Merge 8e977a039e into 5904853ae5

Closes #7442
Authored by: bashonly

test/test_networking.py

@@ -785,6 +785,25 @@ def test_supported_impersonate_targets(self, handler):
                 assert res.status == 200
                 assert std_headers['user-agent'].lower() not in res.read().decode().lower()
 
+    def test_response_extensions(self, handler):
+        with handler() as rh:
+            for target in rh.supported_targets:
+                request = Request(
+                    f'http://127.0.0.1:{self.http_port}/gen_200', extensions={'impersonate': target})
+                res = validate_and_send(rh, request)
+                assert res.extensions['impersonate'] == rh._get_request_target(request)
+
+    def test_http_error_response_extensions(self, handler):
+        with handler() as rh:
+            for target in rh.supported_targets:
+                request = Request(
+                    f'http://127.0.0.1:{self.http_port}/gen_404', extensions={'impersonate': target})
+                try:
+                    validate_and_send(rh, request)
+                except HTTPError as e:
+                    res = e.response
+                assert res.extensions['impersonate'] == rh._get_request_target(request)
+
 
 class TestRequestHandlerMisc:
     """Misc generic tests for request handlers, not related to request or validation testing"""

yt_dlp/YoutubeDL.py

@@ -2136,6 +2136,11 @@ def _filter(f):
 
     def _check_formats(self, formats):
         for f in formats:
+            working = f.get('__working')
+            if working is not None:
+                if working:
+                    yield f
+                continue
             self.to_screen('[info] Testing format %s' % f['format_id'])
             path = self.get_output_path('temp')
             if not self._ensure_dir_exists(f'{path}/'):
@@ -2152,33 +2157,44 @@ def _check_formats(self, formats):
                         os.remove(temp_file.name)
                     except OSError:
                         self.report_warning('Unable to delete temporary file "%s"' % temp_file.name)
+            f['__working'] = success
             if success:
                 yield f
             else:
                 self.to_screen('[info] Unable to download format %s. Skipping...' % f['format_id'])
 
+    def _select_formats(self, formats, selector):
+        return list(selector({
+            'formats': formats,
+            'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
+            'incomplete_formats': (all(f.get('vcodec') == 'none' for f in formats)  # No formats with video
+                                   or all(f.get('acodec') == 'none' for f in formats)),  # OR, No formats with audio
+        }))
+
     def _default_format_spec(self, info_dict, download=True):
+        download = download and not self.params.get('simulate')
+        prefer_best = download and (
+            self.params['outtmpl']['default'] == '-'
+            or info_dict.get('is_live') and not self.params.get('live_from_start'))
 
         def can_merge():
             merger = FFmpegMergerPP(self)
             return merger.available and merger.can_merge()
 
-        prefer_best = (
-            not self.params.get('simulate')
-            and download
-            and (
-                not can_merge()
-                or info_dict.get('is_live') and not self.params.get('live_from_start')
-                or self.params['outtmpl']['default'] == '-'))
-        compat = (
-            prefer_best
-            or self.params.get('allow_multiple_audio_streams', False)
-            or 'format-spec' in self.params['compat_opts'])
+        if not prefer_best and download and not can_merge():
+            prefer_best = True
+            formats = self._get_formats(info_dict)
+            evaluate_formats = lambda spec: self._select_formats(formats, self.build_format_selector(spec))
+            if evaluate_formats('b/bv+ba') != evaluate_formats('bv*+ba/b'):
+                self.report_warning('ffmpeg not found. The downloaded format may not be the best available. '
+                                    'Installing ffmpeg is strongly recommended: https://github.com/yt-dlp/yt-dlp#dependencies')
 
-        return (
-            'best/bestvideo+bestaudio' if prefer_best
-            else 'bestvideo*+bestaudio/best' if not compat
-            else 'bestvideo+bestaudio/best')
+        compat = (self.params.get('allow_multiple_audio_streams')
+                  or 'format-spec' in self.params['compat_opts'])
+
+        return ('best/bestvideo+bestaudio' if prefer_best
+                else 'bestvideo+bestaudio/best' if compat
+                else 'bestvideo*+bestaudio/best')
 
     def build_format_selector(self, format_spec):
         def syntax_error(note, start):
@@ -2928,12 +2944,7 @@ def is_wellformed(f):
                 self.write_debug(f'Default format spec: {req_format}')
                 format_selector = self.build_format_selector(req_format)
 
