[ie/nebula] Overhaul extractors (#8566)

Closes #4300, Closes #5814, Closes #7588, Closes #6334, Closes #6538
Authored by: elyse0, pukkandan, seproDev

Co-authored-by: Elyse <26639800+elyse0@users.noreply.github.com>
Co-authored-by: pukkandan <pukkandan.ytdlp@gmail.com>

test/helper.py

@@ -214,8 +214,9 @@ def sanitize_got_info_dict(got_dict):
 
     test_info_dict = {
         key: sanitize(key, value) for key, value in got_dict.items()
-        if value is not None and key not in IGNORED_FIELDS and not any(
+        if value is not None and key not in IGNORED_FIELDS and (
-            key.startswith(f'{prefix}_') for prefix in IGNORED_PREFIXES)
+            not any(key.startswith(f'{prefix}_') for prefix in IGNORED_PREFIXES)
+            or key == '_old_archive_ids')
     }
 
     # display_id may be generated from id

yt_dlp/extractor/_extractors.py

@@ -81,16 +81,20 @@ from .airmozilla import AirMozillaIE
 from .airtv import AirTVIE
 from .aitube import AitubeKZVideoIE
 from .aljazeera import AlJazeeraIE
+from .allstar import (
+    AllstarIE,
+    AllstarProfileIE,
+)
 from .alphaporno import AlphaPornoIE
 from .altcensored import (
     AltCensoredIE,
     AltCensoredChannelIE,
 )
-from .amara import AmaraIE
 from .alura import (
     AluraIE,
     AluraCourseIE
 )
+from .amara import AmaraIE
 from .amcnetworks import AMCNetworksIE
 from .amazon import (
     AmazonStoreIE,
@@ -216,6 +220,8 @@ from .bilibili import (
     BiliBiliBangumiIE,
     BiliBiliBangumiSeasonIE,
     BiliBiliBangumiMediaIE,
+    BilibiliCheeseIE,
+    BilibiliCheeseSeasonIE,
     BiliBiliSearchIE,
     BilibiliCategoryIE,
     BilibiliAudioIE,
@@ -1241,6 +1247,7 @@ from .ndr import (
 from .ndtv import NDTVIE
 from .nebula import (
     NebulaIE,
+    NebulaClassIE,
     NebulaSubscriptionsIE,
     NebulaChannelIE,
 )

yt_dlp/extractor/allstar.py

@@ -0,0 +1,253 @@
+import functools
+import json
+
+from .common import InfoExtractor
+from ..utils import (
+    ExtractorError,
+    OnDemandPagedList,
+    int_or_none,
+    join_nonempty,
+    parse_qs,
+    urljoin,
+)
+from ..utils.traversal import traverse_obj
+
+
+_FIELDS = '''
+    _id
+    clipImageSource
+    clipImageThumb
+    clipLink
+    clipTitle
+    createdDate
+    shareId
+    user { _id }
+    username
+    views'''
+
+_EXTRA_FIELDS = '''
+    clipLength
+    clipSizeBytes'''
+
+_QUERIES = {
+    'clip': '''query ($id: String!) {
+        video: getClip(clipIdentifier: $id) {
+            %s %s
+        }
+    }''' % (_FIELDS, _EXTRA_FIELDS),
+    'montage': '''query ($id: String!) {
+        video: getMontage(clipIdentifier: $id) {
+            %s
+        }
+    }''' % _FIELDS,
+    'Clips': '''query ($page: Int!, $user: String!, $game: Int) {
+        videos: clips(search: createdDate, page: $page, user: $user, mobile: false, game: $game) {
+            data { %s %s }
+        }
+    }''' % (_FIELDS, _EXTRA_FIELDS),
+    'Montages': '''query ($page: Int!, $user: String!) {
+        videos: montages(search: createdDate, page: $page, user: $user) {
+            data { %s }
+        }
+    }''' % _FIELDS,
+    'Mobile Clips': '''query ($page: Int!, $user: String!) {
+        videos: clips(search: createdDate, page: $page, user: $user, mobile: true) {
+            data { %s %s }
+        }
+    }''' % (_FIELDS, _EXTRA_FIELDS),
+}
+
+
+class AllstarBaseIE(InfoExtractor):
+    @staticmethod
+    def _parse_video_data(video_data):
+        def media_url_or_none(path):
+            return urljoin('https://media.allstar.gg/', path)
+
+        info = traverse_obj(video_data, {
+            'id': ('_id', {str}),
+            'display_id': ('shareId', {str}),
+            'title': ('clipTitle', {str}),
+            'url': ('clipLink', {media_url_or_none}),
+            'thumbnails': (('clipImageThumb', 'clipImageSource'), {'url': {media_url_or_none}}),
+            'duration': ('clipLength', {int_or_none}),
+            'filesize': ('clipSizeBytes', {int_or_none}),
+            'timestamp': ('createdDate', {functools.partial(int_or_none, scale=1000)}),
+            'uploader': ('username', {str}),
+            'uploader_id': ('user', '_id', {str}),
+            'view_count': ('views', {int_or_none}),
+        })
+
+        if info.get('id') and info.get('url'):
+            basename = 'clip' if '/clips/' in info['url'] else 'montage'
+            info['webpage_url'] = f'https://allstar.gg/{basename}?{basename}={info["id"]}'
+
+        info.update({
+            'extractor_key': AllstarIE.ie_key(),
+            'extractor': AllstarIE.IE_NAME,
+            'uploader_url': urljoin('https://allstar.gg/u/', info.get('uploader_id')),
+        })
+
+        return info
+
+    def _call_api(self, query, variables, path, video_id=None, note=None):
+        response = self._download_json(
+            'https://a1.allstar.gg/graphql', video_id, note=note,
+            headers={'content-type': 'application/json'},
+            data=json.dumps({'variables': variables, 'query': query}).encode())
+
+        errors = traverse_obj(response, ('errors', ..., 'message', {str}))
+        if errors:
+            raise ExtractorError('; '.join(errors))
+
+        return traverse_obj(response, path)
+
+
+class AllstarIE(AllstarBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?allstar\.gg/(?P<type>(?:clip|montage))\?(?P=type)=(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'https://allstar.gg/clip?clip=64482c2da9eec30008a67d1b',
+        'info_dict': {
+            'id': '64482c2da9eec30008a67d1b',
+            'title': '4K on Inferno',
+            'url': 'md5:66befb5381eef0c9456026386c25fa55',
+            'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
+            'uploader': 'chrk.',
+            'ext': 'mp4',
+            'duration': 20,
+            'filesize': 21199257,
+            'timestamp': 1682451501,
+            'uploader_id': '62b8bdfc9021052f7905882d',
+            'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
+            'upload_date': '20230425',
+            'view_count': int,
+        }
+    }, {
+        'url': 'https://allstar.gg/clip?clip=8LJLY4JKB',
+        'info_dict': {
+            'id': '64a1ec6b887f4c0008dc50b8',
+            'display_id': '8LJLY4JKB',
+            'title': 'AK-47 3K on Mirage',
+            'url': 'md5:dde224fd12f035c0e2529a4ae34c4283',
+            'ext': 'mp4',
+            'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
+            'duration': 16,
+            'filesize': 30175859,
+            'timestamp': 1688333419,
+            'uploader': 'cherokee',
+            'uploader_id': '62b8bdfc9021052f7905882d',
+            'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
+            'upload_date': '20230702',
+            'view_count': int,
+        }
+    }, {
+        'url': 'https://allstar.gg/montage?montage=643e64089da7e9363e1fa66c',
+        'info_dict': {
+            'id': '643e64089da7e9363e1fa66c',
+            'display_id': 'APQLGM2IMXW',
+            'title': 'cherokee Rapid Fire Snipers Montage',
+            'url': 'md5:a3ee356022115db2b27c81321d195945',
+            'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
+            'ext': 'mp4',
+            'timestamp': 1681810448,
+            'uploader': 'cherokee',
+            'uploader_id': '62b8bdfc9021052f7905882d',
+            'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
+            'upload_date': '20230418',
+            'view_count': int,
+        }
+    }, {
+        'url': 'https://allstar.gg/montage?montage=RILJMH6QOS',
+        'info_dict': {
+            'id': '64a2697372ce3703de29e868',
+            'display_id': 'RILJMH6QOS',
+            'title': 'cherokee Rapid Fire Snipers Montage',
+            'url': 'md5:d5672e6f88579730c2310a80fdbc4030',
+            'thumbnail': r're:https://media\.allstar\.gg/.+\.(?:png|jpg)$',
+            'ext': 'mp4',
+            'timestamp': 1688365434,
+            'uploader': 'cherokee',
+            'uploader_id': '62b8bdfc9021052f7905882d',
+            'uploader_url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d',
+            'upload_date': '20230703',
+            'view_count': int,
+        }
+    }]
+
+    def _real_extract(self, url):
+        query_id, video_id = self._match_valid_url(url).group('type', 'id')
+
+        return self._parse_video_data(
+            self._call_api(
+                _QUERIES.get(query_id), {'id': video_id}, ('data', 'video'), video_id))
+
+
+class AllstarProfileIE(AllstarBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?allstar\.gg/(?:profile\?user=|u/)(?P<id>[^/?#&]+)'
+
+    _TESTS = [{
+        'url': 'https://allstar.gg/profile?user=62b8bdfc9021052f7905882d',
+        'info_dict': {
+            'id': '62b8bdfc9021052f7905882d-clips',
+            'title': 'cherokee - Clips',
+        },
+        'playlist_mincount': 15
+    }, {
+        'url': 'https://allstar.gg/u/cherokee?game=730&view=Clips',
+        'info_dict': {
+            'id': '62b8bdfc9021052f7905882d-clips-730',
+            'title': 'cherokee - Clips - 730',
+        },
+        'playlist_mincount': 15
+    }, {
+        'url': 'https://allstar.gg/u/62b8bdfc9021052f7905882d?view=Montages',
+        'info_dict': {
+            'id': '62b8bdfc9021052f7905882d-montages',
+            'title': 'cherokee - Montages',
+        },
+        'playlist_mincount': 4
+    }, {
+        'url': 'https://allstar.gg/profile?user=cherokee&view=Mobile Clips',
+        'info_dict': {
+            'id': '62b8bdfc9021052f7905882d-mobile',
+            'title': 'cherokee - Mobile Clips',
+        },
+        'playlist_mincount': 1
+    }]
+
+    _PAGE_SIZE = 10
+
+    def _get_page(self, user_id, display_id, game, query, page_num):
+        page_num += 1
+
+        for video_data in self._call_api(
+                query, {
+                    'user': user_id,
+                    'page': page_num,
+                    'game': game,
+                }, ('data', 'videos', 'data'), display_id, f'Downloading page {page_num}'):
+            yield self._parse_video_data(video_data)
+
+    def _real_extract(self, url):
+        display_id = self._match_id(url)
+        profile_data = self._download_json(
+            urljoin('https://api.allstar.gg/v1/users/profile/', display_id), display_id)
+        user_id = traverse_obj(profile_data, ('data', ('_id'), {str}))
+        if not user_id:
+            raise ExtractorError('Unable to extract the user id')
+
+        username = traverse_obj(profile_data, ('data', 'profile', ('username'), {str}))
+        url_query = parse_qs(url)
+        game = traverse_obj(url_query, ('game', 0, {int_or_none}))
+        query_id = traverse_obj(url_query, ('view', 0), default='Clips')
+
+        if query_id not in ('Clips', 'Montages', 'Mobile Clips'):
+            raise ExtractorError(f'Unsupported playlist URL type {query_id!r}')
+
+        return self.playlist_result(
+            OnDemandPagedList(
+                functools.partial(
+                    self._get_page, user_id, display_id, game, _QUERIES.get(query_id)), self._PAGE_SIZE),
+            playlist_id=join_nonempty(user_id, query_id.lower().split()[0], game),
+            playlist_title=join_nonempty((username or display_id), query_id, game, delim=' - '))

