yt_dlp/extractor/afreecatv.py

   1 import functools
   2
   3 from .common import InfoExtractor
   4 from ..networking import Request
   5 from ..utils import (
   6     ExtractorError,
   7     OnDemandPagedList,
   8     UserNotLive,
   9     determine_ext,
  10     filter_dict,
  11     int_or_none,
  12     orderedSet,
  13     unified_timestamp,
  14     url_or_none,
  15     urlencode_postdata,
  16     urljoin,
  17 )
  18 from ..utils.traversal import traverse_obj
  19
  20
  21 class AfreecaTVBaseIE(InfoExtractor):
  22     _NETRC_MACHINE = 'afreecatv'
  23
  24     def _perform_login(self, username, password):
  25         login_form = {
  26             'szWork': 'login',
  27             'szType': 'json',
  28             'szUid': username,
  29             'szPassword': password,
  30             'isSaveId': 'false',
  31             'szScriptVar': 'oLoginRet',
  32             'szAction': '',
  33         }
  34
  35         response = self._download_json(
  36             'https://login.sooplive.co.kr/app/LoginAction.php', None,
  37             'Logging in', data=urlencode_postdata(login_form))
  38
  39         _ERRORS = {
  40             -4: 'Your account has been suspended due to a violation of our terms and policies.',
  41             -5: 'https://member.sooplive.co.kr/app/user_delete_progress.php',
  42             -6: 'https://login.sooplive.co.kr/membership/changeMember.php',
  43             -8: "Hello! Soop here.\nThe username you have entered belongs to \n an account that requires a legal guardian's consent. \nIf you wish to use our services without restriction, \nplease make sure to go through the necessary verification process.",
  44             -9: 'https://member.sooplive.co.kr/app/pop_login_block.php',
  45             -11: 'https://login.sooplive.co.kr/afreeca/second_login.php',
  46             -12: 'https://member.sooplive.co.kr/app/user_security.php',
  47             0: 'The username does not exist or you have entered the wrong password.',
  48             -1: 'The username does not exist or you have entered the wrong password.',
  49             -3: 'You have entered your username/password incorrectly.',
  50             -7: 'You cannot use your Global Soop account to access Korean Soop.',
  51             -10: 'Sorry for the inconvenience. \nYour account has been blocked due to an unauthorized access. \nPlease contact our Help Center for assistance.',
  52             -32008: 'You have failed to log in. Please contact our Help Center.',
  53         }
  54
  55         result = int_or_none(response.get('RESULT'))
  56         if result != 1:
  57             error = _ERRORS.get(result, 'You have failed to log in.')
  58             raise ExtractorError(
  59                 f'Unable to login: {self.IE_NAME} said: {error}',
  60                 expected=True)
  61
  62     def _call_api(self, endpoint, display_id, data=None, headers=None, query=None):
  63         return self._download_json(Request(
  64             f'https://api.m.sooplive.co.kr/{endpoint}',
  65             data=data, headers=headers, query=query,
  66             extensions={'legacy_ssl': True}), display_id,
  67             'Downloading API JSON', 'Unable to download API JSON')
  68
  69     @staticmethod
  70     def _fixup_thumb(thumb_url):
  71         if not url_or_none(thumb_url):
  72             return None
  73         # Core would determine_ext as 'php' from the url, so we need to provide the real ext
  74         # See: https://github.com/yt-dlp/yt-dlp/issues/11537
  75         return [{'url': thumb_url, 'ext': 'jpg'}]
  76
  77
  78 class AfreecaTVIE(AfreecaTVBaseIE):
  79     IE_NAME = 'soop'
  80     IE_DESC = 'sooplive.co.kr'
  81     _VALID_URL = r'https?://vod\.(?:sooplive\.co\.kr|afreecatv\.com)/(?:PLAYER/STATION|player)/(?P<id>\d+)/?(?:$|[?#&])'