-            formats_to_download = list(format_selector({
-                'formats': formats,
-                'has_merged_format': any('none' not in (f.get('acodec'), f.get('vcodec')) for f in formats),
-                'incomplete_formats': (all(f.get('vcodec') == 'none' for f in formats)  # No formats with video
-                                       or all(f.get('acodec') == 'none' for f in formats)),  # OR, No formats with audio
-            }))
+            formats_to_download = self._select_formats(formats, format_selector)
             if interactive_format_selection and not formats_to_download:
                 self.report_error('Requested format is not available', tb=False, is_error=False)
                 continue

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)
 
@@ -394,10 +398,11 @@ def entries():
         if not self._IS_PREMIUM and traverse_obj(response, (f'{object_type}_metadata', 'is_premium_only')):
             message = f'This {object_type} is for premium members only'
             if CrunchyrollBaseIE._REFRESH_TOKEN:
-                raise ExtractorError(message, expected=True)
-            self.raise_login_required(message, method='password')
-
-        result['formats'], result['subtitles'] = self._extract_stream(internal_id)
+                self.raise_no_formats(message, expected=True, video_id=internal_id)
+            else:
+                self.raise_login_required(message, method='password', metadata_available=True)
+        else:
+            result['formats'], result['subtitles'] = self._extract_stream(internal_id)
 
         result['chapters'] = self._extract_chapters(internal_id)
 
@@ -583,14 +588,16 @@ def _real_extract(self, url):
         if not response:
             raise ExtractorError(f'No video with id {internal_id} could be found (possibly region locked?)', expected=True)
 
+        result = self._transform_music_response(response)
+
         if not self._IS_PREMIUM and response.get('isPremiumOnly'):
             message = f'This {response.get("type") or "media"} is for premium members only'
             if CrunchyrollBaseIE._REFRESH_TOKEN:
-                raise ExtractorError(message, expected=True)
-            self.raise_login_required(message, method='password')
-
-        result = self._transform_music_response(response)
-        result['formats'], _ = self._extract_stream(f'music/{internal_id}', internal_id)
+                self.raise_no_formats(message, expected=True, video_id=internal_id)
+            else:
+                self.raise_login_required(message, method='password', metadata_available=True)
+        else:
+            result['formats'], _ = self._extract_stream(f'music/{internal_id}', internal_id)
 
         return result
 

yt_dlp/extractor/patreon.py

@@ -219,7 +219,29 @@ class PatreonIE(PatreonBaseIE):
             'thumbnail': r're:^https?://.+',
         },
         'params': {'skip_download': 'm3u8'},
+    }, {
+        # multiple attachments/embeds
+        'url': 'https://www.patreon.com/posts/holy-wars-solos-100601977',
+        'playlist_count': 3,
+        'info_dict': {
+            'id': '100601977',
+            'title': '"Holy Wars" (Megadeth) Solos Transcription & Lesson/Analysis',
+            'description': 'md5:d099ab976edfce6de2a65c2b169a88d3',
+            'uploader': 'Bradley Hall',
+            'uploader_id': '24401883',
+            'uploader_url': 'https://www.patreon.com/bradleyhallguitar',
+            'channel_id': '3193932',
+            'channel_url': 'https://www.patreon.com/bradleyhallguitar',
+            'channel_follower_count': int,
+            'timestamp': 1710777855,
+            'upload_date': '20240318',
+            'like_count': int,
+            'comment_count': int,
+            'thumbnail': r're:^https?://.+',
+        },
+        'skip': 'Patron-only content',
     }]
+    _RETURN_TYPE = 'video'
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
@@ -234,58 +256,54 @@ def _real_extract(self, url):
                 'include': 'audio,user,user_defined_tags,campaign,attachments_media',
             })
         attributes = post['data']['attributes']
-        title = attributes['title'].strip()
-        image = attributes.get('image') or {}
-        info = {
-            'id': video_id,
-            'title': title,
-            'description': clean_html(attributes.get('content')),
-            'thumbnail': image.get('large_url') or image.get('url'),
-            'timestamp': parse_iso8601(attributes.get('published_at')),
-            'like_count': int_or_none(attributes.get('like_count')),
-            'comment_count': int_or_none(attributes.get('comment_count')),
-        }
-        can_view_post = traverse_obj(attributes, 'current_user_can_view')
-        if can_view_post and info['comment_count']:
-            info['__post_extractor'] = self.extract_comments(video_id)
+        info = traverse_obj(attributes, {
+            'title': ('title', {str.strip}),
+            'description': ('content', {clean_html}),
+            'thumbnail': ('image', ('large_url', 'url'), {url_or_none}, any),
+            'timestamp': ('published_at', {parse_iso8601}),
+            'like_count': ('like_count', {int_or_none}),
+            'comment_count': ('comment_count', {int_or_none}),
+        })
 