yt_dlp/extractor/bilibili.py

@@ -2,6 +2,7 @@ import base64
 import functools
 import hashlib
 import itertools
+import json
 import math
 import re
 import time
@@ -16,9 +17,11 @@ from ..utils import (
     InAdvancePagedList,
     OnDemandPagedList,
     bool_or_none,
+    clean_html,
     filter_dict,
     float_or_none,
     format_field,
+    get_element_by_class,
     int_or_none,
     join_nonempty,
     make_archive_id,
@@ -88,6 +91,12 @@ class BilibiliBaseIE(InfoExtractor):
 
         return formats
 
+    def _download_playinfo(self, video_id, cid):
+        return self._download_json(
+            'https://api.bilibili.com/x/player/playurl', video_id,
+            query={'bvid': video_id, 'cid': cid, 'fnval': 4048},
+            note=f'Downloading video formats for cid {cid}')['data']
+
     def json2srt(self, json_data):
         srt_data = ''
         for idx, line in enumerate(json_data.get('body') or []):
@@ -96,7 +105,7 @@ class BilibiliBaseIE(InfoExtractor):
                          f'{line["content"]}\n\n')
         return srt_data
 
-    def _get_subtitles(self, video_id, aid, cid):
+    def _get_subtitles(self, video_id, cid, aid=None):
         subtitles = {
             'danmaku': [{
                 'ext': 'xml',
@@ -104,8 +113,15 @@ class BilibiliBaseIE(InfoExtractor):
             }]
         }
 
-        video_info_json = self._download_json(f'https://api.bilibili.com/x/player/v2?aid={aid}&cid={cid}', video_id)
+        subtitle_info = traverse_obj(self._download_json(
-        for s in traverse_obj(video_info_json, ('data', 'subtitle', 'subtitles', ...)):
+            'https://api.bilibili.com/x/player/v2', video_id,
+            query={'aid': aid, 'cid': cid} if aid else {'bvid': video_id, 'cid': cid},
+            note=f'Extracting subtitle info {cid}'), ('data', 'subtitle'))
+        subs_list = traverse_obj(subtitle_info, ('subtitles', lambda _, v: v['subtitle_url'] and v['lan']))
+        if not subs_list and traverse_obj(subtitle_info, 'allow_submit'):
+            if not self._get_cookies('https://api.bilibili.com').get('SESSDATA'):  # no login session cookie
+                self.report_warning(f'CC subtitles (if any) are only visible when logged in. {self._login_hint()}', only_once=True)
+        for s in subs_list:
             subtitles.setdefault(s['lan'], []).append({
                 'ext': 'srt',
                 'data': self.json2srt(self._download_json(s['subtitle_url'], video_id))
@@ -155,7 +171,54 @@ class BilibiliBaseIE(InfoExtractor):
         for entry in traverse_obj(season_info, (
                 'result', 'main_section', 'episodes',
                 lambda _, v: url_or_none(v['share_url']) and v['id'])):
-            yield self.url_result(entry['share_url'], BiliBiliBangumiIE, f'ep{entry["id"]}')
+            yield self.url_result(entry['share_url'], BiliBiliBangumiIE, str_or_none(entry.get('id')))
+
+    def _get_divisions(self, video_id, graph_version, edges, edge_id, cid_edges=None):
+        cid_edges = cid_edges or {}
+        division_data = self._download_json(
+            'https://api.bilibili.com/x/stein/edgeinfo_v2', video_id,
+            query={'graph_version': graph_version, 'edge_id': edge_id, 'bvid': video_id},
+            note=f'Extracting divisions from edge {edge_id}')
+        edges.setdefault(edge_id, {}).update(
+            traverse_obj(division_data, ('data', 'story_list', lambda _, v: v['edge_id'] == edge_id, {
+                'title': ('title', {str}),
+                'cid': ('cid', {int_or_none}),
+            }), get_all=False))
+
+        edges[edge_id].update(traverse_obj(division_data, ('data', {
+            'title': ('title', {str}),
+            'choices': ('edges', 'questions', ..., 'choices', ..., {
+                'edge_id': ('id', {int_or_none}),
+                'cid': ('cid', {int_or_none}),
+                'text': ('option', {str}),
+            }),
+        })))
+        # use dict to combine edges that use the same video section (same cid)
+        cid_edges.setdefault(edges[edge_id]['cid'], {})[edge_id] = edges[edge_id]
+        for choice in traverse_obj(edges, (edge_id, 'choices', ...)):
+            if choice['edge_id'] not in edges:
+                edges[choice['edge_id']] = {'cid': choice['cid']}
+                self._get_divisions(video_id, graph_version, edges, choice['edge_id'], cid_edges=cid_edges)
+        return cid_edges
+
+    def _get_interactive_entries(self, video_id, cid, metainfo):
+        graph_version = traverse_obj(
+            self._download_json(
+                'https://api.bilibili.com/x/player/wbi/v2', video_id,
+                'Extracting graph version', query={'bvid': video_id, 'cid': cid}),
+            ('data', 'interaction', 'graph_version', {int_or_none}))
+        cid_edges = self._get_divisions(video_id, graph_version, {1: {'cid': cid}}, 1)
+        for cid, edges in cid_edges.items():
+            play_info = self._download_playinfo(video_id, cid)
+            yield {
+                **metainfo,
+                'id': f'{video_id}_{cid}',
+                'title': f'{metainfo.get("title")} - {list(edges.values())[0].get("title")}',
+                'formats': self.extract_formats(play_info),
+                'description': f'{json.dumps(edges, ensure_ascii=False)}\n{metainfo.get("description", "")}',
+                'duration': float_or_none(play_info.get('timelength'), scale=1000),
+                'subtitles': self.extract_subtitles(video_id, cid),
+            }
 