  82     _TESTS = [{
  83         'url': 'https://vod.sooplive.co.kr/player/96753363',
  84         'info_dict': {
  85             'id': '20230108_9FF5BEE1_244432674_1',
  86             'ext': 'mp4',
  87             'uploader_id': 'rlantnghks',
  88             'uploader': '페이즈으',
  89             'duration': 10840,
  90             'thumbnail': r're:https?://videoimg\.sooplive\.co/.kr/.+',
  91             'upload_date': '20230108',
  92             'timestamp': 1673218805,
  93             'title': '젠지 페이즈',
  94         },
  95         'params': {
  96             'skip_download': True,
  97         },
  98     }, {
  99         # non standard key
 100         'url': 'http://vod.sooplive.co.kr/PLAYER/STATION/20515605',
 101         'info_dict': {
 102             'id': '20170411_BE689A0E_190960999_1_2_h',
 103             'ext': 'mp4',
 104             'title': '혼자사는여자집',
 105             'thumbnail': r're:https?://(?:video|st)img\.sooplive\.co\.kr/.+',
 106             'uploader': '♥이슬이',
 107             'uploader_id': 'dasl8121',
 108             'upload_date': '20170411',
 109             'timestamp': 1491929865,
 110             'duration': 213,
 111         },
 112         'params': {
 113             'skip_download': True,
 114         },
 115     }, {
 116         # adult content
 117         'url': 'https://vod.sooplive.co.kr/player/97267690',
 118         'info_dict': {
 119             'id': '20180327_27901457_202289533_1',
 120             'ext': 'mp4',
 121             'title': '[생]빨개요♥ (part 1)',
 122             'thumbnail': r're:https?://(?:video|st)img\.sooplive\.co\.kr/.+',
 123             'uploader': '[SA]서아',
 124             'uploader_id': 'bjdyrksu',
 125             'upload_date': '20180327',
 126             'duration': 3601,
 127         },
 128         'params': {
 129             'skip_download': True,
 130         },
 131         'skip': 'The VOD does not exist',
 132     }, {
 133         # adult content
 134         'url': 'https://vod.sooplive.co.kr/player/70395877',
 135         'only_matching': True,
 136     }, {
 137         # subscribers only
 138         'url': 'https://vod.sooplive.co.kr/player/104647403',
 139         'only_matching': True,
 140     }, {
 141         # private
 142         'url': 'https://vod.sooplive.co.kr/player/81669846',
 143         'only_matching': True,
 144     }]
 145
 146     def _real_extract(self, url):
 147         video_id = self._match_id(url)
 148         data = self._call_api(
 149             'station/video/a/view', video_id, headers={'Referer': url},
 150             data=urlencode_postdata({
 151                 'nTitleNo': video_id,
 152                 'nApiLevel': 10,
 153             }))['data']
 154
 155         error_code = traverse_obj(data, ('code', {int}))
 156         if error_code == -6221:
 157             raise ExtractorError('The VOD does not exist', expected=True)
 158         elif error_code == -6205:
 159             raise ExtractorError('This VOD is private', expected=True)
 160
 161         common_info = traverse_obj(data, {
 162             'title': ('title', {str}),
 163             'uploader': ('writer_nick', {str}),
 164             'uploader_id': ('bj_id', {str}),
 165             'duration': ('total_file_duration', {int_or_none(scale=1000)}),
 166             'thumbnails': ('thumb', {self._fixup_thumb}),
 167         })
 168
 169         entries = []
 170         for file_num, file_element in enumerate(
 171                 traverse_obj(data, ('files', lambda _, v: url_or_none(v['file']))), start=1):
 172             file_url = file_element['file']
 173             if determine_ext(file_url) == 'm3u8':
 174                 formats = self._extract_m3u8_formats(
 175                     file_url, video_id, 'mp4', m3u8_id='hls',