-        for i in post.get('included', []):
-            i_type = i.get('type')
-            if i_type == 'media':
-                media_attributes = i.get('attributes') or {}
-                download_url = media_attributes.get('download_url')
+        entries = []
+        idx = 0
+        for include in traverse_obj(post, ('included', lambda _, v: v['type'])):
+            include_type = include['type']
+            if include_type == 'media':
+                media_attributes = traverse_obj(include, ('attributes', {dict})) or {}
+                download_url = url_or_none(media_attributes.get('download_url'))
                 ext = mimetype2ext(media_attributes.get('mimetype'))
 
                 # if size_bytes is None, this media file is likely unavailable
                 # See: https://github.com/yt-dlp/yt-dlp/issues/4608
                 size_bytes = int_or_none(media_attributes.get('size_bytes'))
                 if download_url and ext in KNOWN_EXTENSIONS and size_bytes is not None:
-                    # XXX: what happens if there are multiple attachments?
-                    return {
-                        **info,
+                    idx += 1
+                    entries.append({
+                        'id': f'{video_id}-{idx}',
                         'ext': ext,
                         'filesize': size_bytes,
                         'url': download_url,
-                    }
-            elif i_type == 'user':
-                user_attributes = i.get('attributes')
-                if user_attributes:
-                    info.update({
-                        'uploader': user_attributes.get('full_name'),
-                        'uploader_id': str_or_none(i.get('id')),
-                        'uploader_url': user_attributes.get('url'),
                     })
 
-            elif i_type == 'post_tag':
-                info.setdefault('tags', []).append(traverse_obj(i, ('attributes', 'value')))
+            elif include_type == 'user':
+                info.update(traverse_obj(include, {
+                    'uploader': ('attributes', 'full_name', {str}),
+                    'uploader_id': ('id', {str_or_none}),
+                    'uploader_url': ('attributes', 'url', {url_or_none}),
+                }))
 
-            elif i_type == 'campaign':
-                info.update({
-                    'channel': traverse_obj(i, ('attributes', 'title')),
-                    'channel_id': str_or_none(i.get('id')),
-                    'channel_url': traverse_obj(i, ('attributes', 'url')),
-                    'channel_follower_count': int_or_none(traverse_obj(i, ('attributes', 'patron_count'))),
-                })
+            elif include_type == 'post_tag':
+                if post_tag := traverse_obj(include, ('attributes', 'value', {str})):
+                    info.setdefault('tags', []).append(post_tag)
+
+            elif include_type == 'campaign':
+                info.update(traverse_obj(include, {
+                    'channel': ('attributes', 'title', {str}),
+                    'channel_id': ('id', {str_or_none}),
+                    'channel_url': ('attributes', 'url', {url_or_none}),
+                    'channel_follower_count': ('attributes', 'patron_count', {int_or_none}),
+                }))
 
         # handle Vimeo embeds
         if traverse_obj(attributes, ('embed', 'provider')) == 'Vimeo':
@@ -296,36 +314,50 @@ def _real_extract(self, url):
                     v_url, video_id, 'Checking Vimeo embed URL',
                     headers={'Referer': 'https://patreon.com/'},
                     fatal=False, errnote=False):
-                return self.url_result(
+                entries.append(self.url_result(
                     VimeoIE._smuggle_referrer(v_url, 'https://patreon.com/'),
-                    VimeoIE, url_transparent=True, **info)
+                    VimeoIE, url_transparent=True))
 
         embed_url = traverse_obj(attributes, ('embed', 'url', {url_or_none}))
         if embed_url and self._request_webpage(embed_url, video_id, 'Checking embed URL', fatal=False, errnote=False):
-            return self.url_result(embed_url, **info)
+            entries.append(self.url_result(embed_url))
 
