0,
             'ext': 'mp4',
             'upload_date': '20201123',
             'uploader': 'Ersties',
             'tags': [],
-            'thumbnail': 'https://fi1.ypncdn.com/202011/23/16290308/original/8/tinderspecial-trailer1-8(m=eaAaaEPbaaaa).jpg',
-            'timestamp': 1606089600,
+            'thumbnail': r're:https://.+\.jpg',
+            'timestamp': 1606147564,
             'title': 'Tinder In Real Life',
             'view_count': int,
         }
@@ -88,11 +88,17 @@ class YouPornIE(InfoExtractor):
 
     def _real_extract(self, url):
         video_id, display_id = self._match_valid_url(url).group('id', 'display_id')
-        definitions = self._download_json(
-            f'https://www.youporn.com/api/video/media_definitions/{video_id}/', display_id or video_id)
+        self._set_cookie('.youporn.com', 'age_verified', '1')
+        webpage = self._download_webpage(f'https://www.youporn.com/watch/{video_id}', video_id)
+        definitions = self._search_json(r'\bplayervars\s*:', webpage, 'player vars', video_id)['mediaDefinitions']
 
-        def get_format_data(data, f):
-            return traverse_obj(data, lambda _, v: v['format'] == f and url_or_none(v['videoUrl']))
+        def get_format_data(data, stream_type):
+            info_url = traverse_obj(data, (lambda _, v: v['format'] == stream_type, 'videoUrl', {url_or_none}, any))
+            if not info_url:
+                return []
+            return traverse_obj(
+                self._download_json(info_url, video_id, f'Downloading {stream_type} info JSON', fatal=False),
+                lambda _, v: v['format'] == stream_type and url_or_none(v['videoUrl']))
 
         formats = []
         # Try to extract only the actual master m3u8 first, avoiding the duplicate single resolution "master" m3u8s
@@ -123,10 +129,6 @@ def get_format_data(data, f):
             f['height'] = height
             formats.append(f)
 
-        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(