 176                     note=f'Downloading part {file_num} m3u8 information')
 177             else:
 178                 formats = [{
 179                     'url': file_url,
 180                     'format_id': 'http',
 181                 }]
 182
 183             entries.append({
 184                 **common_info,
 185                 'id': file_element.get('file_info_key') or f'{video_id}_{file_num}',
 186                 'title': f'{common_info.get("title") or "Untitled"} (part {file_num})',
 187                 'formats': formats,
 188                 **traverse_obj(file_element, {
 189                     'duration': ('duration', {int_or_none(scale=1000)}),
 190                     'timestamp': ('file_start', {unified_timestamp}),
 191                 }),
 192             })
 193
 194         if traverse_obj(data, ('adult_status', {str})) == 'notLogin':
 195             if not entries:
 196                 self.raise_login_required(
 197                     'Only users older than 19 are able to watch this video', method='password')
 198             self.report_warning(
 199                 'In accordance with local laws and regulations, underage users are '
 200                 'restricted from watching adult content. Only content suitable for all '
 201                 f'ages will be downloaded. {self._login_hint("password")}')
 202
 203         if not entries and traverse_obj(data, ('sub_upload_type', {str})):
 204             self.raise_login_required('This VOD is for subscribers only', method='password')
 205
 206         if len(entries) == 1:
 207             return {
 208                 **entries[0],
 209                 'title': common_info.get('title'),
 210             }
 211
 212         common_info['timestamp'] = traverse_obj(entries, (..., 'timestamp'), get_all=False)
 213
 214         return self.playlist_result(entries, video_id, multi_video=True, **common_info)
 215
 216
 217 class AfreecaTVCatchStoryIE(AfreecaTVBaseIE):
 218     IE_NAME = 'soop:catchstory'
 219     IE_DESC = 'sooplive.co.kr catch story'
 220     _VALID_URL = r'https?://vod\.(?:sooplive\.co\.kr|afreecatv\.com)/player/(?P<id>\d+)/catchstory'
 221     _TESTS = [{
 222         'url': 'https://vod.sooplive.co.kr/player/103247/catchstory',
 223         'info_dict': {
 224             'id': '103247',
 225         },
 226         'playlist_count': 2,
 227     }]
 228
 229     def _real_extract(self, url):
 230         video_id = self._match_id(url)
 231         data = self._call_api(
 232             'catchstory/a/view', video_id, headers={'Referer': url},
 233             query={'aStoryListIdx': '', 'nStoryIdx': video_id})
 234
 235         return self.playlist_result(self._entries(data), video_id)
 236
 237     def _entries(self, data):
 238         # 'files' is always a list with 1 element
 239         yield from traverse_obj(data, (
 240             'data', lambda _, v: v['story_type'] == 'catch',
 241             'catch_list', lambda _, v: v['files'][0]['file'], {
 242                 'id': ('files', 0, 'file_info_key', {str}),
 243                 'url': ('files', 0, 'file', {url_or_none}),
 244                 'duration': ('files', 0, 'duration', {int_or_none(scale=1000)}),
 245                 'title': ('title', {str}),
 246                 'uploader': ('writer_nick', {str}),
 247                 'uploader_id': ('writer_id', {str}),
 248                 'thumbnails': ('thumb', {self._fixup_thumb}),
 249                 'timestamp': ('write_timestamp', {int_or_none}),
 250             }))
 251
 252
 253 class AfreecaTVLiveIE(AfreecaTVBaseIE):
 254     IE_NAME = 'soop:live'
 255     IE_DESC = 'sooplive.co.kr livestreams'
 256     _VALID_URL = r'https?://play\.(?:sooplive\.co\.kr|afreecatv\.com)/(?P<id>[^/?#]+)(?:/(?P<bno>\d+))?'