-        post_file = traverse_obj(attributes, 'post_file')
+        post_file = traverse_obj(attributes, ('post_file', {dict}))
         if post_file:
             name = post_file.get('name')
             ext = determine_ext(name)
             if ext in KNOWN_EXTENSIONS:
-                return {
-                    **info,
+                entries.append({
+                    'id': video_id,
                     'ext': ext,
                     'url': post_file['url'],
-                }
+                })
             elif name == 'video' or determine_ext(post_file.get('url')) == 'm3u8':
                 formats, subtitles = self._extract_m3u8_formats_and_subtitles(post_file['url'], video_id)
-                return {
-                    **info,
+                entries.append({
+                    'id': video_id,
                     'formats': formats,
                     'subtitles': subtitles,
-                }
+                })
 
-        if can_view_post is False:
+        can_view_post = traverse_obj(attributes, 'current_user_can_view')
+        comments = None
+        if can_view_post and info.get('comment_count'):
+            comments = self.extract_comments(video_id)
+
+        if not entries and can_view_post is False:
             self.raise_no_formats('You do not have access to this post', video_id=video_id, expected=True)
-        else:
+        elif not entries:
             self.raise_no_formats('No supported media found in this post', video_id=video_id, expected=True)
+        elif len(entries) == 1:
+            info.update(entries[0])
+        else:
+            for entry in entries:
+                entry.update(info)
+            return self.playlist_result(entries, video_id, **info, __post_extractor=comments)
+
+        info['id'] = video_id
+        info['__post_extractor'] = comments
         return info
 
     def _get_comments(self, post_id):

yt_dlp/extractor/rtp.py

@@ -1,5 +1,11 @@
 from .common import InfoExtractor
-from ..utils import js_to_json
+from ..utils import (
+    ExtractorError,
+    RegexNotFoundError,
+    determine_ext,
+    join_nonempty,
+    js_to_json,
+)
 import re
 import json
 import urllib.parse
@@ -7,19 +13,72 @@
 
 
 class RTPIE(InfoExtractor):
-    _VALID_URL = r'https?://(?:www\.)?rtp\.pt/play/p(?P<program_id>[0-9]+)/(?P<id>[^/?#]+)/?'
+    _VALID_URL = r'https?://(?:(?:(?:www\.)?rtp\.pt/play/(?P<subarea>.*/)?p(?P<program_id>[0-9]+)/)|(?:arquivos\.rtp\.pt/conteudos/))(?P<id>[^/?#]+)/?'
     _TESTS = [{
-        'url': 'http://www.rtp.pt/play/p405/e174042/paixoes-cruzadas',
-        'md5': 'e736ce0c665e459ddb818546220b4ef8',
+        'url': 'https://www.rtp.pt/play/p9165/e562949/por-do-sol',
         'info_dict': {
-            'id': 'e174042',
-            'ext': 'mp3',
-            'title': 'Paixões Cruzadas',
-            'description': 'As paixões musicais de António Cartaxo e António Macedo',
-            'thumbnail': r're:^https?://.*\.jpg',
+            'id': 'e562949',
+            'ext': 'mp4',
+            'title': 'Pôr do Sol Episódio 1',
+            'description': 'Madalena Bourbon de Linhaça vive atormentada pelo segredo que esconde desde 1990. Matilde Bourbon de Linhaça sonha fugir com o seu amor proibido. O',
+            'thumbnail': r're:^https?://.*\.(jpg|png)'
         },
     }, {
-        'url': 'http://www.rtp.pt/play/p831/a-quimica-das-coisas',
+        'url': 'https://www.rtp.pt/play/p12646/e738493/telejornal',
+        'info_dict': {
+            'id': 'e738493',
+            'ext': 'mp4',
+            'title': 'Telejornal de 01 jan 2024 PARTE 1',
+            'description': 'A mais rigorosa seleção de notícias, todos os dias às 20h00. De segunda a domingo, João Adelino Faria, José Rodrigues dos Santos e Ana Lourenço',
+            'thumbnail': r're:^https?://.*\.(jpg|png)'
+        },
+    }, {
+        'url': 'https://www.rtp.pt/play/p6646/e457262/grande-entrevista',
+        'info_dict': {
+            'id': 'e457262',
+            'ext': 'mp4',
+            'title': 'Grande Entrevista Episódio 7 - de 19 fev 2020',
+            'description': 'Bruno Nogueira - É um dos mais originais humoristas portugueses e de maior êxito! Bruno Nogueira na Grande Entrevista com Vítor Gonçalves.',
+            'thumbnail': r're:^https?://.*\.(jpg|png)'
+        },
+    }, {
+        'url': 'https://www.rtp.pt/play/p8064/e750623/fronteira',
+        'info_dict': {
+            'id': 'e750623',
+            'ext': 'mp4',
+            'title': 'Fronteira de 26 fev 2024',
+            'description': '1970. À aldeia de Fronteira chega um novo chefe de posto da Guarda Fiscal. Com convicções inabaláveis sobre a aplicação da Lei, rapidamente entr',
+            'thumbnail': r're:^https?://.*\.(jpg|png)'
+        },
+    }, {
+        'url': 'https://www.rtp.pt/play/estudoemcasa/p7776/e539826/portugues-1-ano',
+        'info_dict': {
+            'id': 'e539826',
+            'ext': 'mp4',
+            'title': 'Português - 1.º ano , aula 45 - 27 abr 2021 - Estudo Em Casa - RTP',
+            'description': 'A História do Pedrito Coelho, de Beatrix Potter. O dígrafo \'lh\' - A História do Pedrito Coelho, de Beatrix Potter. O dígrafo \'lh\'.',
+            'thumbnail': r're:^https?://.*\.(jpg|png)'
+        },
+    }, {
+        'url': 'https://www.rtp.pt/play/zigzag/p11099/e747372/coelhos-corajosos',
+        'info_dict': {
+            'id': 'e747372',
+            'ext': 'mp4',
+            'title': 'Coelhos Corajosos Episódio 1 - de 12 fev 2024 - Zig Zag Play - RTP',
+            'description': 'Boo e o seu irmão mais velho, Bop, vivem grandes aventuras com os seus amigos, e com os seus quatro irmãos pequeninos. Juntos e com muita coragem, e',
+            'thumbnail': r're:^https?://.*\.(jpg|png)'
+        },
+    }, {
+        'url': 'https://arquivos.rtp.pt/conteudos/liga-dos-ultimos-152/',
+        'info_dict': {
+            'id': 'liga-dos-ultimos-152',
+            'ext': 'mp4',
+            'title': 'Liga dos Últimos – RTP Arquivos',
+            'description': 'Magazine desportivo, com apresentação de Álvaro Costa e comentários em estúdio do professor Hernâni Gonçalves e do sociólogo João Nuno Coelho. Destaque para os jogos de futebol das equipas dos escalões secundários de Portugal, com momentos dos jogos: Agrário de Lamas vs Pampilhoense e Apúlia vs Fragoso.',
+            'thumbnail': r're:^https?://.*\.(jpg|png)'
+        },
+    }, {
+        'url': 'https://www.rtp.pt/play/p510/aleixo-fm',
         'only_matching': True,
     }]
 
@@ -43,42 +102,78 @@ 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', webpage, display_name='title', fatal=True)
 
-        f, config = self._search_regex(
-            r'''(?sx)
-                var\s+f\s*=\s*(?P<f>".*?"|{[^;]+?});\s*
-                var\s+player1\s+=\s+new\s+RTPPlayer\s*\((?P<config>{(?:(?!\*/).)+?})\);(?!\s*\*/)
-            ''', webpage,
-            'player config', group=('f', 'config'))
+        title = self._html_search_regex(r'<title>(.+?)</title>', webpage, 'title', default='')
 
-        f = self._parse_json(
-            f, video_id,
-            lambda data: self.__unobfuscate(data, video_id=video_id))
-        config = self._parse_json(
-            config, video_id,
-            lambda data: self.__unobfuscate(data, video_id=video_id))
+        # Raise error if episode is unavailable
+        if 'Este episódio não se encontra disponível' in title:
+            raise ExtractorError('Episode unavailable', expected=True)
+
+        # Replace irrelevant string in title
+        title = re.sub(r' -  ?RTP Play - RTP', '', title)
+
+        # Check if it's a program split in parts
+        part = self._html_search_regex(r'section\-parts.*<span.*>(.+?)</span>.*</ul>', webpage, 'part', default=None)
+
+        # Add program part identification to title if it exists
+        title = join_nonempty(title, part, delim=' ')
+
+        try:
+            # Extract f and config from page
+            f, config = self._search_regex(
+                r'''(?sx)
+                    var\s+f\s*=\s*(?P<f>".*?"|{[^;]+?});\s*
+                    var\s+player1\s+=\s+new\s+RTPPlayer\s*\((?P<config>{(?:(?!\*/).)+?})\);(?!\s*\*/)
+                ''', webpage,
+                'player config', group=('f', 'config'))
+
+            f = self._parse_json(
+                f, video_id,
+                lambda data: self.__unobfuscate(data, video_id=video_id))
+
+            config = self._parse_json(
+                config, video_id,
+                lambda data: self.__unobfuscate(data, video_id=video_id))
+
+            config['file'] = f
+        except RegexNotFoundError:
+            # Estudo em Casa / Zig Zag / RTP Arquivos pages don't include f
+            config = self._search_regex(
+                r'''(?sx)
+                    var\s+player1\s+=\s+new\s+RTPPlayer\s*\((?P<config>{(?:(?!\*/).)+?})\);(?!\s*\*/)
+                ''', webpage,
+                'just player config')
+
+            config = self._parse_json(
+                config, video_id,
+                lambda data: self.__unobfuscate(data, video_id=video_id))
 
         formats = []
-        if isinstance(f, dict):
-            f_hls = f.get('hls')
-            if f_hls is not None:
-                formats.extend(self._extract_m3u8_formats(
-                    f_hls, video_id, 'mp4', 'm3u8_native', m3u8_id='hls'))
+        file = config.get('file')
+        if isinstance(file, dict):
+            file_hls = file.get('hls')
+            file_fps = file.get('fps')
 
-            f_dash = f.get('dash')
-            if f_dash is not None:
-                formats.extend(self._extract_mpd_formats(f_dash, video_id, mpd_id='dash'))
+            if file_hls is None and file_fps is not None:
+                file_hls = file_fps.replace('drm-fps', 'hls')
+
+            formats.extend(self._extract_m3u8_formats(
+                file_hls, video_id, 'mp4', 'm3u8_native', m3u8_id='hls'))
         else:
-            formats.append({
-                'format_id': 'f',
-                'url': f,
-                'vcodec': 'none' if config.get('mediaType') == 'audio' else None,
-            })
+            ext = determine_ext(file)
+
+            if ext == 'm3u8':
+                formats.extend(self._extract_m3u8_formats(
+                    file, video_id, 'mp4', 'm3u8_native', m3u8_id='hls'))
+
+            else:
+                formats.append({
+                    'format_id': 'f',
+                    'url': file,
+                    'vcodec': 'none' if config.get('mediaType') == 'audio' else None,
+                })
 
         subtitles = {}
-
         vtt = config.get('vtt')
         if vtt is not None:
             for lcode, lname, url in vtt:

yt_dlp/extractor/soundcloud.py

@@ -361,7 +361,7 @@ def extract_count(key):
             'like_count': extract_count('favoritings') or extract_count('likes'),
             'comment_count': extract_count('comment'),
             'repost_count': extract_count('reposts'),
-            'genre': info.get('genre'),
+            'genres': traverse_obj(info, ('genre', {str}, {lambda x: x or None}, all)),
             'formats': formats if not extract_flat else None
         }
 
@@ -395,10 +395,10 @@ class SoundcloudIE(SoundcloudBaseIE):
     _TESTS = [
         {
             'url': 'http://soundcloud.com/ethmusic/lostin-powers-she-so-heavy',
-            'md5': 'ebef0a451b909710ed1d7787dddbf0d7',
+            'md5': 'de9bac153e7427a7333b4b0c1b6a18d2',
             'info_dict': {
                 'id': '62986583',
-                'ext': 'mp3',
+                'ext': 'opus',
                 'title': 'Lostin Powers - She so Heavy (SneakPreview) Adrian Ackers Blueprint 1',
                 'description': 'No Downloads untill we record the finished version this weekend, i was too pumped n i had to post it , earl is prolly gonna b hella p.o\'d',
                 'uploader': 'E.T. ExTerrestrial Music',
@@ -411,6 +411,9 @@ class SoundcloudIE(SoundcloudBaseIE):
                 'like_count': int,
                 'comment_count': int,
                 'repost_count': int,
+                'thumbnail': 'https://i1.sndcdn.com/artworks-000031955188-rwb18x-original.jpg',
+                'uploader_url': 'https://soundcloud.com/ethmusic',
+                'genres': [],
             }
         },
         # geo-restricted
@@ -418,7 +421,7 @@ class SoundcloudIE(SoundcloudBaseIE):
             'url': 'https://soundcloud.com/the-concept-band/goldrushed-mastered?in=the-concept-band/sets/the-royal-concept-ep',
             'info_dict': {
                 'id': '47127627',
-                'ext': 'mp3',
+                'ext': 'opus',
                 'title': 'Goldrushed',
                 'description': 'From Stockholm Sweden\r\nPovel / Magnus / Filip / David\r\nwww.theroyalconcept.com',
                 'uploader': 'The Royal Concept',
@@ -431,6 +434,9 @@ class SoundcloudIE(SoundcloudBaseIE):
                 'like_count': int,
                 'comment_count': int,
                 'repost_count': int,
+                'uploader_url': 'https://soundcloud.com/the-concept-band',
+                'thumbnail': 'https://i1.sndcdn.com/artworks-v8bFHhXm7Au6-0-original.jpg',
+                'genres': ['Alternative'],
             },
         },
         # private link
@@ -452,6 +458,9 @@ class SoundcloudIE(SoundcloudBaseIE):
                 'like_count': int,
                 'comment_count': int,
                 'repost_count': int,
+                'uploader_url': 'https://soundcloud.com/jaimemf',
+                'thumbnail': 'https://a1.sndcdn.com/images/default_avatar_large.png',
+                'genres': ['youtubedl'],
             },
         },
         # private link (alt format)
@@ -473,6 +482,9 @@ class SoundcloudIE(SoundcloudBaseIE):
                 'like_count': int,
                 'comment_count': int,
                 'repost_count': int,
+                'uploader_url': 'https://soundcloud.com/jaimemf',
+                'thumbnail': 'https://a1.sndcdn.com/images/default_avatar_large.png',
+                'genres': ['youtubedl'],
             },
         },
         # downloadable song
@@ -482,6 +494,21 @@ class SoundcloudIE(SoundcloudBaseIE):
             'info_dict': {
                 'id': '343609555',
                 'ext': 'wav',
+                'title': 'The Following',
+                'description': '',
+                'uploader': '80M',
+                'uploader_id': '312384765',
+                'uploader_url': 'https://soundcloud.com/the80m',
+                'upload_date': '20170922',
+                'timestamp': 1506120436,
+                'duration': 397.228,
+                'thumbnail': 'https://i1.sndcdn.com/artworks-000243916348-ktoo7d-original.jpg',
+                'license': 'all-rights-reserved',
+                'like_count': int,
+                'comment_count': int,
+                'repost_count': int,
+                'view_count': int,
+                'genres': ['Dance & EDM'],
             },
         },
         # private link, downloadable format
@@ -503,6 +530,9 @@ class SoundcloudIE(SoundcloudBaseIE):
                 'like_count': int,
                 'comment_count': int,
                 'repost_count': int,
+                'thumbnail': 'https://i1.sndcdn.com/artworks-000240712245-kedn4p-original.jpg',
+                'uploader_url': 'https://soundcloud.com/oriuplift',
+                'genres': ['Trance'],
             },
         },
         # no album art, use avatar pic for thumbnail
@@ -525,6 +555,8 @@ class SoundcloudIE(SoundcloudBaseIE):
                 'like_count': int,
                 'comment_count': int,
                 'repost_count': int,
+                'uploader_url': 'https://soundcloud.com/garyvee',
+                'genres': [],
             },
             'params': {
                 'skip_download': True,
@@ -532,13 +564,13 @@ class SoundcloudIE(SoundcloudBaseIE):
         },
         {
             'url': 'https://soundcloud.com/giovannisarani/mezzo-valzer',
-            'md5': 'e22aecd2bc88e0e4e432d7dcc0a1abf7',
+            'md5': '8227c3473a4264df6b02ad7e5b7527ac',
             'info_dict': {
                 'id': '583011102',
-                'ext': 'mp3',
+                'ext': 'opus',
                 'title': 'Mezzo Valzer',
-                'description': 'md5:4138d582f81866a530317bae316e8b61',
-                'uploader': 'Micronie',
+                'description': 'md5:f4d5f39d52e0ccc2b4f665326428901a',
+                'uploader': 'Giovanni Sarani',
                 'uploader_id': '3352531',
                 'timestamp': 1551394171,
                 'upload_date': '20190228',
@@ -549,6 +581,8 @@ class SoundcloudIE(SoundcloudBaseIE):
                 'like_count': int,
                 'comment_count': int,
                 'repost_count': int,
+                'genres': ['Piano'],
+                'uploader_url': 'https://soundcloud.com/giovannisarani',
             },
         },
         {

yt_dlp/extractor/wrestleuniverse.py

@@ -12,6 +12,7 @@
     jwt_decode_hs256,
     traverse_obj,
     try_call,
+    url_basename,
     url_or_none,
     urlencode_postdata,
     variadic,
@@ -194,8 +195,7 @@ def _real_extract(self, url):
 
         return {
             'id': video_id,
-            'formats': self._get_formats(video_data, (
-                (('protocolHls', 'url'), ('chromecastUrls', ...)), {url_or_none}), video_id),
+            'formats': self._get_formats(video_data, ('protocolHls', 'url', {url_or_none}), video_id),
             **traverse_obj(metadata, {
                 'title': ('displayName', {str}),
                 'description': ('description', {str}),
@@ -259,6 +259,10 @@ class WrestleUniversePPVIE(WrestleUniverseBaseIE):
         'params': {
             'skip_download': 'm3u8',
         },
+    }, {
+        'note': 'manifest provides live-a (partial) and live-b (full) streams',
+        'url': 'https://www.wrestle-universe.com/en/lives/umc99R9XsexXrxr9VjTo9g',
+        'only_matching': True,
     }]
 
     _API_PATH = 'events'
@@ -285,12 +289,16 @@ def _real_extract(self, url):
 
         video_data, decrypt = self._call_encrypted_api(
             video_id, ':watchArchive', 'watch archive', data={'method': 1})
-        info['formats'] = self._get_formats(video_data, (
-            ('hls', None), ('urls', 'chromecastUrls'), ..., {url_or_none}), video_id)
+        # 'chromecastUrls' can be only partial videos, avoid
+        info['formats'] = self._get_formats(video_data, ('hls', (('urls', ...), 'url'), {url_or_none}), video_id)
         for f in info['formats']:
             # bitrates are exaggerated in PPV playlists, so avoid wrong/huge filesize_approx values
             if f.get('tbr'):
                 f['tbr'] = int(f['tbr'] / 2.5)
+            # prefer variants with the same basename as the master playlist to avoid partial streams
+            f['format_id'] = url_basename(f['url']).partition('.')[0]
+            if not f['format_id'].startswith(url_basename(f['manifest_url']).partition('.')[0]):
+                f['preference'] = -10
 
         hls_aes_key = traverse_obj(video_data, ('hls', 'key', {decrypt}))
         if hls_aes_key:

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(

yt_dlp/networking/_curlcffi.py

@@ -132,6 +132,16 @@ def _check_extensions(self, extensions):
         extensions.pop('cookiejar', None)
         extensions.pop('timeout', None)
 
+    def send(self, request: Request) -> Response:
+        target = self._get_request_target(request)
+        try:
+            response = super().send(request)
+        except HTTPError as e:
+            e.response.extensions['impersonate'] = target
+            raise
+        response.extensions['impersonate'] = target
+        return response
+
     def _send(self, request: Request):
         max_redirects_exceeded = False
         session: curl_cffi.requests.Session = self._get_instance(

yt_dlp/networking/common.py

@@ -497,6 +497,7 @@ class Response(io.IOBase):
     @param headers: response headers.
     @param status: Response HTTP status code. Default is 200 OK.
     @param reason: HTTP status reason. Will use built-in reasons based on status code if not provided.
+    @param extensions: Dictionary of handler-specific response extensions.
     """
 
     def __init__(
@@ -505,7 +506,9 @@ def __init__(
             url: str,
             headers: Mapping[str, str],
             status: int = 200,
-            reason: str = None):
+            reason: str = None,
+            extensions: dict = None
+    ):
 
         self.fp = fp
         self.headers = Message()
@@ -517,6 +520,7 @@ def __init__(
             self.reason = reason or HTTPStatus(status).phrase
         except ValueError:
             self.reason = None
+        self.extensions = extensions or {}
 
     def readable(self):
         return self.fp.readable()