[ie/NovaEmbed] Fix extractor (#7910)

Closes #8025
Authored by: std-move

yt_dlp/compat/compat_utils.py

@@ -15,7 +15,7 @@ def get_package_info(module):
         name=getattr(module, '_yt_dlp__identifier', module.__name__),
         version=str(next(filter(None, (
             getattr(module, attr, None)
-            for attr in ('__version__', 'version_string', 'version')
+            for attr in ('_yt_dlp__version', '__version__', 'version_string', 'version')
         )), None)))
 
 

yt_dlp/dependencies/__init__.py

@@ -43,6 +43,8 @@ except Exception as _err:
 
 try:
     import sqlite3
+    # We need to get the underlying `sqlite` version, see https://github.com/yt-dlp/yt-dlp/issues/8152
+    sqlite3._yt_dlp__version = sqlite3.sqlite_version
 except ImportError:
     # although sqlite3 is part of the standard library, it is possible to compile python without
     # sqlite support. See: https://github.com/yt-dlp/yt-dlp/issues/544

yt_dlp/extractor/_extractors.py

@@ -296,9 +296,11 @@ from .cammodels import CamModelsIE
 from .camsoda import CamsodaIE
 from .camtasia import CamtasiaEmbedIE
 from .camwithher import CamWithHerIE
+from .canal1 import Canal1IE
 from .canalalpha import CanalAlphaIE
 from .canalplus import CanalplusIE
 from .canalc2 import Canalc2IE
+from .caracoltv import CaracolTvPlayIE
 from .carambatv import (
     CarambaTVIE,
     CarambaTVPageIE,
@@ -565,6 +567,7 @@ from .epicon import (
     EpiconIE,
     EpiconSeriesIE,
 )
+from .eplus import EplusIbIE
 from .epoch import EpochIE
 from .eporner import EpornerIE
 from .eroprofile import (
@@ -1598,6 +1601,7 @@ from .rbmaradio import RBMARadioIE
 from .rbgtum import (
     RbgTumIE,
     RbgTumCourseIE,
+    RbgTumNewCourseIE,
 )
 from .rcs import (
     RCSIE,

yt_dlp/extractor/bilibili.py

@@ -49,14 +49,14 @@ class BilibiliBaseIE(InfoExtractor):
             for r in traverse_obj(play_info, ('support_formats', lambda _, v: v['quality']))
         }
 
-        audios = traverse_obj(play_info, ('dash', 'audio', ...))
+        audios = traverse_obj(play_info, ('dash', (None, 'dolby'), 'audio', ..., {dict}))
         flac_audio = traverse_obj(play_info, ('dash', 'flac', 'audio'))
         if flac_audio:
             audios.append(flac_audio)
         formats = [{
             'url': traverse_obj(audio, 'baseUrl', 'base_url', 'url'),
             'ext': mimetype2ext(traverse_obj(audio, 'mimeType', 'mime_type')),
-            'acodec': audio.get('codecs'),
+            'acodec': traverse_obj(audio, ('codecs', {str.lower})),
             'vcodec': 'none',
             'tbr': float_or_none(audio.get('bandwidth'), scale=1000),
             'filesize': int_or_none(audio.get('size')),
@@ -71,6 +71,7 @@ class BilibiliBaseIE(InfoExtractor):
             'height': int_or_none(video.get('height')),
             'vcodec': video.get('codecs'),
             'acodec': 'none' if audios else None,
+            'dynamic_range': {126: 'DV', 125: 'HDR10'}.get(int_or_none(video.get('id'))),
             'tbr': float_or_none(video.get('bandwidth'), scale=1000),
             'filesize': int_or_none(video.get('size')),
             'quality': int_or_none(video.get('id')),

yt_dlp/extractor/canal1.py

@@ -0,0 +1,39 @@
+from .common import InfoExtractor
+
+
+class Canal1IE(InfoExtractor):
+    _VALID_URL = r'https?://(?:www\.|noticias\.)?canal1\.com\.co/(?:[^?#&])+/(?P<id>[\w-]+)'
+
+    _TESTS = [{
+        'url': 'https://canal1.com.co/noticias/napa-i-una-cadena-de-produccion-de-arroz-que-se-quedo-en-veremos-y-abandonada-en-el-departamento-del-choco/',
+        'info_dict': {
+            'id': '63b39f6b354977084b85ab54',
+            'display_id': 'napa-i-una-cadena-de-produccion-de-arroz-que-se-quedo-en-veremos-y-abandonada-en-el-departamento-del-choco',
+            'title': 'Ñapa I Una cadena de producción de arroz que se quedó en veremos y abandonada en el departamento del Chocó',
+            'description': 'md5:bc49c6d64d20610ea1e7daf079a0d013',
+            'thumbnail': r're:^https?://[^?#]+63b39f6b354977084b85ab54',
+            'ext': 'mp4',
+        },
+    }, {
+        'url': 'https://noticias.canal1.com.co/noticias/tres-i-el-triste-record-que-impuso-elon-musk-el-dueno-de-tesla-y-de-twitter/',
+        'info_dict': {
+            'id': '63b39e93f5fd223aa32250fb',
+            'display_id': 'tres-i-el-triste-record-que-impuso-elon-musk-el-dueno-de-tesla-y-de-twitter',
+            'title': 'Tres I El triste récord que impuso Elon Musk, el dueño de Tesla y de Twitter',
+            'description': 'md5:d9f691f131a21ce6767ca6c05d17d791',
+            'thumbnail': r're:^https?://[^?#]+63b39e93f5fd223aa32250fb',
+            'ext': 'mp4',
+        },
+    }, {
+        # Geo-restricted to Colombia
+        'url': 'https://canal1.com.co/programas/guerreros-canal-1/video-inedito-guerreros-despedida-kewin-zarate/',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        webpage = self._download_webpage(url, display_id)
+
+        return self.url_result(
+            self._search_regex(r'"embedUrl"\s*:\s*"([^"]+)', webpage, 'embed url'),
+            display_id=display_id, url_transparent=True)