 257     _TESTS = [{
 258         'url': 'https://play.sooplive.co.kr/pyh3646/237852185',
 259         'info_dict': {
 260             'id': '237852185',
 261             'ext': 'mp4',
 262             'title': '【 우루과이 오늘은 무슨일이? 】',
 263             'uploader': '박진우[JINU]',
 264             'uploader_id': 'pyh3646',
 265             'timestamp': 1640661495,
 266             'is_live': True,
 267         },
 268         'skip': 'Livestream has ended',
 269     }, {
 270         'url': 'https://play.sooplive.co.kr/pyh3646/237852185',
 271         'only_matching': True,
 272     }, {
 273         'url': 'https://play.sooplive.co.kr/pyh3646',
 274         'only_matching': True,
 275     }]
 276
 277     _LIVE_API_URL = 'https://live.sooplive.co.kr/afreeca/player_live_api.php'
 278     _WORKING_CDNS = [
 279         'gcp_cdn',  # live-global-cdn-v02.sooplive.co.kr
 280         'gs_cdn_pc_app',  # pc-app.stream.sooplive.co.kr
 281         'gs_cdn_mobile_web',  # mobile-web.stream.sooplive.co.kr
 282         'gs_cdn_pc_web',  # pc-web.stream.sooplive.co.kr
 283     ]
 284     _BAD_CDNS = [
 285         'gs_cdn',  # chromecast.afreeca.gscdn.com (cannot resolve)
 286         'gs_cdn_chromecast',  # chromecast.stream.sooplive.co.kr (HTTP Error 400)
 287         'azure_cdn',  # live-global-cdn-v01.sooplive.co.kr (cannot resolve)
 288         'aws_cf',  # live-global-cdn-v03.sooplive.co.kr (cannot resolve)
 289         'kt_cdn',  # kt.stream.sooplive.co.kr (HTTP Error 400)
 290     ]
 291
 292     def _extract_formats(self, channel_info, broadcast_no, aid):
 293         stream_base_url = channel_info.get('RMD') or 'https://livestream-manager.sooplive.co.kr'
 294
 295         # If user has not passed CDN IDs, try API-provided CDN ID followed by other working CDN IDs
 296         default_cdn_ids = orderedSet([
 297             *traverse_obj(channel_info, ('CDN', {str}, all, lambda _, v: v not in self._BAD_CDNS)),
 298             *self._WORKING_CDNS,
 299         ])
 300         cdn_ids = self._configuration_arg('cdn', default_cdn_ids)
 301
 302         for attempt, cdn_id in enumerate(cdn_ids, start=1):
 303             m3u8_url = traverse_obj(self._download_json(
 304                 urljoin(stream_base_url, 'broad_stream_assign.html'), broadcast_no,
 305                 f'Downloading {cdn_id} stream info', f'Unable to download {cdn_id} stream info',
 306                 fatal=False, query={
 307                     'return_type': cdn_id,
 308                     'broad_key': f'{broadcast_no}-common-master-hls',
 309                 }), ('view_url', {url_or_none}))
 310             try:
 311                 return self._extract_m3u8_formats(
 312                     m3u8_url, broadcast_no, 'mp4', m3u8_id='hls', query={'aid': aid},
 313                     headers={'Referer': 'https://play.sooplive.co.kr/'})
 314             except ExtractorError as e:
 315                 if attempt == len(cdn_ids):
 316                     raise
 317                 self.report_warning(
 318                     f'{e.cause or e.msg}. Retrying... (attempt {attempt} of {len(cdn_ids)})')
 319
 320     def _real_extract(self, url):
 321         broadcaster_id, broadcast_no = self._match_valid_url(url).group('id', 'bno')
 322         channel_info = traverse_obj(self._download_json(
 323             self._LIVE_API_URL, broadcaster_id, data=urlencode_postdata({'bid': broadcaster_id})),
 324             ('CHANNEL', {dict})) or {}
 325
 326         broadcaster_id = channel_info.get('BJID') or broadcaster_id
 327         broadcast_no = channel_info.get('BNO') or broadcast_no
 328         if not broadcast_no:
 329             result = channel_info.get('RESULT')
 330             if result == 0:
 331                 raise UserNotLive(video_id=broadcaster_id)
 332             elif result == -6:
 333                 self.raise_login_required(
 334                     'This channel is streaming for subscribers only', method='password')
 335             raise ExtractorError('Unable to extract broadcast number')
 336
 337         password = self.get_param('videopassword')
 338         if channel_info.get('BPWD') == 'Y' and password is None:
 339             raise ExtractorError(
 340                 'This livestream is protected by a password, use the --video-password option',
 341                 expected=True)
 342
 343         token_info = traverse_obj(self._download_json(
 344             self._LIVE_API_URL, broadcast_no, 'Downloading access token for stream',
 345             'