Merge e287aec51d into 5904853ae5

Closes #7442
Authored by: bashonly

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/tiktok.py

@@ -370,13 +370,19 @@ def extract_addr(addr, add_meta={}):
         contained_music_track = traverse_obj(
             music_info, ('matched_song', 'title'), ('matched_pgc_sound', 'title'), expected_type=str)
         contained_music_author = traverse_obj(
-            music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), 'author', expected_type=str)
+            music_info, ('matched_song', 'author'), ('matched_pgc_sound', 'author'), expected_type=str)
 
         is_generic_og_trackname = music_info.get('is_original_sound') and music_info.get('title') == 'original sound - %s' % music_info.get('owner_handle')
-        if is_generic_og_trackname:
-            music_track, music_author = contained_music_track or 'original sound', contained_music_author
+        music_track, music_author = 'original sound' if is_generic_og_trackname else music_info.get('title'), traverse_obj(music_info, ('author', {str}))
+
+        if contained_music_track or contained_music_author:
+            contained_music_info = [{
+                'relation': 'music',
+                'track': contained_music_track,
+                'artist': contained_music_author,
+            }]
         else:
-            music_track, music_author = music_info.get('title'), traverse_obj(music_info, ('author', {str}))
+            contained_music_info = None
 
         return {
             'id': aweme_id,
@@ -400,7 +406,10 @@ def extract_addr(addr, add_meta={}):
             }),
             'uploader_url': user_url,
             'track': music_track,
+            'track_id': str_or_none(music_info.get('id')),
             'album': str_or_none(music_info.get('album')) or None,
+            'attributions': contained_music_info,
+            'timestamp': int_or_none(aweme_detail.get('create_time')),
             'artists': re.split(r'(?:, | & )', music_author) if music_author else None,
             'formats': formats,
             'subtitles': self.extract_subtitles(aweme_detail, aweme_id),
@@ -475,6 +484,7 @@ def _parse_aweme_video_web(self, aweme_detail, webpage_url, video_id):
             }, expected_type=int_or_none),
             **traverse_obj(music_info, {
                 'track': ('title', {str}),
+                'track_id': ('id', {str}, {lambda x: x or None}),
                 'album': ('album', {str}, {lambda x: x or None}),
                 'artists': ('authorName', {str}, {lambda x: [x] if x else None}),
             }),
@@ -517,6 +527,7 @@ class TikTokIE(TikTokBaseIE):
             'artist': 'Ysrbeats',
             'album': 'Lehanga',
             'track': 'Lehanga',
+            'track_id': '6716465478027447045',
         },
         'skip': '404 Not Found',
     }, {
@@ -545,6 +556,7 @@ class TikTokIE(TikTokBaseIE):
             'comment_count': int,
             'artists': ['Evan Todd', 'Jessica Keenan Wynn', 'Alice Lee', 'Barrett Wilbert Weed', 'Jon Eidson'],
             'track': 'Big Fun',
+            'track_id': '209649576000286720',
         },
     }, {
         # Banned audio, only available on the app
@@ -561,7 +573,15 @@ class TikTokIE(TikTokBaseIE):
             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
             'channel_id': 'MS4wLjABAAAAbhBwQC-R1iKoix6jDFsF-vBdfx2ABoDjaZrM9fX6arU3w71q3cOWgWuTXn1soZ7d',
             'track': 'Boka Dance',
+            'track_id': '6984138615588653826',
             'artists': ['md5:29f238c49bc0c176cb3cef1a9cea9fa6'],
+            'attributions': [
+                {
+                    'relation': 'music',
+                    'track': 'Boka Loka Dance',
+                    'artist': 'FAUJI NOSS',
+                }
+            ],
             'timestamp': 1626121503,
             'duration': 18,
             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
@@ -586,6 +606,7 @@ class TikTokIE(TikTokBaseIE):
             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
             'channel_id': 'MS4wLjABAAAATh8Vewkn0LYM7Fo03iec3qKdeCUOcBIouRk1mkiag6h3o_pQu_dUXvZ2EZlGST7_',
             'track': 'Promoted Music',
+            'track_id': '7042692955299203841',
             'timestamp': 1639754738,
             'duration': 30,
             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',
@@ -611,6 +632,7 @@ class TikTokIE(TikTokBaseIE):
             'uploader_url': 'https://www.tiktok.com/@MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
             'channel_id': 'MS4wLjABAAAA0tF1nBwQVVMyrGu3CqttkNgM68Do1OXUFuCY0CRQk8fEtSVDj89HqoqvbSTmUP2W',
             'track': 'original sound',
+            'track_id': '7059698289792273198',
             'timestamp': 1643714123,
             'duration': 6,
             'thumbnail': r're:^https?://[\w\/\.\-]+(~[\w\-]+\.image)?',

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(