yt_dlp/extractor/caracoltv.py

@@ -0,0 +1,136 @@
+import base64
+import json
+import uuid
+
+from .common import InfoExtractor
+from ..utils import (
+    int_or_none,
+    js_to_json,
+    traverse_obj,
+    urljoin,
+)
+
+
+class CaracolTvPlayIE(InfoExtractor):
+    _VALID_URL = r'https?://play\.caracoltv\.com/videoDetails/(?P<id>[^/?#]+)'
+    _NETRC_MACHINE = 'caracoltv-play'
+
+    _TESTS = [{
+        'url': 'https://play.caracoltv.com/videoDetails/OTo4NGFmNjUwOWQ2ZmM0NTg2YWRiOWU0MGNhOWViOWJkYQ==',
+        'info_dict': {
+            'id': 'OTo4NGFmNjUwOWQ2ZmM0NTg2YWRiOWU0MGNhOWViOWJkYQ==',
+            'title': 'La teoría del promedio',
+            'description': 'md5:1cdd6d2c13f19ef0d9649ab81a023ac3',
+        },
+        'playlist_count': 6,
+    }, {
+        'url': 'https://play.caracoltv.com/videoDetails/OTo3OWM4ZTliYzQxMmM0MTMxYTk4Mjk2YjdjNGQ4NGRkOQ==/ella?season=0',
+        'info_dict': {
+            'id': 'OTo3OWM4ZTliYzQxMmM0MTMxYTk4Mjk2YjdjNGQ4NGRkOQ==',
+            'title': 'Ella',
+            'description': 'md5:a639b1feb5ddcc0cff92a489b4e544b8',
+        },
+        'playlist_count': 10,
+    }, {
+        'url': 'https://play.caracoltv.com/videoDetails/OTpiYTY1YTVmOTI5MzI0ZWJhOGZiY2Y3MmRlOWZlYmJkOA==/la-vuelta-al-mundo-en-80-risas-2022?season=0',
+        'info_dict': {
+            'id': 'OTpiYTY1YTVmOTI5MzI0ZWJhOGZiY2Y3MmRlOWZlYmJkOA==',
+            'title': 'La vuelta al mundo en 80 risas 2022',
+            'description': 'md5:e97aac36106e5c37ebf947b3350106a4',
+        },
+        'playlist_count': 17,
+    }, {
+        'url': 'https://play.caracoltv.com/videoDetails/MzoxX3BwbjRmNjB1',
+        'only_matching': True,
+    }]
+
+    _USER_TOKEN = None
+
+    def _extract_app_token(self, webpage):
+        config_js_path = self._search_regex(
+            r'<script[^>]+src\s*=\s*"([^"]+coreConfig.js[^"]+)', webpage, 'config js url', fatal=False)
+
+        mediation_config = {} if not config_js_path else self._search_json(
+            r'mediation\s*:', self._download_webpage(
+                urljoin('https://play.caracoltv.com/', config_js_path), None, fatal=False, note='Extracting JS config'),
+            'mediation_config', None, transform_source=js_to_json, fatal=False)
+
+        key = traverse_obj(
+            mediation_config, ('live', 'key')) or '795cd9c089a1fc48094524a5eba85a3fca1331817c802f601735907c8bbb4f50'
+        secret = traverse_obj(
+            mediation_config, ('live', 'secret')) or '64dec00a6989ba83d087621465b5e5d38bdac22033b0613b659c442c78976fa0'
+
+        return base64.b64encode(f'{key}:{secret}'.encode()).decode()
+
+    def _perform_login(self, email, password):
+        webpage = self._download_webpage('https://play.caracoltv.com/', None, fatal=False)
+        app_token = self._extract_app_token(webpage)
+
+        bearer_token = self._download_json(
+            'https://eu-gateway.inmobly.com/applications/oauth', None, data=b'', note='Retrieving bearer token',
+            headers={'Authorization': f'Basic {app_token}'})['token']
+
+        self._USER_TOKEN = self._download_json(
+            'https://eu-gateway.inmobly.com/user/login', None, note='Performing login', headers={
+                'Content-Type': 'application/json',
+                'Authorization': f'Bearer {bearer_token}',
+            }, data=json.dumps({
+                'device_data': {
+                    'device_id': str(uuid.uuid4()),
+                    'device_token': '',
+                    'device_type': 'web'
+                },
+                'login_data': {
+                    'enabled': True,
+                    'email': email,
+                    'password': password,
+                }
+            }).encode())['user_token']
+
+    def _extract_video(self, video_data, series_id=None, season_id=None, season_number=None):
+        formats, subtitles = self._extract_m3u8_formats_and_subtitles(video_data['stream_url'], series_id, 'mp4')
+
+        return {
+            'id': video_data['id'],
+            'title': video_data.get('name'),
+            'description': video_data.get('description'),
+            'formats': formats,
+            'subtitles': subtitles,
+            'thumbnails': traverse_obj(
+                video_data, ('extra_thumbs', ..., {'url': 'thumb_url', 'height': 'height', 'width': 'width'})),
+            'series_id': series_id,
+            'season_id': season_id,
+            'season_number': int_or_none(season_number),
+            'episode_number': int_or_none(video_data.get('item_order')),
+            'is_live': video_data.get('entry_type') == 3,
+        }
+
+    def _extract_series_seasons(self, seasons, series_id):
+        for season in seasons:
+            api_response = self._download_json(
+                'https://eu-gateway.inmobly.com/feed', series_id, query={'season_id': season['id']},
+                headers={'Authorization': f'Bearer {self._USER_TOKEN}'})
+
+            season_number = season.get('order')
+            for episode in api_response['items']:
+                yield self._extract_video(episode, series_id, season['id'], season_number)
+
+    def _real_extract(self, url):
+        series_id = self._match_id(url)
+
+        if self._USER_TOKEN is None:
+            self._perform_login('guest@inmobly.com', 'Test@gus1')
+
+        api_response = self._download_json(
+            'https://eu-gateway.inmobly.com/feed', series_id, query={'include_ids': series_id},
+            headers={'Authorization': f'Bearer {self._USER_TOKEN}'})['items'][0]
+
+        if not api_response.get('seasons'):
+            return self._extract_video(api_response)
+
+        return self.playlist_result(
+            self._extract_series_seasons(api_response['seasons'], series_id),
+            series_id, **traverse_obj(api_response, {
+                'title': 'name',
+                'description': 'description',
+            }))

