Merge 9dd8574b68 into 0b7ec08816

Closes #10986, Closes #11106
Authored by: DarkZeros, bashonly

Co-authored-by: bashonly <88596187+bashonly@users.noreply.github.com>

yt_dlp/extractor/cbc.py

@@ -4,7 +4,6 @@ import json
 import re
 import time
 import urllib.parse
-import xml.etree.ElementTree
 
 from .common import InfoExtractor
 from ..networking import HEADRequest
@@ -12,7 +11,6 @@ from ..utils import (
     ExtractorError,
     float_or_none,
     int_or_none,
-    join_nonempty,
     js_to_json,
     mimetype2ext,
     orderedSet,
@@ -524,14 +522,13 @@ class CBCGemIE(InfoExtractor):
     _TESTS = [{
         # This is a normal, public, TV show video
         'url': 'https://gem.cbc.ca/media/schitts-creek/s06e01',
-        'md5': '93dbb31c74a8e45b378cf13bd3f6f11e',
         'info_dict': {
             'id': 'schitts-creek/s06e01',
             'ext': 'mp4',
             'title': 'Smoke Signals',
             'description': 'md5:929868d20021c924020641769eb3e7f1',
-            'thumbnail': 'https://images.radio-canada.ca/v1/synps-cbc/episode/perso/cbc_schitts_creek_season_06e01_thumbnail_v01.jpg?im=Resize=(Size)',
-            'duration': 1314,
+            'thumbnail': r're:https://images\.radio-canada\.ca/[^#?]+/cbc_schitts_creek_season_06e01_thumbnail_v01\.jpg',
+            'duration': 1324,
             'categories': ['comedy'],
             'series': 'Schitt\'s Creek',
             'season': 'Season 6',
@@ -539,19 +536,21 @@ class CBCGemIE(InfoExtractor):
             'episode': 'Smoke Signals',
             'episode_number': 1,
             'episode_id': 'schitts-creek/s06e01',
+            'upload_date': '20210618',
+            'timestamp': 1623988800,
+            'release_date': '20200107',
+            'release_timestamp': 1578427200,
         },
         'params': {'format': 'bv'},
-        'skip': 'Geo-restricted to Canada',
     }, {
         # This video requires an account in the browser, but works fine in yt-dlp
         'url': 'https://gem.cbc.ca/media/schitts-creek/s01e01',
-        'md5': '297a9600f554f2258aed01514226a697',
         'info_dict': {
             'id': 'schitts-creek/s01e01',
             'ext': 'mp4',
             'title': 'The Cup Runneth Over',
             'description': 'md5:9bca14ea49ab808097530eb05a29e797',
-            'thumbnail': 'https://images.radio-canada.ca/v1/synps-cbc/episode/perso/cbc_schitts_creek_season_01e01_thumbnail_v01.jpg?im=Resize=(Size)',
+            'thumbnail': r're:https://images\.radio-canada\.ca/[^#?]+/cbc_schitts_creek_season_01e01_thumbnail_v01\.jpg',
             'series': 'Schitt\'s Creek',
             'season_number': 1,
             'season': 'Season 1',
@@ -560,9 +559,12 @@ class CBCGemIE(InfoExtractor):
             'episode_id': 'schitts-creek/s01e01',
             'duration': 1309,
             'categories': ['comedy'],
+            'upload_date': '20210617',
+            'timestamp': 1623902400,
+            'release_date': '20151124',
+            'release_timestamp': 1448323200,
         },
         'params': {'format': 'bv'},
-        'skip': 'Geo-restricted to Canada',
     }, {
         'url': 'https://gem.cbc.ca/nadiyas-family-favourites/s01e01',
         'only_matching': True,
@@ -631,38 +633,6 @@ class CBCGemIE(InfoExtractor):
             return
         self._claims_token = self.cache.load(self._NETRC_MACHINE, 'claims_token')
 
-    def _find_secret_formats(self, formats, video_id):
-        """ Find a valid video url and convert it to the secret variant """
-        base_format = next((f for f in formats if f.get('vcodec') != 'none'), None)
-        if not base_format:
-            return
-
-        base_url = re.sub(r'(Manifest\(.*?),filter=[\w-]+(.*?\))', r'\1\2', base_format['url'])
-        url = re.sub(r'(Manifest\(.*?),format=[\w-]+(.*?\))', r'\1\2', base_url)
-
-        secret_xml = self._download_xml(url, video_id, note='Downloading secret XML', fatal=False)
-        if not isinstance(secret_xml, xml.etree.ElementTree.Element):
-            return
-
-        for child in secret_xml:
-            if child.attrib.get('Type') != 'video':
-                continue
-            for video_quality in child:
-                bitrate = int_or_none(video_quality.attrib.get('Bitrate'))
-                if not bitrate or 'Index' not in video_quality.attrib:
-                    continue
-                height = int_or_none(video_quality.attrib.get('MaxHeight'))
-
-                yield {
-                    **base_format,
-                    'format_id': join_nonempty('sec', height),
-                    # Note: \g<1> is necessary instead of \1 since bitrate is a number
-                    'url': re.sub(r'(QualityLevels\()\d+(\))', fr'\g<1>{bitrate}\2', base_url),
-                    'width': int_or_none(video_quality.attrib.get('MaxWidth')),
-                    'tbr': bitrate / 1000.0,
-                    'height': height,
-                }
-
     def _real_extract(self, url):
         video_id = self._match_id(url)
         video_info = self._download_json(
@@ -676,7 +646,6 @@ class CBCGemIE(InfoExtractor):
         else:
             headers = {}
         m3u8_info = self._download_json(video_info['playSession']['url'], video_id, headers=headers)
-        m3u8_url = m3u8_info.get('url')
 
         if m3u8_info.get('errorCode') == 1:
             self.raise_geo_restricted(countries=['CA'])
@@ -685,9 +654,9 @@ class CBCGemIE(InfoExtractor):
         elif m3u8_info.get('errorCode') != 0:
             raise ExtractorError(f'{self.IE_NAME} said: {m3u8_info.get("errorCode")} - {m3u8_info.get("message")}')
 
-        formats = self._extract_m3u8_formats(m3u8_url, video_id, m3u8_id='hls')
+        formats = self._extract_m3u8_formats(
+            m3u8_info['url'], video_id, 'mp4', m3u8_id='hls', query={'manifestType': ''})
         self._remove_duplicate_formats(formats)
-        formats.extend(self._find_secret_formats(formats, video_id))
 
         for fmt in formats:
             if fmt.get('vcodec') == 'none':
@@ -703,20 +672,21 @@ class CBCGemIE(InfoExtractor):
 
         return {
             'id': video_id,
-            'title': video_info['title'],
-            'description': video_info.get('description'),
-            'thumbnail': video_info.get('image'),
-            'series': video_info.get('series'),
-            'season_number': video_info.get('season'),
-            'season': f'Season {video_info.get("season")}',
-            'episode_number': video_info.get('episode'),
-            'episode': video_info.get('title'),
             'episode_id': video_id,
-            'duration': video_info.get('duration'),
-            'categories': [video_info.get('category')],
             'formats': formats,
-            'release_timestamp': video_info.get('airDate'),
-            'timestamp': video_info.get('availableDate'),
+            **traverse_obj(video_info, {
+                'title': ('title', {str}),
+                'episode': ('title', {str}),
+                'description': ('description', {str}),
+                'thumbnail': ('image', {url_or_none}),
+                'series': ('series', {str}),
+                'season_number': ('season', {int_or_none}),
+                'episode_number': ('episode', {int_or_none}),
+                'duration': ('duration', {int_or_none}),
+                'categories': ('category', {str}, all),
+                'release_timestamp': ('airDate', {int_or_none(scale=1000)}),
+                'timestamp': ('availableDate', {int_or_none(scale=1000)}),
+            }),
         }
 
 

yt_dlp/extractor/mitele.py

@@ -1,14 +1,13 @@
-from .telecinco import TelecincoIE
+from .telecinco import TelecincoBaseIE
 from ..utils import (
     int_or_none,
     parse_iso8601,
 )
 
 
-class MiTeleIE(TelecincoIE):  # XXX: Do not subclass from concrete IE
+class MiTeleIE(TelecincoBaseIE):
     IE_DESC = 'mitele.es'
     _VALID_URL = r'https?://(?:www\.)?mitele\.es/(?:[^/]+/)+(?P<id>[^/]+)/player'
-
     _TESTS = [{
         'url': 'http://www.mitele.es/programas-tv/diario-de/57b0dfb9c715da65618b4afa/player',
         'info_dict': {
@@ -27,6 +26,7 @@ class MiTeleIE(TelecincoIE):  # XXX: Do not subclass from concrete IE
             'timestamp': 1471209401,
             'upload_date': '20160814',
         },
+        'skip': 'HTTP Error 404 Not Found',
     }, {
         # no explicit title
         'url': 'http://www.mitele.es/programas-tv/cuarto-milenio/57b0de3dc915da14058b4876/player',
@@ -49,6 +49,26 @@ class MiTeleIE(TelecincoIE):  # XXX: Do not subclass from concrete IE
         'params': {
             'skip_download': True,
         },
+        'skip': 'HTTP Error 404 Not Found',
+    }, {
+        'url': 'https://www.mitele.es/programas-tv/horizonte/temporada-5/programa-171-40_013480051/player/',
+        'info_dict': {
+            'id': '7adbe22e-cd41-4787-afa4-36f3da7c2c6f',
+            'ext': 'mp4',
+            'title': 'Horizonte Temporada 5 Programa 171',
+            'description': 'md5:97f1fb712c5ac27e5693a8b3c5c0c6e3',
+            'episode': 'Las Zonas de Bajas Emisiones, a debate',
+            'episode_number': 171,
+            'season': 'Season 5',
+            'season_number': 5,
+            'series': 'Horizonte',
+            'duration': 7012,
+            'upload_date': '20240927',
+            'timestamp': 1727416450,
+            'thumbnail': 'https://album.mediaset.es/eimg/2024/09/27/horizonte-171_9f02.jpg',
+            'age_limit': 12,
+        },
+        'params': {'geo_bypass_country': 'ES'},
     }, {
         'url': 'http://www.mitele.es/series-online/la-que-se-avecina/57aac5c1c915da951a8b45ed/player',
         'only_matching': True,

yt_dlp/extractor/telecinco.py

@@ -2,15 +2,69 @@ import json
 import re
 
 from .common import InfoExtractor
+from ..networking.exceptions import HTTPError
 from ..utils import (
+    ExtractorError,
     clean_html,
     int_or_none,
+    join_nonempty,
     str_or_none,
-    try_get,
+    traverse_obj,
+    update_url,
+    url_or_none,
 )
 
 
-class TelecincoIE(InfoExtractor):
+class TelecincoBaseIE(InfoExtractor):
+    def _parse_content(self, content, url):
+        video_id = content['dataMediaId']
+        config = self._download_json(
+            content['dataConfig'], video_id, 'Downloading config JSON')
+        services = config['services']
+        caronte = self._download_json(services['caronte'], video_id)
+        if traverse_obj(caronte, ('dls', 0, 'drm', {bool})):
+            self.report_drm(video_id)
+
+        stream = caronte['dls'][0]['stream']
+        headers = {
+            'Referer': url,
+            'Origin': re.match(r'https?://[^/]+', url).group(0),
+        }
+        geo_headers = {**headers, **self.geo_verification_headers()}
+
+        try:
+            cdn = self._download_json(
+                caronte['cerbero'], video_id, data=json.dumps({
+                    'bbx': caronte['bbx'],
+                    'gbx': self._download_json(services['gbx'], video_id)['gbx'],
+                }).encode(), headers={
+                    'Content-Type': 'application/json',
+                    **geo_headers,
+                })['tokens']['1']['cdn']
+        except ExtractorError as error:
+            if isinstance(error.cause, HTTPError) and error.cause.status == 403:
+                error_code = traverse_obj(
+                    self._webpage_read_content(error.cause.response, caronte['cerbero'], video_id, fatal=False),
+                    ({json.loads}, 'code', {int}))
+                if error_code == 4038:
+                    self.raise_geo_restricted(countries=['ES'])
+            raise
+
+        formats = self._extract_m3u8_formats(
+            update_url(stream, query=cdn), video_id, 'mp4', m3u8_id='hls', headers=geo_headers)
+
+        return {
+            'id': video_id,
+            'title': traverse_obj(config, ('info', 'title', {str})),
+            'formats': formats,
+            'thumbnail': (traverse_obj(content, ('dataPoster', {url_or_none}))
+                          or traverse_obj(config, 'poster', 'imageUrl', expected_type=url_or_none)),
+            'duration': traverse_obj(content, ('dataDuration', {int_or_none})),
+            'http_headers': headers,
+        }
+
+
+class TelecincoIE(TelecincoBaseIE):
     IE_DESC = 'telecinco.es, cuatro.com and mediaset.es'
     _VALID_URL = r'https?://(?:www\.)?(?:telecinco\.es|cuatro\.com|mediaset\.es)/(?:[^/]+/)+(?P<id>.+?)\.html'
 
@@ -30,6 +84,7 @@ class TelecincoIE(InfoExtractor):
                 'duration': 662,
             },
         }],
+        'skip': 'HTTP Error 410 Gone',
     }, {
         'url': 'http://www.cuatro.com/deportes/futbol/barcelona/Leo_Messi-Champions-Roma_2_2052780128.html',
         'md5': 'c86fe0d99e3bdb46b7950d38bf6ef12a',
@@ -40,23 +95,24 @@ class TelecincoIE(InfoExtractor):
             'description': 'md5:a62ecb5f1934fc787107d7b9a2262805',
             'duration': 79,
         },
+        'skip': 'Redirects to main page',
     }, {
         'url': 'http://www.mediaset.es/12meses/campanas/doylacara/conlatratanohaytrato/Ayudame-dar-cara-trata-trato_2_1986630220.html',
-        'md5': 'eddb50291df704ce23c74821b995bcac',
+        'md5': '5ce057f43f30b634fbaf0f18c71a140a',
         'info_dict': {
             'id': 'aywerkD2Sv1vGNqq9b85Q2',
             'ext': 'mp4',
             'title': '#DOYLACARA. Con la trata no hay trato',
-            'description': 'md5:2771356ff7bfad9179c5f5cd954f1477',
             'duration': 50,
+            'thumbnail': 'https://album.mediaset.es/eimg/2017/11/02/1tlQLO5Q3mtKT24f3EaC24.jpg',
         },
     }, {
         # video in opening's content
         'url': 'https://www.telecinco.es/vivalavida/fiorella-sobrina-edmundo-arrocet-entrevista_18_2907195140.html',
         'info_dict': {
-            'id': '2907195140',
+            'id': '1691427',
             'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
-            'description': 'md5:73f340a7320143d37ab895375b2bf13a',
+            'description': r're:Fiorella, la sobrina de Edmundo Arrocet, concedió .{727}',
         },
         'playlist': [{
             'md5': 'adb28c37238b675dad0f042292f209a7',
@@ -65,6 +121,7 @@ class TelecincoIE(InfoExtractor):
                 'ext': 'mp4',
                 'title': 'La surrealista entrevista a la sobrina de Edmundo Arrocet: "No puedes venir aquí y tomarnos por tontos"',
                 'duration': 1015,
+                'thumbnail': 'https://album.mediaset.es/eimg/2020/02/29/5opaC37lUhKlZ7FoDhiVC.jpg',
             },
         }],
         'params': {
@@ -81,66 +138,29 @@ class TelecincoIE(InfoExtractor):
         'only_matching': True,
     }]
 
-    def _parse_content(self, content, url):
-        video_id = content['dataMediaId']
-        config = self._download_json(
-            content['dataConfig'], video_id, 'Downloading config JSON')
-        title = config['info']['title']
-        services = config['services']
-        caronte = self._download_json(services['caronte'], video_id)
-        stream = caronte['dls'][0]['stream']
-        headers = self.geo_verification_headers()
-        headers.update({
-            'Content-Type': 'application/json;charset=UTF-8',
-            'Origin': re.match(r'https?://[^/]+', url).group(0),
-        })
-        cdn = self._download_json(
-            caronte['cerbero'], video_id, data=json.dumps({
-                'bbx': caronte['bbx'],
-                'gbx': self._download_json(services['gbx'], video_id)['gbx'],
-            }).encode(), headers=headers)['tokens']['1']['cdn']
-        formats = self._extract_m3u8_formats(
-            stream + '?' + cdn, video_id, 'mp4', 'm3u8_native', m3u8_id='hls')
-
-        return {
-            'id': video_id,
-            'title': title,
-            'formats': formats,
-            'thumbnail': content.get('dataPoster') or config.get('poster', {}).get('imageUrl'),
-            'duration': int_or_none(content.get('dataDuration')),
-        }
-
     def _real_extract(self, url):
         display_id = self._match_id(url)
         webpage = self._download_webpage(url, display_id)
-        article = self._parse_json(self._search_regex(
-            r'window\.\$REACTBASE_STATE\.article(?:_multisite)?\s*=\s*({.+})',
-            webpage, 'article'), display_id)['article']
-        title = article.get('title')
-        description = clean_html(article.get('leadParagraph')) or ''
+        article = self._search_json(
+            r'window\.\$REACTBASE_STATE\.article(?:_multisite)?\s*=',
+            webpage, 'article', display_id)['article']
+        description = traverse_obj(article, ('leadParagraph', {clean_html}, filter))
+
         if article.get('editorialType') != 'VID':
             entries = []
-            body = [article.get('opening')]
-            body.extend(try_get(article, lambda x: x['body'], list) or [])
-            for p in body:
-                if not isinstance(p, dict):
-                    continue
-                content = p.get('content')
-                if not content:
-                    continue
+
+            for p in traverse_obj(article, ((('opening', all), 'body'), lambda _, v: v['content'])):
+                content = p['content']
                 type_ = p.get('type')
-                if type_ == 'paragraph':
-                    content_str = str_or_none(content)
-                    if content_str:
-                        description += content_str
-                    continue
-                if type_ == 'video' and isinstance(content, dict):
+                if type_ == 'paragraph' and isinstance(content, str):
+                    description = join_nonempty(description, content, delim='')
+                elif type_ == 'video' and isinstance(content, dict):
                     entries.append(self._parse_content(content, url))
+
             return self.playlist_result(
-                entries, str_or_none(article.get('id')), title, description)
-        content = article['opening']['content']
-        info = self._parse_content(content, url)
-        info.update({
-            'description': description,
-        })
+                entries, str_or_none(article.get('id')),
+                traverse_obj(article, ('title', {str})), clean_html(description))
+
+        info = self._parse_content(article['opening']['content'], url)
+        info['description'] = description
         return info

yt_dlp/extractor/youtube.py

@@ -114,6 +114,7 @@ INNERTUBE_CLIENTS = {
         },
         'INNERTUBE_CONTEXT_CLIENT_NAME': 67,
     },
+    # This client now requires sign-in for every video
     'web_creator': {
         'INNERTUBE_CONTEXT': {
             'client': {
@@ -153,6 +154,7 @@ INNERTUBE_CLIENTS = {
         'REQUIRE_JS_PLAYER': False,
         'REQUIRE_PO_TOKEN': True,
     },
+    # This client now requires sign-in for every video
     'android_creator': {
         'INNERTUBE_CONTEXT': {
             'client': {
@@ -200,21 +202,6 @@ INNERTUBE_CLIENTS = {
         'REQUIRE_JS_PLAYER': False,
         'PLAYER_PARAMS': '2AMB',
     },
-    # This client only has legacy formats and storyboards
-    'android_producer': {
-        'INNERTUBE_CONTEXT': {
-            'client': {
-                'clientName': 'ANDROID_PRODUCER',
-                'clientVersion': '0.111.1',
-                'androidSdkVersion': 30,
-                'userAgent': 'com.google.android.apps.youtube.producer/0.111.1 (Linux; U; Android 11) gzip',
-                'osName': 'Android',
-                'osVersion': '11',
-            },
-        },
-        'INNERTUBE_CONTEXT_CLIENT_NAME': 91,
-        'REQUIRE_JS_PLAYER': False,
-    },
     # iOS clients have HLS live streams. Setting device model to get 60fps formats.
     # See: https://github.com/TeamNewPipe/NewPipeExtractor/issues/680#issuecomment-1002724558
     'ios': {
@@ -247,6 +234,7 @@ INNERTUBE_CLIENTS = {
         'INNERTUBE_CONTEXT_CLIENT_NAME': 26,
         'REQUIRE_JS_PLAYER': False,
     },
+    # This client now requires sign-in for every video
     'ios_creator': {
         'INNERTUBE_CONTEXT': {
             'client': {
@@ -282,8 +270,9 @@ INNERTUBE_CLIENTS = {
         },
         'INNERTUBE_CONTEXT_CLIENT_NAME': 7,
     },
-    # This client can access age restricted videos (unless the uploader has disabled the 'allow embedding' option)
-    # See: https://github.com/zerodytrash/YouTube-Internal-Clients
+    # This client now requires sign-in for every video
+    # It was previously an age-gate workaround for videos that were `playable_in_embed`
+    # It may still be useful if signed into an EU account that is not age-verified
     'tv_embedded': {
         'INNERTUBE_CONTEXT': {
             'client': {
@@ -1525,6 +1514,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'heatmap': 'count:100',
                 'timestamp': 1401991663,
             },
+            'skip': 'Age-restricted; requires authentication',
         },
         {
             'note': 'Age-gate video with embed allowed in public site',
@@ -1555,6 +1545,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'comment_count': int,
                 'channel_is_verified': True,
             },
+            'skip': 'Age-restricted; requires authentication',
         },
         {
             'note': 'Age-gate video embedable only with clientScreen=EMBED',
@@ -1585,6 +1576,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'uploader_id': '@ProjektMelody',
                 'timestamp': 1577508724,
             },
+            'skip': 'Age-restricted; requires authentication',
         },
         {
             'note': 'Non-Agegated non-embeddable video',
@@ -2356,6 +2348,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'channel_is_verified': True,
                 'timestamp': 1405513526,
             },
+            'skip': 'Age-restricted; requires authentication',
         },
         {
             # restricted location, https://github.com/ytdl-org/youtube-dl/issues/28685
@@ -2726,6 +2719,7 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 'timestamp': 1577508724,
             },
             'params': {'extractor_args': {'youtube': {'player_client': ['tv_embedded']}}, 'format': '251-drc'},
+            'skip': 'Age-restricted; requires authentication',
         },
         {
             'url': 'https://www.youtube.com/live/qVv6vCqciTM',
@@ -3982,26 +3976,15 @@ class YoutubeIE(YoutubeBaseInfoExtractor):
                 else:
                     prs.append(pr)
 
-            # tv_embedded can work around age-gate and age-verification IF the video is embeddable
-            if self._is_agegated(pr) and variant != 'tv_embedded':
-                append_client(f'tv_embedded.{base_client}')
-
-            # Unauthenticated users will only get tv_embedded client formats if age-gated
-            if self._is_agegated(pr) and not self.is_authenticated:
-                self.to_screen(
-                    f'{video_id}: This video is age-restricted; some formats may be missing '
-                    f'without authentication. {self._login_hint()}', only_once=True)
-
             # EU countries require age-verification for accounts to access age-restricted videos
             # If account is not age-verified, _is_agegated() will be truthy for non-embedded clients
-            # If embedding is disabled for the video, _is_unplayable() will be truthy for tv_embedded
-            embedding_is_disabled = variant == 'tv_embedded' and self._is_unplayable(pr)
-            if self.is_authenticated and (self._is_agegated(pr) or embedding_is_disabled):
+            if self.is_authenticated and self._is_agegated(pr):
                 self.to_screen(
                     f'{video_id}: This video is age-restricted and YouTube is requiring '
                     'account age-verification; some formats may be missing', only_once=True)
                 # web_creator and mediaconnect can work around the age-verification requirement
-                # _producer, _testsuite, & _vr variants can also work around age-verification
+                # _testsuite & _vr variants can also work around age-verification
+                # tv_embedded may(?) still work around age-verification if the video is embeddable
                 append_client('web_creator', 'mediaconnect')
 
         prs.extend(deprioritized_prs)