 
 class BiliBiliIE(BilibiliBaseIE):
@@ -180,7 +243,7 @@ class BiliBiliIE(BilibiliBaseIE):
             'view_count': int,
         },
     }, {
-        # old av URL version
+        'note': 'old av URL version',
         'url': 'http://www.bilibili.com/video/av1074402/',
         'info_dict': {
             'thumbnail': r're:^https?://.*\.(jpg|jpeg)$',
@@ -212,7 +275,7 @@ class BiliBiliIE(BilibiliBaseIE):
                 'id': 'BV1bK411W797_p1',
                 'ext': 'mp4',
                 'title': '物语中的人物是如何吐槽自己的OP的 p01 Staple Stable/战场原+羽川',
-                'tags': 'count:11',
+                'tags': 'count:10',
                 'timestamp': 1589601697,
                 'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
                 'uploader': '打牌还是打桩',
@@ -232,7 +295,7 @@ class BiliBiliIE(BilibiliBaseIE):
             'id': 'BV1bK411W797_p1',
             'ext': 'mp4',
             'title': '物语中的人物是如何吐槽自己的OP的 p01 Staple Stable/战场原+羽川',
-            'tags': 'count:11',
+            'tags': 'count:10',
             'timestamp': 1589601697,
             'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
             'uploader': '打牌还是打桩',
@@ -343,18 +406,120 @@ class BiliBiliIE(BilibiliBaseIE):
             'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
         },
         'params': {'skip_download': True},
+    }, {
+        'note': 'interactive/split-path video',
+        'url': 'https://www.bilibili.com/video/BV1af4y1H7ga/',
+        'info_dict': {
+            'id': 'BV1af4y1H7ga',
+            'title': '【互动游戏】花了大半年时间做的自我介绍~请查收！！',
+            'timestamp': 1630500414,
+            'upload_date': '20210901',
+            'description': 'md5:01113e39ab06e28042d74ac356a08786',
+            'tags': list,
+            'uploader': '钉宫妮妮Ninico',
+            'duration': 1503,
+            'uploader_id': '8881297',
+            'comment_count': int,
+            'view_count': int,
+            'like_count': int,
+            'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+        },
+        'playlist_count': 33,
+        'playlist': [{
+            'info_dict': {
+                'id': 'BV1af4y1H7ga_400950101',
+                'ext': 'mp4',
+                'title': '【互动游戏】花了大半年时间做的自我介绍~请查收！！ - 听见猫猫叫~',
+                'timestamp': 1630500414,
+                'upload_date': '20210901',
+                'description': 'md5:db66ac7a2813a94b8291dbce990cc5b2',
+                'tags': list,
+                'uploader': '钉宫妮妮Ninico',
+                'duration': 11.605,
+                'uploader_id': '8881297',
+                'comment_count': int,
+                'view_count': int,
+                'like_count': int,
+                'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+            },
+        }],
+    }, {
+        'note': '301 redirect to bangumi link',
+        'url': 'https://www.bilibili.com/video/BV1TE411f7f1',
+        'info_dict': {
+            'id': '288525',
+            'title': '李永乐老师 钱学森弹道和乘波体飞行器是什么？',
+            'ext': 'mp4',
+            'series': '我和我的祖国',
+            'series_id': '4780',
+            'season': '幕后纪实',
+            'season_id': '28609',
+            'season_number': 1,
+            'episode': '钱学森弹道和乘波体飞行器是什么？',
+            'episode_id': '288525',
+            'episode_number': 105,
+            'duration': 1183.957,
+            'timestamp': 1571648124,
+            'upload_date': '20191021',
+            'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+        },
+    }, {
+        'url': 'https://www.bilibili.com/video/BV1jL41167ZG/',
+        'info_dict': {
+            'id': 'BV1jL41167ZG',
+            'title': '一场大火引发的离奇死亡！古典推理经典短篇集《不可能犯罪诊断书》！',
+            'ext': 'mp4',
+        },
+        'skip': 'supporter-only video',
+    }, {
+        'url': 'https://www.bilibili.com/video/BV1Ks411f7aQ/',
+        'info_dict': {
+            'id': 'BV1Ks411f7aQ',
+            'title': '【BD1080P】狼与香辛料I【华盟】',
+            'ext': 'mp4',
+        },
+        'skip': 'login required',
+    }, {
+        'url': 'https://www.bilibili.com/video/BV1GJ411x7h7/',
+        'info_dict': {
+            'id': 'BV1GJ411x7h7',
+            'title': '【官方 MV】Never Gonna Give You Up - Rick Astley',
+            'ext': 'mp4',
+        },
+        'skip': 'geo-restricted',
     }]
 
     def _real_extract(self, url):
         video_id = self._match_id(url)
-        webpage = self._download_webpage(url, video_id)
+        webpage, urlh = self._download_webpage_handle(url, video_id)
+        if not self._match_valid_url(urlh.url):
+            return self.url_result(urlh.url)
+
         initial_state = self._search_json(r'window\.__INITIAL_STATE__\s*=', webpage, 'initial state', video_id)
 
         is_festival = 'videoData' not in initial_state
         if is_festival:
             video_data = initial_state['videoInfo']
         else:
-            play_info = self._search_json(r'window\.__playinfo__\s*=', webpage, 'play info', video_id)['data']
+            play_info_obj = self._search_json(
+                r'window\.__playinfo__\s*=', webpage, 'play info', video_id, fatal=False)
+            if not play_info_obj:
+                if traverse_obj(initial_state, ('error', 'trueCode')) == -403:
+                    self.raise_login_required()
+                if traverse_obj(initial_state, ('error', 'trueCode')) == -404:
+                    raise ExtractorError(
+                        'This video may be deleted or geo-restricted. '
+                        'You might want to try a VPN or a proxy server (with --proxy)', expected=True)
+            play_info = traverse_obj(play_info_obj, ('data', {dict}))
+            if not play_info:
+                if traverse_obj(play_info_obj, 'code') == 87007:
+                    toast = get_element_by_class('tips-toast', webpage) or ''
+                    msg = clean_html(
+                        f'{get_element_by_class("belongs-to", toast) or ""}，'
+                        + (get_element_by_class('level', toast) or ''))
+                    raise ExtractorError(
+                        f'This is a supporter-only video: {msg}. {self._login_hint()}', expected=True)
+                raise ExtractorError('Failed to extract play info')
             video_data = initial_state['videoData']
 
         video_id, title = video_data['bvid'], video_data.get('title')
@@ -385,10 +550,7 @@ class BiliBiliIE(BilibiliBaseIE):
 
         festival_info = {}
         if is_festival:
-            play_info = self._download_json(
+            play_info = self._download_playinfo(video_id, cid)
-                'https://api.bilibili.com/x/player/playurl', video_id,
-                query={'bvid': video_id, 'cid': cid, 'fnval': 4048},
-                note='Extracting festival video formats')['data']
 
             festival_info = traverse_obj(initial_state, {
                 'uploader': ('videoInfo', 'upName'),
@@ -397,7 +559,7 @@ class BiliBiliIE(BilibiliBaseIE):
                 'thumbnail': ('sectionEpisodes', lambda _, v: v['bvid'] == video_id, 'cover'),
             }, get_all=False)
 
-        return {
+        metainfo = {
             **traverse_obj(initial_state, {
                 'uploader': ('upData', 'name'),
                 'uploader_id': ('upData', 'mid', {str_or_none}),
@@ -413,28 +575,59 @@ class BiliBiliIE(BilibiliBaseIE):
                 'comment_count': ('stat', 'reply', {int_or_none}),
             }, get_all=False),
             'id': f'{video_id}{format_field(part_id, None, "_p%d")}',
-            'formats': self.extract_formats(play_info),
             '_old_archive_ids': [make_archive_id(self, old_video_id)] if old_video_id else None,
             'title': title,
+            'http_headers': {'Referer': url},
+        }
+
+        is_interactive = traverse_obj(video_data, ('rights', 'is_stein_gate'))
+        if is_interactive:
+            return self.playlist_result(
+                self._get_interactive_entries(video_id, cid, metainfo), **metainfo, **{
+                    'duration': traverse_obj(initial_state, ('videoData', 'duration', {int_or_none})),
+                    '__post_extractor': self.extract_comments(aid),
+                })
+        else:
+            return {
+                **metainfo,
                 'duration': float_or_none(play_info.get('timelength'), scale=1000),
                 'chapters': self._get_chapters(aid, cid),
-            'subtitles': self.extract_subtitles(video_id, aid, cid),
+                'subtitles': self.extract_subtitles(video_id, cid),
+                'formats': self.extract_formats(play_info),
                 '__post_extractor': self.extract_comments(aid),
-            'http_headers': {'Referer': url},
             }
 
 
 class BiliBiliBangumiIE(BilibiliBaseIE):
-    _VALID_URL = r'https?://(?:www\.)?bilibili\.com/bangumi/play/(?P<id>ep\d+)'
+    _VALID_URL = r'https?://(?:www\.)?bilibili\.com/bangumi/play/ep(?P<id>\d+)'
 
     _TESTS = [{
+        'url': 'https://www.bilibili.com/bangumi/play/ep21495/',
+        'info_dict': {
+            'id': '21495',
+            'ext': 'mp4',
+            'series': '悠久之翼',
+            'series_id': '774',
+            'season': '第二季',
+            'season_id': '1182',
+            'season_number': 2,
+            'episode': 'forever／ef',
+            'episode_id': '21495',
+            'episode_number': 12,
+            'title': '12 forever／ef',
+            'duration': 1420.791,
+            'timestamp': 1320412200,
+            'upload_date': '20111104',
+            'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+        },
+    }, {
         'url': 'https://www.bilibili.com/bangumi/play/ep267851',
         'info_dict': {
             'id': '267851',
             'ext': 'mp4',
             'series': '鬼灭之刃',
             'series_id': '4358',
-            'season': '鬼灭之刃',
+            'season': '立志篇',
             'season_id': '26801',
             'season_number': 1,
             'episode': '残酷',
@@ -446,13 +639,32 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
             'upload_date': '20190406',
             'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$'
         },
-        'skip': 'According to the copyright owner\'s request, you may only watch the video after you are premium member.'
+        'skip': 'Geo-restricted',
+    }, {
+        'note': 'a making-of which falls outside main section',
+        'url': 'https://www.bilibili.com/bangumi/play/ep345120',
+        'info_dict': {
+            'id': '345120',
+            'ext': 'mp4',
+            'series': '鬼灭之刃',
+            'series_id': '4358',
+            'season': '立志篇',
+            'season_id': '26801',
+            'season_number': 1,
+            'episode': '炭治郎篇',
+            'episode_id': '345120',
+            'episode_number': 27,
+            'title': '#1 炭治郎篇',
+            'duration': 1922.129,
+            'timestamp': 1602853860,
+            'upload_date': '20201016',
+            'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$'
+        },
     }]
 
     def _real_extract(self, url):
-        video_id = self._match_id(url)
+        episode_id = self._match_id(url)
-        episode_id = video_id[2:]
+        webpage = self._download_webpage(url, episode_id)
-        webpage = self._download_webpage(url, video_id)
 
         if '您所在的地区无法观看本片' in webpage:
             raise GeoRestrictedError('This video is restricted')
@@ -461,7 +673,7 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
 
         headers = {'Referer': url, **self.geo_verification_headers()}
         play_info = self._download_json(
-            'https://api.bilibili.com/pgc/player/web/v2/playurl', video_id,
+            'https://api.bilibili.com/pgc/player/web/v2/playurl', episode_id,
             'Extracting episode', query={'fnval': '4048', 'ep_id': episode_id},
             headers=headers)
         premium_only = play_info.get('code') == -10403
@@ -472,40 +684,43 @@ class BiliBiliBangumiIE(BilibiliBaseIE):
             self.raise_login_required('This video is for premium members only')
 
         bangumi_info = self._download_json(
-            'https://api.bilibili.com/pgc/view/web/season', video_id, 'Get episode details',
+            'https://api.bilibili.com/pgc/view/web/season', episode_id, 'Get episode details',
             query={'ep_id': episode_id}, headers=headers)['result']
 
         episode_number, episode_info = next((
             (idx, ep) for idx, ep in enumerate(traverse_obj(
-                bangumi_info, ('episodes', ..., {dict})), 1)
+                bangumi_info, (('episodes', ('section', ..., 'episodes')), ..., {dict})), 1)
             if str_or_none(ep.get('id')) == episode_id), (1, {}))
 
         season_id = bangumi_info.get('season_id')
-        season_number = season_id and next((
+        season_number, season_title = season_id and next((
-            idx + 1 for idx, e in enumerate(
+            (idx + 1, e.get('season_title')) for idx, e in enumerate(
                 traverse_obj(bangumi_info, ('seasons', ...)))
             if e.get('season_id') == season_id
-        ), None)
+        ), (None, None))
 
         aid = episode_info.get('aid')
 
         return {
-            'id': video_id,
+            'id': episode_id,
             'formats': formats,
             **traverse_obj(bangumi_info, {
                 'series': ('series', 'series_title', {str}),
                 'series_id': ('series', 'series_id', {str_or_none}),
                 'thumbnail': ('square_cover', {url_or_none}),
             }),
-            'title': join_nonempty('title', 'long_title', delim=' ', from_dict=episode_info),
+            **traverse_obj(episode_info, {
-            'episode': episode_info.get('long_title'),
+                'episode': ('long_title', {str}),
+                'episode_number': ('title', {int_or_none}, {lambda x: x or episode_number}),
+                'timestamp': ('pub_time', {int_or_none}),
+                'title': {lambda v: v and join_nonempty('title', 'long_title', delim=' ', from_dict=v)},
+            }),
             'episode_id': episode_id,
-            'episode_number': int_or_none(episode_info.get('title')) or episode_number,
+            'season': str_or_none(season_title),
             'season_id': str_or_none(season_id),
             'season_number': season_number,
-            'timestamp': int_or_none(episode_info.get('pub_time')),
             'duration': float_or_none(play_info.get('timelength'), scale=1000),
-            'subtitles': self.extract_subtitles(video_id, aid, episode_info.get('cid')),
+            'subtitles': self.extract_subtitles(episode_id, episode_info.get('cid'), aid=aid),
             '__post_extractor': self.extract_comments(aid),
             'http_headers': headers,
         }
@@ -517,17 +732,53 @@ class BiliBiliBangumiMediaIE(BilibiliBaseIE):
         'url': 'https://www.bilibili.com/bangumi/media/md24097891',
         'info_dict': {
             'id': '24097891',
+            'title': 'CAROLE & TUESDAY',
+            'description': 'md5:42417ad33d1eaa1c93bfd2dd1626b829',
         },
         'playlist_mincount': 25,
+    }, {
+        'url': 'https://www.bilibili.com/bangumi/media/md1565/',
+        'info_dict': {
+            'id': '1565',
+            'title': '攻壳机动队 S.A.C. 2nd GIG',
+            'description': 'md5:46cac00bafd645b97f4d6df616fc576d',
+        },
+        'playlist_count': 26,
+        'playlist': [{
+            'info_dict': {
+                'id': '68540',
+                'ext': 'mp4',
+                'series': '攻壳机动队',
+                'series_id': '1077',
+                'season': '第二季',
+                'season_id': '1565',
+                'season_number': 2,
+                'episode': '再启动 REEMBODY',
+                'episode_id': '68540',
+                'episode_number': 1,
+                'title': '1 再启动 REEMBODY',
+                'duration': 1525.777,
+                'timestamp': 1425074413,
+                'upload_date': '20150227',
+                'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$'
+            },
+        }],
     }]
 
     def _real_extract(self, url):
         media_id = self._match_id(url)
         webpage = self._download_webpage(url, media_id)
-        ss_id = self._search_json(
-            r'window\.__INITIAL_STATE__\s*=', webpage, 'initial_state', media_id)['mediaInfo']['season_id']
 
-        return self.playlist_result(self._get_episodes_from_season(ss_id, url), media_id)
+        initial_state = self._search_json(
+            r'window\.__INITIAL_STATE__\s*=', webpage, 'initial_state', media_id)
+        ss_id = initial_state['mediaInfo']['season_id']
+
+        return self.playlist_result(
+            self._get_episodes_from_season(ss_id, url), media_id,
+            **traverse_obj(initial_state, ('mediaInfo', {
+                'title': ('title', {str}),
+                'description': ('evaluate', {str}),
+            })))
 
 
 class BiliBiliBangumiSeasonIE(BilibiliBaseIE):
@@ -535,15 +786,183 @@ class BiliBiliBangumiSeasonIE(BilibiliBaseIE):
     _TESTS = [{
         'url': 'https://www.bilibili.com/bangumi/play/ss26801',
         'info_dict': {
-            'id': '26801'
+            'id': '26801',
+            'title': '鬼灭之刃',
+            'description': 'md5:e2cc9848b6f69be6db79fc2a82d9661b',
         },
         'playlist_mincount': 26
+    }, {
+        'url': 'https://www.bilibili.com/bangumi/play/ss2251',
+        'info_dict': {
+            'id': '2251',
+            'title': '玲音',
+            'description': 'md5:1fd40e3df4c08d4d9d89a6a34844bdc4',
+        },
+        'playlist_count': 13,
+        'playlist': [{
+            'info_dict': {
+                'id': '50188',
+                'ext': 'mp4',
+                'series': '玲音',
+                'series_id': '1526',
+                'season': 'TV',
+                'season_id': '2251',
+                'season_number': 1,
+                'episode': 'WEIRD',
+                'episode_id': '50188',
+                'episode_number': 1,
+                'title': '1 WEIRD',
+                'duration': 1436.992,
+                'timestamp': 1343185080,
+                'upload_date': '20120725',
+                'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$'
+            },
+        }],
     }]
 
     def _real_extract(self, url):
         ss_id = self._match_id(url)
+        webpage = self._download_webpage(url, ss_id)
+        metainfo = traverse_obj(
+            self._search_json(r'<script[^>]+type="application/ld\+json"[^>]*>', webpage, 'info', ss_id),
+            ('itemListElement', ..., {
+                'title': ('name', {str}),
+                'description': ('description', {str}),
+            }), get_all=False)
 
-        return self.playlist_result(self._get_episodes_from_season(ss_id, url), ss_id)
+        return self.playlist_result(self._get_episodes_from_season(ss_id, url), ss_id, **metainfo)
+
+
+class BilibiliCheeseBaseIE(BilibiliBaseIE):
+    _HEADERS = {'Referer': 'https://www.bilibili.com/'}
+
+    def _extract_episode(self, season_info, ep_id):
+        episode_info = traverse_obj(season_info, (
+            'episodes', lambda _, v: v['id'] == int(ep_id)), get_all=False)
+        aid, cid = episode_info['aid'], episode_info['cid']
+
+        if traverse_obj(episode_info, 'ep_status') == -1:
+            raise ExtractorError('This course episode is not yet available.', expected=True)
+        if not traverse_obj(episode_info, 'playable'):
+            self.raise_login_required('You need to purchase the course to download this episode')
+
+        play_info = self._download_json(
+            'https://api.bilibili.com/pugv/player/web/playurl', ep_id,
+            query={'avid': aid, 'cid': cid, 'ep_id': ep_id, 'fnval': 16, 'fourk': 1},
+            headers=self._HEADERS, note='Downloading playinfo')['data']
+
+        return {
+            'id': str_or_none(ep_id),
+            'episode_id': str_or_none(ep_id),
+            'formats': self.extract_formats(play_info),
+            'extractor_key': BilibiliCheeseIE.ie_key(),
+            'extractor': BilibiliCheeseIE.IE_NAME,
+            'webpage_url': f'https://www.bilibili.com/cheese/play/ep{ep_id}',
+            **traverse_obj(episode_info, {
+                'episode': ('title', {str}),
+                'title': {lambda v: v and join_nonempty('index', 'title', delim=' - ', from_dict=v)},
+                'alt_title': ('subtitle', {str}),
+                'duration': ('duration', {int_or_none}),
+                'episode_number': ('index', {int_or_none}),
+                'thumbnail': ('cover', {url_or_none}),
+                'timestamp': ('release_date', {int_or_none}),
+                'view_count': ('play', {int_or_none}),
+            }),
+            **traverse_obj(season_info, {
+                'uploader': ('up_info', 'uname', {str}),
+                'uploader_id': ('up_info', 'mid', {str_or_none}),
+            }),
+            'subtitles': self.extract_subtitles(ep_id, cid, aid=aid),
+            '__post_extractor': self.extract_comments(aid),
+            'http_headers': self._HEADERS,
+        }
+
+    def _download_season_info(self, query_key, video_id):
+        return self._download_json(
+            f'https://api.bilibili.com/pugv/view/web/season?{query_key}={video_id}', video_id,
+            headers=self._HEADERS, note='Downloading season info')['data']
+
+
+class BilibiliCheeseIE(BilibiliCheeseBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?bilibili\.com/cheese/play/ep(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'https://www.bilibili.com/cheese/play/ep229832',
+        'info_dict': {
+            'id': '229832',
+            'ext': 'mp4',
+            'title': '1 - 课程先导片',
+            'alt_title': '视频课 · 3分41秒',
+            'uploader': '马督工',
+            'uploader_id': '316568752',
+            'episode': '课程先导片',
+            'episode_id': '229832',
+            'episode_number': 1,
+            'duration': 221,
+            'timestamp': 1695549606,
+            'upload_date': '20230924',
+            'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+            'view_count': int,
+        }
+    }]
+
+    def _real_extract(self, url):
+        ep_id = self._match_id(url)
+        return self._extract_episode(self._download_season_info('ep_id', ep_id), ep_id)
+
+
+class BilibiliCheeseSeasonIE(BilibiliCheeseBaseIE):
+    _VALID_URL = r'https?://(?:www\.)?bilibili\.com/cheese/play/ss(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'https://www.bilibili.com/cheese/play/ss5918',
+        'info_dict': {
+            'id': '5918',
+            'title': '【限时五折】新闻系学不到：马督工教你做自媒体',
+            'description': '帮普通人建立世界模型，降低人与人的沟通门槛',
+        },
+        'playlist': [{
+            'info_dict': {
+                'id': '229832',
+                'ext': 'mp4',
+                'title': '1 - 课程先导片',
+                'alt_title': '视频课 · 3分41秒',
+                'uploader': '马督工',
+                'uploader_id': '316568752',
+                'episode': '课程先导片',
+                'episode_id': '229832',
+                'episode_number': 1,
+                'duration': 221,
+                'timestamp': 1695549606,
+                'upload_date': '20230924',
+                'thumbnail': r're:^https?://.*\.(jpg|jpeg|png)$',
+                'view_count': int,
+            }
+        }],
+        'params': {'playlist_items': '1'},
+    }, {
+        'url': 'https://www.bilibili.com/cheese/play/ss5918',
+        'info_dict': {
+            'id': '5918',
+            'title': '【限时五折】新闻系学不到：马督工教你做自媒体',
+            'description': '帮普通人建立世界模型，降低人与人的沟通门槛',
+        },
+        'playlist_mincount': 5,
+        'skip': 'paid video in list',
+    }]
+
+    def _get_cheese_entries(self, season_info):
+        for ep_id in traverse_obj(season_info, ('episodes', lambda _, v: v['episode_can_view'], 'id')):
+            yield self._extract_episode(season_info, ep_id)
+
+    def _real_extract(self, url):
+        season_id = self._match_id(url)
+        season_info = self._download_season_info('season_id', season_id)
+
+        return self.playlist_result(
+            self._get_cheese_entries(season_info), season_id,
+            **traverse_obj(season_info, {
+                'title': ('title', {str}),
+                'description': ('subtitle', {str}),
+            }))
 
 
 class BilibiliSpaceBaseIE(InfoExtractor):

yt_dlp/extractor/nebula.py

@@ -3,140 +3,140 @@ import json
 
 from .common import InfoExtractor
 from ..networking.exceptions import HTTPError
-from ..utils import ExtractorError, make_archive_id, parse_iso8601, remove_start
+from ..utils import (
+    ExtractorError,
+    int_or_none,
+    make_archive_id,
+    parse_iso8601,
+    smuggle_url,
+    try_call,
+    unsmuggle_url,
+    update_url_query,
+    url_or_none,
+    urljoin,
+)
+from ..utils.traversal import traverse_obj
 
 _BASE_URL_RE = r'https?://(?:www\.|beta\.)?(?:watchnebula\.com|nebula\.app|nebula\.tv)'
 
 
 class NebulaBaseIE(InfoExtractor):
     _NETRC_MACHINE = 'watchnebula'
+    _token = _api_token = None
 
-    _nebula_api_token = None
+    def _perform_login(self, username, password):
-    _nebula_bearer_token = None
-
-    def _perform_nebula_auth(self, username, password):
-        if not username or not password:
-            self.raise_login_required(method='password')
-
-        data = json.dumps({'email': username, 'password': password}).encode('utf8')
-        response = self._download_json(
-            'https://api.watchnebula.com/api/v1/auth/login/',
-            data=data, fatal=False, video_id=None,
-            headers={
-                'content-type': 'application/json',
-                # Submitting the 'sessionid' cookie always causes a 403 on auth endpoint
-                'cookie': ''
-            },
-            note='Logging in to Nebula with supplied credentials',
-            errnote='Authentication failed or rejected')
-        if not response or not response.get('key'):
-            self.raise_login_required(method='password')
-
-        return response['key']
-
-    def _call_nebula_api(self, url, video_id=None, method='GET', auth_type='api', note=''):
-        assert method in ('GET', 'POST',)
-        assert auth_type in ('api', 'bearer',)
-
-        def inner_call():
-            authorization = f'Token {self._nebula_api_token}' if auth_type == 'api' else f'Bearer {self._nebula_bearer_token}'
-            return self._download_json(
-                url, video_id, note=note, headers={'Authorization': authorization},
-                data=b'' if method == 'POST' else None)
-
         try:
-            return inner_call()
+            response = self._download_json(
-        except ExtractorError as exc:
+                'https://nebula.tv/auth/login/', None,
-            # if 401 or 403, attempt credential re-auth and retry
+                'Logging in to Nebula', 'Login failed',
-            if exc.cause and isinstance(exc.cause, HTTPError) and exc.cause.status in (401, 403):
+                data=json.dumps({'email': username, 'password': password}).encode(),
-                self.to_screen(f'Reauthenticating to Nebula and retrying, because last {auth_type} call resulted in error {exc.cause.code}')
+                headers={'content-type': 'application/json'})
-                self._perform_login()
+        except ExtractorError as e:
-                return inner_call()
+            if isinstance(e.cause, HTTPError) and e.cause.status == 400:
-            else:
+                raise ExtractorError('Login failed: Invalid username or password', expected=True)
+            raise
+        self._api_token = traverse_obj(response, ('key', {str}))
+        if not self._api_token:
+            raise ExtractorError('Login failed: No token')
+
+    def _call_api(self, *args, **kwargs):
+        if self._token:
+            kwargs.setdefault('headers', {})['Authorization'] = f'Bearer {self._token}'
+        try:
+            return self._download_json(*args, **kwargs)
+        except ExtractorError as e:
+            if not isinstance(e.cause, HTTPError) or e.cause.status not in (401, 403):
+                raise
+            self.to_screen(
+                f'Reauthorizing with Nebula and retrying, because last API call resulted in error {e.cause.status}')
+            self._real_initialize()
+            if self._token:
+                kwargs.setdefault('headers', {})['Authorization'] = f'Bearer {self._token}'
+            return self._download_json(*args, **kwargs)
+
+    def _real_initialize(self):
+        if not self._api_token:
+            self._api_token = try_call(
+                lambda: self._get_cookies('https://nebula.tv')['nebula_auth.apiToken'].value)
+        self._token = self._download_json(
+            'https://users.api.nebula.app/api/v1/authorization/', None,
+            headers={'Authorization': f'Token {self._api_token}'} if self._api_token else None,
+            note='Authorizing to Nebula', data=b'')['token']
+
+    def _extract_formats(self, content_id, slug):
+        for retry in (False, True):
+            try:
+                fmts, subs = self._extract_m3u8_formats_and_subtitles(
+                    f'https://content.api.nebula.app/{content_id.split(":")[0]}s/{content_id}/manifest.m3u8',
+                    slug, 'mp4', query={
+                        'token': self._token,
+                        'app_version': '23.10.0',
+                        'platform': 'ios',
+                    })
+                return {'formats': fmts, 'subtitles': subs}
+            except ExtractorError as e:
+                if isinstance(e.cause, HTTPError) and e.cause.status == 401:
+                    self.raise_login_required()
+                if not retry and isinstance(e.cause, HTTPError) and e.cause.status == 403:
+                    self.to_screen('Reauthorizing with Nebula and retrying, because fetching video resulted in error')
+                    self._real_initialize()
+                    continue
                 raise
 
-    def _fetch_nebula_bearer_token(self):
+    def _extract_video_metadata(self, episode):
-        """
+        channel_url = traverse_obj(
-        Get a Bearer token for the Nebula API. This will be required to fetch video meta data.
+            episode, (('channel_slug', 'class_slug'), {lambda x: urljoin('https://nebula.tv/', x)}), get_all=False)
-        """
-        response = self._call_nebula_api('https://api.watchnebula.com/api/v1/authorization/',
-                                         method='POST',
-                                         note='Authorizing to Nebula')
-        return response['token']
-
-    def _fetch_video_formats(self, slug):
-        stream_info = self._call_nebula_api(f'https://content.api.nebula.app/video/{slug}/stream/',
-                                            video_id=slug,
-                                            auth_type='bearer',
-                                            note='Fetching video stream info')
-        manifest_url = stream_info['manifest']
-        return self._extract_m3u8_formats_and_subtitles(manifest_url, slug, 'mp4')
-
-    def _build_video_info(self, episode):
-        fmts, subs = self._fetch_video_formats(episode['slug'])
-        channel_slug = episode['channel_slug']
-        channel_title = episode['channel_title']
-        zype_id = episode.get('zype_id')
         return {
-            'id': remove_start(episode['id'], 'video_episode:'),
+            'id': episode['id'].partition(':')[2],
-            'display_id': episode['slug'],
+            **traverse_obj(episode, {
-            'formats': fmts,
+                'display_id': 'slug',
-            'subtitles': subs,
+                'title': 'title',
-            'webpage_url': f'https://nebula.tv/{episode["slug"]}',
+                'description': 'description',
-            'title': episode['title'],
+                'timestamp': ('published_at', {parse_iso8601}),
-            'description': episode['description'],
+                'duration': ('duration', {int_or_none}),
-            'timestamp': parse_iso8601(episode['published_at']),
+                'channel_id': 'channel_slug',
-            'thumbnails': [{
+                'uploader_id': 'channel_slug',
-                # 'id': tn.get('name'),  # this appears to be null
+                'channel': 'channel_title',
-                'url': tn['original'],
+                'uploader': 'channel_title',
-                'height': key,
+                'series': 'channel_title',
-            } for key, tn in episode['assets']['thumbnail'].items()],
+                'creator': 'channel_title',
-            'duration': episode['duration'],
+                'thumbnail': ('images', 'thumbnail', 'src', {url_or_none}),
-            'channel': channel_title,
+                'episode_number': ('order', {int_or_none}),
-            'channel_id': channel_slug,
+                # Old code was wrongly setting extractor_key from NebulaSubscriptionsIE
-            'channel_url': f'https://nebula.tv/{channel_slug}',
+                '_old_archive_ids': ('zype_id', {lambda x: [
-            'uploader': channel_title,
+                    make_archive_id(NebulaIE, x), make_archive_id(NebulaSubscriptionsIE, x)] if x else None}),
-            'uploader_id': channel_slug,
+            }),
-            'uploader_url': f'https://nebula.tv/{channel_slug}',
+            'channel_url': channel_url,
-            'series': channel_title,
+            'uploader_url': channel_url,
-            'creator': channel_title,
-            'extractor_key': NebulaIE.ie_key(),
-            'extractor': NebulaIE.IE_NAME,
-            '_old_archive_ids': [make_archive_id(NebulaIE, zype_id)] if zype_id else None,
         }
 
-    def _perform_login(self, username=None, password=None):
-        self._nebula_api_token = self._perform_nebula_auth(username, password)
-        self._nebula_bearer_token = self._fetch_nebula_bearer_token()
-
 
 class NebulaIE(NebulaBaseIE):
     _VALID_URL = rf'{_BASE_URL_RE}/videos/(?P<id>[-\w]+)'
-    _TESTS = [
+    _TESTS = [{
-        {
         'url': 'https://nebula.tv/videos/that-time-disney-remade-beauty-and-the-beast',
-            'md5': '14944cfee8c7beeea106320c47560efc',
         'info_dict': {
             'id': '84ed544d-4afd-4723-8cd5-2b95261f0abf',
             'ext': 'mp4',
             'title': 'That Time Disney Remade Beauty and the Beast',
-                'description': 'Note: this video was originally posted on YouTube with the sponsor read included. We weren’t able to remove it without reducing video quality, so it’s presented here in its original context.',
+            'description': 'md5:2aae3c4cfc5ee09a1ecdff0909618cf4',
             'upload_date': '20180731',
             'timestamp': 1533009600,
             'channel': 'Lindsay Ellis',
             'channel_id': 'lindsayellis',
             'uploader': 'Lindsay Ellis',
             'uploader_id': 'lindsayellis',
-                'uploader_url': 'https://nebula.tv/lindsayellis',
+            'uploader_url': r're:https://nebula\.(tv|app)/lindsayellis',
             'series': 'Lindsay Ellis',
             'display_id': 'that-time-disney-remade-beauty-and-the-beast',
-                'channel_url': 'https://nebula.tv/lindsayellis',
+            'channel_url': r're:https://nebula\.(tv|app)/lindsayellis',
             'creator': 'Lindsay Ellis',
             'duration': 2212,
-                'thumbnail': r're:https://\w+\.cloudfront\.net/[\w-]+\.jpeg?.*',
+            'thumbnail': r're:https://\w+\.cloudfront\.net/[\w-]+',
+            '_old_archive_ids': ['nebula 5c271b40b13fd613090034fd', 'nebulasubscriptions 5c271b40b13fd613090034fd'],
         },
-        },
+        'params': {'skip_download': 'm3u8'},
-        {
+    }, {
         'url': 'https://nebula.tv/videos/the-logistics-of-d-day-landing-craft-how-the-allies-got-ashore',
         'md5': 'd05739cf6c38c09322422f696b569c23',
         'info_dict': {
@@ -156,10 +156,11 @@ class NebulaIE(NebulaBaseIE):
             'duration': 841,
             'channel_url': 'https://nebula.tv/d-day',
             'uploader_url': 'https://nebula.tv/d-day',
-                'thumbnail': r're:https://\w+\.cloudfront\.net/[\w-]+\.jpeg?.*',
+            'thumbnail': r're:https://\w+\.cloudfront\.net/[\w-]+',
+            '_old_archive_ids': ['nebula 5e7e78171aaf320001fbd6be', 'nebulasubscriptions 5e7e78171aaf320001fbd6be'],
         },
-        },
+        'params': {'skip_download': 'm3u8'},
-        {
+    }, {
         'url': 'https://nebula.tv/videos/money-episode-1-the-draw',
         'md5': 'ebe28a7ad822b9ee172387d860487868',
         'info_dict': {
@@ -178,55 +179,130 @@ class NebulaIE(NebulaBaseIE):
             'channel_url': 'https://nebula.tv/tom-scott-presents-money',
             'series': 'Tom Scott Presents: Money',
             'display_id': 'money-episode-1-the-draw',
-                'thumbnail': r're:https://\w+\.cloudfront\.net/[\w-]+\.jpeg?.*',
+            'thumbnail': r're:https://\w+\.cloudfront\.net/[\w-]+',
             'creator': 'Tom Scott Presents: Money',
+            '_old_archive_ids': ['nebula 5e779ebdd157bc0001d1c75a', 'nebulasubscriptions 5e779ebdd157bc0001d1c75a'],
         },
-        },
+        'params': {'skip_download': 'm3u8'},
-        {
+    }, {
         'url': 'https://watchnebula.com/videos/money-episode-1-the-draw',
         'only_matching': True,
+    }, {
+        'url': 'https://nebula.tv/videos/tldrnewseu-did-the-us-really-blow-up-the-nordstream-pipelines',
+        'info_dict': {
+            'id': 'e389af9d-1dab-44f2-8788-ee24deb7ff0d',
+            'ext': 'mp4',
+            'display_id': 'tldrnewseu-did-the-us-really-blow-up-the-nordstream-pipelines',
+            'title': 'Did the US Really Blow Up the NordStream Pipelines?',
+            'description': 'md5:b4e2a14e3ff08f546a3209c75261e789',
+            'upload_date': '20230223',
+            'timestamp': 1677144070,
+            'channel': 'TLDR News EU',
+            'channel_id': 'tldrnewseu',
+            'uploader': 'TLDR News EU',
+            'uploader_id': 'tldrnewseu',
+            'uploader_url': r're:https://nebula\.(tv|app)/tldrnewseu',
+            'duration': 524,
+            'channel_url': r're:https://nebula\.(tv|app)/tldrnewseu',
+            'series': 'TLDR News EU',
+            'thumbnail': r're:https://\w+\.cloudfront\.net/[\w-]+',
+            'creator': 'TLDR News EU',
+            '_old_archive_ids': ['nebula 63f64c74366fcd00017c1513', 'nebulasubscriptions 63f64c74366fcd00017c1513'],
         },
-        {
+        'params': {'skip_download': 'm3u8'},
+    }, {
         'url': 'https://beta.nebula.tv/videos/money-episode-1-the-draw',
         'only_matching': True,
-        },
+    }]
-    ]
-
-    def _fetch_video_metadata(self, slug):
-        return self._call_nebula_api(f'https://content.api.nebula.app/video/{slug}/',
-                                     video_id=slug,
-                                     auth_type='bearer',
-                                     note='Fetching video meta data')
 