yt_dlp/extractor/douyutv.py

@@ -1,31 +1,72 @@
 import time
 import hashlib
-import re
 import urllib
+import uuid
 
 from .common import InfoExtractor
+from .openload import PhantomJSwrapper
 from ..utils import (
     ExtractorError,
+    UserNotLive,
+    determine_ext,
+    int_or_none,
+    js_to_json,
+    parse_resolution,
+    str_or_none,
+    traverse_obj,
     unescapeHTML,
-    unified_strdate,
+    url_or_none,
+    urlencode_postdata,
     urljoin,
 )
 
 
-class DouyuTVIE(InfoExtractor):
+class DouyuBaseIE(InfoExtractor):
-    IE_DESC = '斗鱼'
+    def _download_cryptojs_md5(self, video_id):
+        for url in [
+            'https://cdnjs.cloudflare.com/ajax/libs/crypto-js/3.1.2/rollups/md5.js',
+            'https://cdn.bootcdn.net/ajax/libs/crypto-js/3.1.2/rollups/md5.js',
+        ]:
+            js_code = self._download_webpage(
+                url, video_id, note='Downloading signing dependency', fatal=False)
+            if js_code:
+                self.cache.store('douyu', 'crypto-js-md5', js_code)
+                return js_code
+        raise ExtractorError('Unable to download JS dependency (crypto-js/md5)')
+
+    def _get_cryptojs_md5(self, video_id):
+        return self.cache.load('douyu', 'crypto-js-md5') or self._download_cryptojs_md5(video_id)
+
+    def _calc_sign(self, sign_func, video_id, a):
+        b = uuid.uuid4().hex
+        c = round(time.time())
+        js_script = f'{self._get_cryptojs_md5(video_id)};{sign_func};console.log(ub98484234("{a}","{b}","{c}"))'
+        phantom = PhantomJSwrapper(self)
+        result = phantom.execute(js_script, video_id,
+                                 note='Executing JS signing script').strip()
+        return {i: v[0] for i, v in urllib.parse.parse_qs(result).items()}
+
+    def _search_js_sign_func(self, webpage, fatal=True):
+        # The greedy look-behind ensures last possible script tag is matched
+        return self._search_regex(
+            r'(?:<script.*)?<script[^>]*>(.*?ub98484234.*?)</script>', webpage, 'JS sign func', fatal=fatal)
+
+
+class DouyuTVIE(DouyuBaseIE):
+    IE_DESC = '斗鱼直播'
     _VALID_URL = r'https?://(?:www\.)?douyu(?:tv)?\.com/(topic/\w+\?rid=|(?:[^/]+/))*(?P<id>[A-Za-z0-9]+)'
     _TESTS = [{
-        'url': 'http://www.douyutv.com/iseven',
+        'url': 'https://www.douyu.com/pigff',
         'info_dict': {
-            'id': '17732',
+            'id': '24422',
-            'display_id': 'iseven',
+            'display_id': 'pigff',
-            'ext': 'flv',
+            'ext': 'mp4',
-            'title': 're:^清晨醒脑！根本停不下来！ [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
+            'title': 're:^【PIGFF】.* [0-9]{4}-[0-9]{2}-[0-9]{2} [0-9]{2}:[0-9]{2}$',
-            'description': r're:.*m7show@163\.com.*',
+            'description': r'≥15级牌子看鱼吧置顶帖进粉丝vx群',
-            'thumbnail': r're:^https?://.*\.png',
+            'thumbnail': str,
-            'uploader': '7师傅',
+            'uploader': 'pigff',
             'is_live': True,
+            'live_status': 'is_live',
         },
         'params': {
             'skip_download': True,
@@ -85,15 +126,43 @@ class DouyuTVIE(InfoExtractor):
         'only_matching': True,
     }]
 
+    def _get_sign_func(self, room_id, video_id):
+        return self._download_json(
+            f'https://www.douyu.com/swf_api/homeH5Enc?rids={room_id}', video_id,
+            note='Getting signing script')['data'][f'room{room_id}']
+
+    def _extract_stream_formats(self, stream_formats):
+        formats = []
+        for stream_info in traverse_obj(stream_formats, (..., 'data')):
+            stream_url = urljoin(
+                traverse_obj(stream_info, 'rtmp_url'), traverse_obj(stream_info, 'rtmp_live'))
+            if stream_url:
+                rate_id = traverse_obj(stream_info, ('rate', {int_or_none}))
+                rate_info = traverse_obj(stream_info, ('multirates', lambda _, v: v['rate'] == rate_id), get_all=False)
+                ext = determine_ext(stream_url)
+                formats.append({
+                    'url': stream_url,
+                    'format_id': str_or_none(rate_id),
+                    'ext': 'mp4' if ext == 'm3u8' else ext,
+                    'protocol': 'm3u8_native' if ext == 'm3u8' else 'https',
+                    'quality': rate_id % -10000 if rate_id is not None else None,
+                    **traverse_obj(rate_info, {
+                        'format': ('name', {str_or_none}),
+                        'tbr': ('bit', {int_or_none}),
+                    }),
+                })
+        return formats
+
     def _real_extract(self, url):
         video_id = self._match_id(url)
 
-        if video_id.isdigit():
+        webpage = self._download_webpage(url, video_id)
-            room_id = video_id
+        room_id = self._search_regex(r'\$ROOM\.room_id\s*=\s*(\d+)', webpage, 'room id')
-        else:
+
-            page = self._download_webpage(url, video_id)
+        if self._search_regex(r'"videoLoop"\s*:\s*(\d+)', webpage, 'loop', default='') == '1':
-            room_id = self._html_search_regex(
+            raise UserNotLive('The channel is auto-playing VODs', video_id=video_id)
-                r'"room_id\\?"\s*:\s*(\d+),', page, 'room id')
+        if self._search_regex(r'\$ROOM\.show_status\s*=\s*(\d+)', webpage, 'status', default='') == '2':
+            raise UserNotLive(video_id=video_id)
 
         # Grab metadata from API
         params = {
@@ -102,110 +171,136 @@ class DouyuTVIE(InfoExtractor):
             'time': int(time.time()),
         }
         params['auth'] = hashlib.md5(
-            f'room/{video_id}?{urllib.parse.urlencode(params)}zNzMV1y4EMxOHS6I5WKm'.encode()).hexdigest()
+            f'room/{room_id}?{urllib.parse.urlencode(params)}zNzMV1y4EMxOHS6I5WKm'.encode()).hexdigest()
-        room = self._download_json(
+        room = traverse_obj(self._download_json(
             f'http://www.douyutv.com/api/v1/room/{room_id}', video_id,
-            note='Downloading room info', query=params)['data']
+            note='Downloading room info', query=params, fatal=False), 'data')
 
         # 1 = live, 2 = offline
-        if room.get('show_status') == '2':
+        if traverse_obj(room, 'show_status') == '2':
-            raise ExtractorError('Live stream is offline', expected=True)
+            raise UserNotLive(video_id=video_id)
 
-        video_url = urljoin('https://hls3-akm.douyucdn.cn/', self._search_regex(r'(live/.*)', room['hls_url'], 'URL'))
+        js_sign_func = self._search_js_sign_func(webpage, fatal=False) or self._get_sign_func(room_id, video_id)
-        formats, subs = self._extract_m3u8_formats_and_subtitles(video_url, room_id)
+        form_data = {
+            'rate': 0,
+            **self._calc_sign(js_sign_func, video_id, room_id),
+        }
+        stream_formats = [self._download_json(
+            f'https://www.douyu.com/lapi/live/getH5Play/{room_id}',
+            video_id, note="Downloading livestream format",
+            data=urlencode_postdata(form_data))]
 
-        title = unescapeHTML(room['room_name'])
+        for rate_id in traverse_obj(stream_formats[0], ('data', 'multirates', ..., 'rate')):
-        description = room.get('show_details')
+            if rate_id != traverse_obj(stream_formats[0], ('data', 'rate')):
-        thumbnail = room.get('room_src')
+                form_data['rate'] = rate_id
-        uploader = room.get('nickname')
+                stream_formats.append(self._download_json(
+                    f'https://www.douyu.com/lapi/live/getH5Play/{room_id}',
+                    video_id, note=f'Downloading livestream format {rate_id}',
+                    data=urlencode_postdata(form_data)))
 
         return {
             'id': room_id,
-            'display_id': video_id,
+            'formats': self._extract_stream_formats(stream_formats),
-            'title': title,
-            'description': description,
-            'thumbnail': thumbnail,
-            'uploader': uploader,
             'is_live': True,
-            'subtitles': subs,
+            **traverse_obj(room, {
-            'formats': formats,
+                'display_id': ('url', {str}, {lambda i: i[1:]}),
+                'title': ('room_name', {unescapeHTML}),
+                'description': ('show_details', {str}),
+                'uploader': ('nickname', {str}),
+                'thumbnail': ('room_src', {url_or_none}),
+            })
         }
 
 
-class DouyuShowIE(InfoExtractor):
+class DouyuShowIE(DouyuBaseIE):
     _VALID_URL = r'https?://v(?:mobile)?\.douyu\.com/show/(?P<id>[0-9a-zA-Z]+)'
 
     _TESTS = [{
-        'url': 'https://v.douyu.com/show/rjNBdvnVXNzvE2yw',
+        'url': 'https://v.douyu.com/show/mPyq7oVNe5Yv1gLY',
-        'md5': '0c2cfd068ee2afe657801269b2d86214',
         'info_dict': {
-            'id': 'rjNBdvnVXNzvE2yw',
+            'id': 'mPyq7oVNe5Yv1gLY',
             'ext': 'mp4',
-            'title': '陈一发儿：砒霜 我有个室友系列！04-01 22点场',
+            'title': '四川人小时候的味道“蒜苗回锅肉”，传统菜不能丢，要常做来吃',
-            'duration': 7150.08,
+            'duration': 633,
-            'thumbnail': r're:^https?://.*\.jpg$',
+            'thumbnail': str,
-            'uploader': '陈一发儿',
+            'uploader': '美食作家王刚V',
-            'uploader_id': 'XrZwYelr5wbK',
+            'uploader_id': 'OVAO4NVx1m7Q',
-            'uploader_url': 'https://v.douyu.com/author/XrZwYelr5wbK',
+            'timestamp': 1661850002,
-            'upload_date': '20170402',
+            'upload_date': '20220830',
+            'view_count': int,
+            'tags': ['美食', '美食综合'],
         },
     }, {
         'url': 'https://vmobile.douyu.com/show/rjNBdvnVXNzvE2yw',
         'only_matching': True,
     }]
 
+    _FORMATS = {
+        'super': '原画',
+        'high': '超清',
+        'normal': '高清',
+    }
+
+    _QUALITIES = {
+        'super': -1,
+        'high': -2,
+        'normal': -3,
+    }
+
+    _RESOLUTIONS = {
+        'super': '1920x1080',
+        'high': '1280x720',
+        'normal': '852x480',
+    }
+
     def _real_extract(self, url):
         url = url.replace('vmobile.', 'v.')
         video_id = self._match_id(url)
 
         webpage = self._download_webpage(url, video_id)
 
-        room_info = self._parse_json(self._search_regex(
+        video_info = self._search_json(
-            r'var\s+\$ROOM\s*=\s*({.+});', webpage, 'room info'), video_id)
+            r'<script>\s*window\.\$DATA\s*=', webpage,
+            'video info', video_id, transform_source=js_to_json)
 
-        video_info = None
+        js_sign_func = self._search_js_sign_func(webpage)
+        form_data = {
+            'vid': video_id,
+            **self._calc_sign(js_sign_func, video_id, video_info['ROOM']['point_id']),
+        }
+        url_info = self._download_json(
+            'https://v.douyu.com/api/stream/getStreamUrl', video_id,
+            data=urlencode_postdata(form_data), note="Downloading video formats")
 
-        for trial in range(5):
+        formats = []
-            # Sometimes Douyu rejects our request. Let's try it more times
+        for name, url in traverse_obj(url_info, ('data', 'thumb_video', {dict.items}, ...)):
-            try:
+            video_url = traverse_obj(url, ('url', {url_or_none}))
-                video_info = self._download_json(
+            if video_url:
-                    'https://vmobile.douyu.com/video/getInfo', video_id,
+                ext = determine_ext(video_url)
-                    query={'vid': video_id},
+                formats.append({
-                    headers={
+                    'format': self._FORMATS.get(name),
-                        'Referer': url,
+                    'format_id': name,
-                        'x-requested-with': 'XMLHttpRequest',
+                    'url': video_url,
+                    'quality': self._QUALITIES.get(name),
+                    'ext': 'mp4' if ext == 'm3u8' else ext,
+                    'protocol': 'm3u8_native' if ext == 'm3u8' else 'https',
+                    **parse_resolution(self._RESOLUTIONS.get(name))
                 })
-                break
+            else:
-            except ExtractorError:
+                self.to_screen(
-                self._sleep(1, video_id)
+                    f'"{self._FORMATS.get(name, name)}" format may require logging in. {self._login_hint()}')
-
-        if not video_info:
-            raise ExtractorError('Can\'t fetch video info')
-
-        formats = self._extract_m3u8_formats(
-            video_info['data']['video_url'], video_id,
-            entry_protocol='m3u8_native', ext='mp4')
-
-        upload_date = unified_strdate(self._html_search_regex(
-            r'<em>上传时间：</em><span>([^<]+)</span>', webpage,
-            'upload date', fatal=False))
-
-        uploader = uploader_id = uploader_url = None
-        mobj = re.search(
-            r'(?m)<a[^>]+href="/author/([0-9a-zA-Z]+)".+?<strong[^>]+title="([^"]+)"',
-            webpage)
-        if mobj:
-            uploader_id, uploader = mobj.groups()
-            uploader_url = urljoin(url, '/author/' + uploader_id)
 
         return {
             'id': video_id,
-            'title': room_info['name'],
             'formats': formats,
-            'duration': room_info.get('duration'),
+            **traverse_obj(video_info, ('DATA', {
-            'thumbnail': room_info.get('pic'),
+                'title': ('content', 'title', {str}),
-            'upload_date': upload_date,
+                'uploader': ('content', 'author', {str}),
-            'uploader': uploader,
+                'uploader_id': ('content', 'up_id', {str_or_none}),
-            'uploader_id': uploader_id,
+                'duration': ('content', 'video_duration', {int_or_none}),
-            'uploader_url': uploader_url,
+                'thumbnail': ('content', 'video_pic', {url_or_none}),
+                'timestamp': ('content', 'create_time', {int_or_none}),
+                'view_count': ('content', 'view_num', {int_or_none}),
+                'tags': ('videoTag', ..., 'tagName', {str}),
+            }))
         }

yt_dlp/extractor/eplus.py

@@ -0,0 +1,96 @@
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    try_call,
+    unified_timestamp,
+)
+
+
+class EplusIbIE(InfoExtractor):
+    IE_NAME = 'eplus:inbound'
+    IE_DESC = 'e+ (イープラス) overseas'
+    _VALID_URL = r'https?://live\.eplus\.jp/ex/player\?ib=(?P<id>(?:\w|%2B|%2F){86}%3D%3D)'
+    _TESTS = [{
+        'url': 'https://live.eplus.jp/ex/player?ib=YEFxb3Vyc2Dombnjg7blkrLlrablnJLjgrnjgq%2Fjg7zjg6vjgqLjgqTjg4njg6vlkIzlpb3kvJpgTGllbGxhIQ%3D%3D',
+        'info_dict': {
+            'id': '354502-0001-002',
+            'title': 'LoveLive!Series Presents COUNTDOWN LoveLive! 2021→2022～LIVE with a smile!～【Streaming+(配信)】',
+            'live_status': 'was_live',
+            'release_date': '20211231',
+            'release_timestamp': 1640952000,
+            'description': str,
+        },
+        'params': {
+            'skip_download': True,
+            'ignore_no_formats_error': True,
+        },
+        'expected_warnings': [
+            'Could not find the playlist URL. This event may not be accessible',
+            'No video formats found!',
+            'Requested format is not available',
+        ],
+    }]
+
+    def _real_extract(self, url):
+        video_id = self._match_id(url)
+        webpage = self._download_webpage(url, video_id)
+
+        data_json = self._search_json(r'<script>\s*var app\s*=', webpage, 'data json', video_id)
+
+        delivery_status = data_json.get('delivery_status')
+        archive_mode = data_json.get('archive_mode')
+        release_timestamp = try_call(lambda: unified_timestamp(data_json['event_datetime']) - 32400)
+        release_timestamp_str = data_json.get('event_datetime_text')  # JST
+
+        self.write_debug(f'delivery_status = {delivery_status}, archive_mode = {archive_mode}')
+
+        if delivery_status == 'PREPARING':
+            live_status = 'is_upcoming'
+        elif delivery_status == 'STARTED':
+            live_status = 'is_live'
+        elif delivery_status == 'STOPPED':
+            if archive_mode != 'ON':
+                raise ExtractorError(
+                    'This event has ended and there is no archive for this event', expected=True)
+            live_status = 'post_live'
+        elif delivery_status == 'WAIT_CONFIRM_ARCHIVED':
+            live_status = 'post_live'
+        elif delivery_status == 'CONFIRMED_ARCHIVE':
+            live_status = 'was_live'
+        else:
+            self.report_warning(f'Unknown delivery_status {delivery_status}, treat it as a live')
+            live_status = 'is_live'
+
+        formats = []
+
+        m3u8_playlist_urls = self._search_json(
+            r'var listChannels\s*=', webpage, 'hls URLs', video_id, contains_pattern=r'\[.+\]', default=[])
+        if not m3u8_playlist_urls:
+            if live_status == 'is_upcoming':
+                self.raise_no_formats(
+                    f'Could not find the playlist URL. This live event will begin at {release_timestamp_str} JST', expected=True)
+            else:
+                self.raise_no_formats(
+                    'Could not find the playlist URL. This event may not be accessible', expected=True)
+        elif live_status == 'is_upcoming':
+            self.raise_no_formats(f'This live event will begin at {release_timestamp_str} JST', expected=True)
+        elif live_status == 'post_live':
+            self.raise_no_formats('This event has ended, and the archive will be available shortly', expected=True)
+        else:
+            for m3u8_playlist_url in m3u8_playlist_urls:
+                formats.extend(self._extract_m3u8_formats(m3u8_playlist_url, video_id))
+            # FIXME: HTTP request headers need to be updated to continue download
+            warning = 'Due to technical limitations, the download will be interrupted after one hour'
+            if live_status == 'is_live':
+                self.report_warning(warning)
+            elif live_status == 'was_live':
+                self.report_warning(f'{warning}. You can restart to continue the download')
+
+        return {
+            'id': data_json['app_id'],
+            'title': data_json.get('app_name'),
+            'formats': formats,
+            'live_status': live_status,
+            'description': data_json.get('content'),
+            'release_timestamp': release_timestamp,
+        }

yt_dlp/extractor/expressen.py

@@ -11,8 +11,8 @@ class ExpressenIE(InfoExtractor):
     _VALID_URL = r'''(?x)
                     https?://
                         (?:www\.)?(?:expressen|di)\.se/
-                        (?:(?:tvspelare/video|videoplayer/embed)/)?
+                        (?:(?:tvspelare/video|video-?player/embed)/)?
-                        tv/(?:[^/]+/)*
+                        (?:tv|nyheter)/(?:[^/?#]+/)*
                         (?P<id>[^/?#&]+)
                     '''
     _EMBED_REGEX = [r'<iframe[^>]+\bsrc=(["\'])(?P<url>(?:https?:)?//(?:www\.)?(?:expressen|di)\.se/(?:tvspelare/video|videoplayer/embed)/tv/.+?)\1']
@@ -42,6 +42,12 @@ class ExpressenIE(InfoExtractor):
     }, {
         'url': 'https://www.di.se/videoplayer/embed/tv/ditv/borsmorgon/implantica-rusar-70--under-borspremiaren-hor-styrelsemedlemmen/?embed=true&external=true&autoplay=true&startVolume=0&partnerId=di',
         'only_matching': True,
+    }, {
+        'url': 'https://www.expressen.se/video-player/embed/tv/nyheter/ekero-fodda-olof-gustafsson-forvaltar-knarkbaronen-pablo-escobars-namn',
+        'only_matching': True,
+    }, {
+        'url': 'https://www.expressen.se/nyheter/efter-egna-telefonbluffen-escobar-stammer-klarna/',
+        'only_matching': True,
     }]
 
     def _real_extract(self, url):

yt_dlp/extractor/mediastream.py

@@ -106,7 +106,11 @@ class MediaStreamIE(MediaStreamBaseIE):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        if 'Debido a tu ubicación no puedes ver el contenido' in webpage:
+        for message in [
+            'Debido a tu ubicación no puedes ver el contenido',
+            'You are not allowed to watch this video: Geo Fencing Restriction'
+        ]:
+            if message in webpage:
                 self.raise_geo_restricted()
 
         player_config = self._search_json(r'window\.MDSTRM\.OPTIONS\s*=', webpage, 'metadata', video_id)

yt_dlp/extractor/nova.py

@@ -6,7 +6,6 @@ from ..utils import (
     determine_ext,
     int_or_none,
     js_to_json,
-    qualities,
     traverse_obj,
     unified_strdate,
     url_or_none,
@@ -49,13 +48,8 @@ class NovaEmbedIE(InfoExtractor):
         duration = None
         formats = []
 
-        player = self._parse_json(
+        def process_format_list(format_list, format_id=""):
-            self._search_regex(
+            nonlocal formats, has_drm
-                (r'(?:(?:replacePlaceholders|processAdTagModifier).*?:\s*)?(?:replacePlaceholders|processAdTagModifier)\s*\(\s*(?P<json>{.*?})\s*\)(?:\s*\))?\s*,',
-                    r'Player\.init\s*\([^,]+,(?P<cndn>\s*\w+\s*\?)?\s*(?P<json>{(?(cndn).+?|.+)})\s*(?(cndn):|,\s*{.+?}\s*\)\s*;)'),
-                webpage, 'player', default='{}', group='json'), video_id, fatal=False)
-        if player:
-            for format_id, format_list in player['tracks'].items():
             if not isinstance(format_list, list):
                 format_list = [format_list]
             for format_dict in format_list:
@@ -82,44 +76,24 @@ class NovaEmbedIE(InfoExtractor):
                     formats.append({
                         'url': format_url,
                     })
-            duration = int_or_none(player.get('duration'))
+
-        else:
+        player = self._search_json(
-            # Old path, not actual as of 08.04.2020
+            r'player:', webpage, 'player', video_id, fatal=False, end_pattern=r';\s*</script>')
-            bitrates = self._parse_json(
+        if player:
+            for src in traverse_obj(player, ('lib', 'source', 'sources', ...)):
+                process_format_list(src)
+            duration = traverse_obj(player, ('sourceInfo', 'duration', {int_or_none}))
+        if not formats and not has_drm:
+            # older code path, in use before August 2023
+            player = self._parse_json(
                 self._search_regex(
-                    r'(?s)(?:src|bitrates)\s*=\s*({.+?})\s*;', webpage, 'formats'),
+                    (r'(?:(?:replacePlaceholders|processAdTagModifier).*?:\s*)?(?:replacePlaceholders|processAdTagModifier)\s*\(\s*(?P<json>{.*?})\s*\)(?:\s*\))?\s*,',
-                video_id, transform_source=js_to_json)
+                     r'Player\.init\s*\([^,]+,(?P<cndn>\s*\w+\s*\?)?\s*(?P<json>{(?(cndn).+?|.+)})\s*(?(cndn):|,\s*{.+?}\s*\)\s*;)'),
-
+                    webpage, 'player', group='json'), video_id)
-            QUALITIES = ('lq', 'mq', 'hq', 'hd')
+            if player:
-            quality_key = qualities(QUALITIES)
+                for format_id, format_list in player['tracks'].items():
-
+                    process_format_list(format_list, format_id)
-            for format_id, format_list in bitrates.items():
+                duration = int_or_none(player.get('duration'))
-                if not isinstance(format_list, list):
-                    format_list = [format_list]
-                for format_url in format_list:
-                    format_url = url_or_none(format_url)
-                    if not format_url:
-                        continue
-                    if format_id == 'hls':
-                        formats.extend(self._extract_m3u8_formats(
-                            format_url, video_id, ext='mp4',
-                            entry_protocol='m3u8_native', m3u8_id='hls',
-                            fatal=False))
-                        continue
-                    f = {
-                        'url': format_url,
-                    }
-                    f_id = format_id
-                    for quality in QUALITIES:
-                        if '%s.mp4' % quality in format_url:
-                            f_id += '-%s' % quality
-                            f.update({
-                                'quality': quality_key(quality),
-                                'format_note': quality.upper(),
-                            })
-                            break
-                    f['format_id'] = f_id
-                    formats.append(f)
 
         if not formats and has_drm:
             self.report_drm(video_id)

yt_dlp/extractor/rbgtum.py

@@ -1,10 +1,11 @@
 import re
 
 from .common import InfoExtractor
+from ..utils import parse_qs, remove_start, traverse_obj, ExtractorError
 
 
 class RbgTumIE(InfoExtractor):
-    _VALID_URL = r'https://live\.rbg\.tum\.de/w/(?P<id>.+)'
+    _VALID_URL = r'https://(?:live\.rbg\.tum\.de|tum\.live)/w/(?P<id>[^?#]+)'
     _TESTS = [{
         # Combined view
         'url': 'https://live.rbg.tum.de/w/cpp/22128',
@@ -35,16 +36,18 @@ class RbgTumIE(InfoExtractor):
             'title': 'Fachschaftsvollversammlung',
             'series': 'Fachschaftsvollversammlung Informatik',
         }
+    }, {
+        'url': 'https://tum.live/w/linalginfo/27102',
+        'only_matching': True,
     }, ]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
 
-        m3u8 = self._html_search_regex(r'(https://.+?\.m3u8)', webpage, 'm3u8')
+        m3u8 = self._html_search_regex(r'"(https://[^"]+\.m3u8[^"]*)', webpage, 'm3u8')
-        lecture_title = self._html_search_regex(r'(?si)<h1.*?>(.*)</h1>', webpage, 'title')
+        lecture_title = self._html_search_regex(r'<h1[^>]*>([^<]+)</h1>', webpage, 'title', fatal=False)
-        lecture_series_title = self._html_search_regex(
+        lecture_series_title = remove_start(self._html_extract_title(webpage), 'TUM-Live | ')
-            r'(?s)<title\b[^>]*>\s*(?:TUM-Live\s\|\s?)?([^:]+):?.*?</title>', webpage, 'series')
 
         formats = self._extract_m3u8_formats(m3u8, video_id, 'mp4', entry_protocol='m3u8_native', m3u8_id='hls')
 
@@ -57,9 +60,9 @@ class RbgTumIE(InfoExtractor):
 
 
 class RbgTumCourseIE(InfoExtractor):
-    _VALID_URL = r'https://live\.rbg\.tum\.de/course/(?P<id>.+)'
+    _VALID_URL = r'https://(?P<hostname>(?:live\.rbg\.tum\.de|tum\.live))/old/course/(?P<id>(?P<year>\d+)/(?P<term>\w+)/(?P<slug>[^/?#]+))'
     _TESTS = [{
-        'url': 'https://live.rbg.tum.de/course/2022/S/fpv',
+        'url': 'https://live.rbg.tum.de/old/course/2022/S/fpv',
         'info_dict': {
             'title': 'Funktionale Programmierung und Verifikation (IN0003)',
             'id': '2022/S/fpv',
@@ -69,7 +72,7 @@ class RbgTumCourseIE(InfoExtractor):
         },
         'playlist_count': 13,
     }, {
-        'url': 'https://live.rbg.tum.de/course/2022/W/set',
+        'url': 'https://live.rbg.tum.de/old/course/2022/W/set',
         'info_dict': {
             'title': 'SET FSMPIC',
             'id': '2022/W/set',
@@ -78,16 +81,62 @@ class RbgTumCourseIE(InfoExtractor):
             'noplaylist': False,
         },
         'playlist_count': 6,
+    }, {
+        'url': 'https://tum.live/old/course/2023/S/linalginfo',
+        'only_matching': True,
     }, ]
 
     def _real_extract(self, url):
-        course_id = self._match_id(url)
+        course_id, hostname, year, term, slug = self._match_valid_url(url).group('id', 'hostname', 'year', 'term', 'slug')
+        meta = self._download_json(
+            f'https://{hostname}/api/courses/{slug}/', course_id, fatal=False,
+            query={'year': year, 'term': term}) or {}
+        lecture_series_title = meta.get('Name')
+        lectures = [self.url_result(f'https://{hostname}/w/{slug}/{stream_id}', RbgTumIE)
+                    for stream_id in traverse_obj(meta, ('Streams', ..., 'ID'))]
+
+        if not lectures:
             webpage = self._download_webpage(url, course_id)
+            lecture_series_title = remove_start(self._html_extract_title(webpage), 'TUM-Live | ')
+            lectures = [self.url_result(f'https://{hostname}{lecture_path}', RbgTumIE)
+                        for lecture_path in re.findall(r'href="(/w/[^/"]+/[^/"]+)"', webpage)]
 
-        lecture_series_title = self._html_search_regex(r'(?si)<h1.*?>(.*)</h1>', webpage, 'title')
+        return self.playlist_result(lectures, course_id, lecture_series_title)
 
-        lecture_urls = []
-        for lecture_url in re.findall(r'(?i)href="/w/(.+)(?<!/cam)(?<!/pres)(?<!/chat)"', webpage):
-            lecture_urls.append(self.url_result('https://live.rbg.tum.de/w/' + lecture_url, ie=RbgTumIE.ie_key()))
 
-        return self.playlist_result(lecture_urls, course_id, lecture_series_title)
+class RbgTumNewCourseIE(InfoExtractor):
+    _VALID_URL = r'https://(?P<hostname>(?:live\.rbg\.tum\.de|tum\.live))/\?'
+    _TESTS = [{
+        'url': 'https://live.rbg.tum.de/?year=2022&term=S&slug=fpv&view=3',
+        'info_dict': {
+            'title': 'Funktionale Programmierung und Verifikation (IN0003)',
+            'id': '2022/S/fpv',
+        },
+        'params': {
+            'noplaylist': False,
+        },
+        'playlist_count': 13,
+    }, {
+        'url': 'https://live.rbg.tum.de/?year=2022&term=W&slug=set&view=3',
+        'info_dict': {
+            'title': 'SET FSMPIC',
+            'id': '2022/W/set',
+        },
+        'params': {
+            'noplaylist': False,
+        },
+        'playlist_count': 6,
+    }, {
+        'url': 'https://tum.live/?year=2023&term=S&slug=linalginfo&view=3',
+        'only_matching': True,
+    }]
+
+    def _real_extract(self, url):
+        query = parse_qs(url)
+        errors = [key for key in ('year', 'term', 'slug') if not query.get(key)]
+        if errors:
+            raise ExtractorError(f'Input URL is missing query parameters: {", ".join(errors)}')
+        year, term, slug = query['year'][0], query['term'][0], query['slug'][0]
+        hostname = self._match_valid_url(url).group('hostname')
+
+        return self.url_result(f'https://{hostname}/old/course/{year}/{term}/{slug}', RbgTumCourseIE)