     def _real_extract(self, url):
         slug = self._match_id(url)
-        video = self._fetch_video_metadata(slug)
+        url, smuggled_data = unsmuggle_url(url, {})
-        return self._build_video_info(video)
+        if smuggled_data.get('id'):
+            return {
+                'id': smuggled_data['id'],
+                'display_id': slug,
+                'title': '',
+                **self._extract_formats(smuggled_data['id'], slug),
+            }
+
+        metadata = self._call_api(
+            f'https://content.api.nebula.app/content/videos/{slug}',
+            slug, note='Fetching video metadata')
+        return {
+            **self._extract_video_metadata(metadata),
+            **self._extract_formats(metadata['id'], slug),
+        }
+
+
+class NebulaClassIE(NebulaBaseIE):
+    IE_NAME = 'nebula:class'
+    _VALID_URL = rf'{_BASE_URL_RE}/(?P<id>[-\w]+)/(?P<ep>\d+)'
+    _TESTS = [{
+        'url': 'https://nebula.tv/copyright-for-fun-and-profit/14',
+        'info_dict': {
+            'id': 'd7432cdc-c608-474d-942c-f74345daed7b',
+            'ext': 'mp4',
+            'display_id': '14',
+            'channel_url': 'https://nebula.tv/copyright-for-fun-and-profit',
+            'episode_number': 14,
+            'thumbnail': 'https://dj423fildxgac.cloudfront.net/d533718d-9307-42d4-8fb0-e283285e99c9',
+            'uploader_url': 'https://nebula.tv/copyright-for-fun-and-profit',
+            'duration': 646,
+            'episode': 'Episode 14',
+            'title': 'Photos, Sculpture, and Video',
+        },
+        'params': {'skip_download': 'm3u8'},
+    }]
+
+    def _real_extract(self, url):
+        slug, episode = self._match_valid_url(url).group('id', 'ep')
+        url, smuggled_data = unsmuggle_url(url, {})
+        if smuggled_data.get('id'):
+            return {
+                'id': smuggled_data['id'],
+                'display_id': slug,
+                'title': '',
+                **self._extract_formats(smuggled_data['id'], slug),
+            }
+
+        metadata = self._call_api(
+            f'https://content.api.nebula.app/content/{slug}/{episode}/?include=lessons',
+            slug, note='Fetching video metadata')
+        return {
+            **self._extract_video_metadata(metadata),
+            **self._extract_formats(metadata['id'], slug),
+        }
 
 
 class NebulaSubscriptionsIE(NebulaBaseIE):
     IE_NAME = 'nebula:subscriptions'
-    _VALID_URL = rf'{_BASE_URL_RE}/myshows'
+    _VALID_URL = rf'{_BASE_URL_RE}/(?P<id>myshows|library/latest-videos)'
-    _TESTS = [
+    _TESTS = [{
-        {
         'url': 'https://nebula.tv/myshows',
         'playlist_mincount': 1,
         'info_dict': {
             'id': 'myshows',
         },
-        },
+    }]
-    ]
 
     def _generate_playlist_entries(self):
-        next_url = 'https://content.watchnebula.com/library/video/?page_size=100'
+        next_url = update_url_query('https://content.api.nebula.app/video_episodes/', {
-        page_num = 1
+            'following': 'true',
-        while next_url:
+            'include': 'engagement',
-            channel = self._call_nebula_api(next_url, 'myshows', auth_type='bearer',
+            'ordering': '-published_at',
-                                            note=f'Retrieving subscriptions page {page_num}')
+        })
+        for page_num in itertools.count(1):
+            channel = self._call_api(
+                next_url, 'myshows', note=f'Retrieving subscriptions page {page_num}')
             for episode in channel['results']:
-                yield self._build_video_info(episode)
+                metadata = self._extract_video_metadata(episode)
-            next_url = channel['next']
+                yield self.url_result(smuggle_url(
-            page_num += 1
+                    f'https://nebula.tv/videos/{metadata["display_id"]}',
+                    {'id': episode['id']}), NebulaIE, url_transparent=True, **metadata)
+            next_url = channel.get('next')
+            if not next_url:
+                return
 
     def _real_extract(self, url):
         return self.playlist_result(self._generate_playlist_entries(), 'myshows')
@@ -234,9 +310,8 @@ class NebulaSubscriptionsIE(NebulaBaseIE):
 
 class NebulaChannelIE(NebulaBaseIE):
     IE_NAME = 'nebula:channel'
-    _VALID_URL = rf'{_BASE_URL_RE}/(?!myshows|videos/)(?P<id>[-\w]+)'
+    _VALID_URL = rf'{_BASE_URL_RE}/(?!myshows|library|videos/)(?P<id>[-\w]+)/?(?:$|[?#])'
-    _TESTS = [
+    _TESTS = [{
-        {
         'url': 'https://nebula.tv/tom-scott-presents-money',
         'info_dict': {
             'id': 'tom-scott-presents-money',
@@ -252,30 +327,57 @@ class NebulaChannelIE(NebulaBaseIE):
             'description': 'Enjoy these hottest of takes on Disney, Transformers, and Musicals.',
         },
         'playlist_mincount': 2,
+    }, {
+        'url': 'https://nebula.tv/johnnyharris',
+        'info_dict': {
+            'id': 'johnnyharris',
+            'title': 'Johnny Harris',
+            'description': 'I make videos about maps and many other things.',
         },
-    ]
+        'playlist_mincount': 90,
+    }, {
+        'url': 'https://nebula.tv/copyright-for-fun-and-profit',
+        'info_dict': {
+            'id': 'copyright-for-fun-and-profit',
+            'title': 'Copyright for Fun and Profit',
+            'description': 'md5:6690248223eed044a9f11cd5a24f9742',
+        },
+        'playlist_count': 23,
+    }]
 
-    def _generate_playlist_entries(self, collection_id, channel):
+    def _generate_playlist_entries(self, collection_id, collection_slug):
-        episodes = channel['episodes']['results']
+        next_url = f'https://content.api.nebula.app/video_channels/{collection_id}/video_episodes/?ordering=-published_at'
-        for page_num in itertools.count(2):
+        for page_num in itertools.count(1):
-            for episode in episodes:
+            episodes = self._call_api(next_url, collection_slug, note=f'Retrieving channel page {page_num}')
-                yield self._build_video_info(episode)
+            for episode in episodes['results']:
-            next_url = channel['episodes']['next']
+                metadata = self._extract_video_metadata(episode)
+                yield self.url_result(smuggle_url(
+                    episode.get('share_url') or f'https://nebula.tv/videos/{metadata["display_id"]}',
+                    {'id': episode['id']}), NebulaIE, url_transparent=True, **metadata)
+            next_url = episodes.get('next')
             if not next_url:
                 break
-            channel = self._call_nebula_api(next_url, collection_id, auth_type='bearer',
+
-                                            note=f'Retrieving channel page {page_num}')
+    def _generate_class_entries(self, channel):
-            episodes = channel['episodes']['results']
+        for lesson in channel['lessons']:
+            metadata = self._extract_video_metadata(lesson)
+            yield self.url_result(smuggle_url(
+                lesson.get('share_url') or f'https://nebula.tv/{metadata["class_slug"]}/{metadata["slug"]}',
+                {'id': lesson['id']}), NebulaClassIE, url_transparent=True, **metadata)
 
     def _real_extract(self, url):
-        collection_id = self._match_id(url)
+        collection_slug = self._match_id(url)
-        channel_url = f'https://content.watchnebula.com/video/channels/{collection_id}/'
+        channel = self._call_api(
-        channel = self._call_nebula_api(channel_url, collection_id, auth_type='bearer', note='Retrieving channel')
+            f'https://content.api.nebula.app/content/{collection_slug}/?include=lessons',
-        channel_details = channel['details']
+            collection_slug, note='Retrieving channel')
+
+        if channel.get('type') == 'class':
+            entries = self._generate_class_entries(channel)
+        else:
+            entries = self._generate_playlist_entries(channel['id'], collection_slug)
 
         return self.playlist_result(
-            entries=self._generate_playlist_entries(collection_id, channel),
+            entries=entries,
-            playlist_id=collection_id,
+            playlist_id=collection_slug,
-            playlist_title=channel_details['title'],
+            playlist_title=channel.get('title'),
-            playlist_description=channel_details['description']
+            playlist_description=channel.get('description'))
-        )

yt_dlp/extractor/vvvvid.py

@@ -1,3 +1,4 @@
+import functools
 import re
 
 from .common import InfoExtractor
@@ -14,21 +15,21 @@ class VVVVIDIE(InfoExtractor):
     _VALID_URL = r'%s(?P<show_id>\d+)/[^/]+/(?P<season_id>\d+)/(?P<id>[0-9]+)' % _VALID_URL_BASE
     _TESTS = [{
         # video_type == 'video/vvvvid'
-        'url': 'https://www.vvvvid.it/#!show/434/perche-dovrei-guardarlo-di-dario-moccia/437/489048/ping-pong',
+        'url': 'https://www.vvvvid.it/show/498/the-power-of-computing/518/505692/playstation-vr-cambiera-il-nostro-modo-di-giocare',
-        'md5': 'b8d3cecc2e981adc3835adf07f6df91b',
         'info_dict': {
-            'id': '489048',
+            'id': '505692',
             'ext': 'mp4',
-            'title': 'Ping Pong',
+            'title': 'Playstation VR cambierà il nostro modo di giocare',
-            'duration': 239,
+            'duration': 93,
-            'series': '"Perché dovrei guardarlo?" di Dario Moccia',
+            'series': 'The Power of Computing',
-            'season_id': '437',
+            'season_id': '518',
-            'episode': 'Ping Pong',
+            'episode': 'Playstation VR cambierà il nostro modo di giocare',
-            'episode_number': 1,
+            'episode_number': None,
-            'episode_id': '3334',
+            'episode_id': '4747',
             'view_count': int,
             'like_count': int,
             'repost_count': int,
+            'thumbnail': 'https://static.vvvvid.it/img/zoomin/28CA2409-E663-34F0-2B02E72356556EA3_500k.jpg',
         },
         'params': {
             'skip_download': True,
@@ -36,7 +37,6 @@ class VVVVIDIE(InfoExtractor):
     }, {
         # video_type == 'video/rcs'
         'url': 'https://www.vvvvid.it/#!show/376/death-note-live-action/377/482493/episodio-01',
-        'md5': '33e0edfba720ad73a8782157fdebc648',
         'info_dict': {
             'id': '482493',
             'ext': 'mp4',
@@ -45,6 +45,7 @@ class VVVVIDIE(InfoExtractor):
         'params': {
             'skip_download': True,
         },
+        'skip': 'Every video/rcs is not working even in real website',
     }, {
         # video_type == 'video/youtube'
         'url': 'https://www.vvvvid.it/show/404/one-punch-man/406/486683/trailer',
@@ -55,19 +56,54 @@ class VVVVIDIE(InfoExtractor):
             'title': 'Trailer',
             'upload_date': '20150906',
             'description': 'md5:a5e802558d35247fee285875328c0b80',
-            'uploader_id': 'BandaiVisual',
+            'uploader_id': '@EMOTIONLabelChannel',
-            'uploader': 'BANDAI NAMCO Arts Channel',
+            'uploader': 'EMOTION Label Channel',
+            'episode_number': None,
+            'episode_id': '3115',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'availability': str,
+            'categories': list,
+            'age_limit': 0,
+            'channel': 'EMOTION Label Channel',
+            'channel_follower_count': int,
+            'channel_id': 'UCQ5URCSs1f5Cz9rh-cDGxNQ',
+            'channel_url': 'https://www.youtube.com/channel/UCQ5URCSs1f5Cz9rh-cDGxNQ',
+            'comment_count': int,
+            'duration': 133,
+            'episode': 'Trailer',
+            'heatmap': list,
+            'live_status': 'not_live',
+            'playable_in_embed': True,
+            'season_id': '406',
+            'series': 'One-Punch Man',
+            'tags': list,
+            'uploader_url': 'https://www.youtube.com/@EMOTIONLabelChannel',
+            'thumbnail': 'https://i.ytimg.com/vi/RzmFKUDOUgw/maxresdefault.jpg',
         },
         'params': {
             'skip_download': True,
         },
     }, {
         # video_type == 'video/dash'
-        'url': 'https://www.vvvvid.it/show/683/made-in-abyss/1542/693786/nanachi',
+        'url': 'https://www.vvvvid.it/show/844/le-bizzarre-avventure-di-jojo-vento-aureo/938/527551/golden-wind',
         'info_dict': {
-            'id': '693786',
+            'id': '527551',
             'ext': 'mp4',
-            'title': 'Nanachi',
+            'title': 'Golden Wind',
+            'duration': 1430,
+            'series': 'Le bizzarre avventure di Jojo - Vento Aureo',
+            'season_id': '938',
+            'episode': 'Golden Wind',
+            'episode_number': 1,
+            'episode_id': '9089',
+            'view_count': int,
+            'like_count': int,
+            'repost_count': int,
+            'thumbnail': 'https://static.vvvvid.it/img/thumbs/Dynit/Jojo/Jojo_S05Ep01-t.jpg',
+            'season': 'Season 5',
+            'season_number': 5,
         },
         'params': {
             'skip_download': True,
@@ -79,10 +115,17 @@ class VVVVIDIE(InfoExtractor):
     }]
     _conn_id = None
 
+    @functools.cached_property
+    def _headers(self):
+        return {
+            **self.geo_verification_headers(),
+            'User-Agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/95.0.4638.50 Safari/537.37',
+        }
+
     def _real_initialize(self):
         self._conn_id = self._download_json(
             'https://www.vvvvid.it/user/login',
-            None, headers=self.geo_verification_headers())['data']['conn_id']
+            None, headers=self._headers)['data']['conn_id']
 
     def _download_info(self, show_id, path, video_id, fatal=True, query=None):
         q = {
@@ -92,7 +135,7 @@ class VVVVIDIE(InfoExtractor):
             q.update(query)
         response = self._download_json(
             'https://www.vvvvid.it/vvvvid/ondemand/%s/%s' % (show_id, path),
-            video_id, headers=self.geo_verification_headers(), query=q, fatal=fatal)
+            video_id, headers=self._headers, query=q, fatal=fatal)
         if not (response or fatal):
             return
         if response.get('result') == 'error':
@@ -219,7 +262,7 @@ class VVVVIDIE(InfoExtractor):
                     embed_code, video_id, 'mp4', m3u8_id='hls', fatal=False))
             else:
                 formats.extend(self._extract_wowza_formats(
-                    'http://sb.top-ix.org/videomg/_definst_/mp4:%s/playlist.m3u8' % embed_code, video_id))
+                    'http://sb.top-ix.org/videomg/_definst_/mp4:%s/playlist.m3u8' % embed_code, video_id, skip_protocols=['f4m']))
             metadata_from_url(embed_code)
 
         if not is_youtube: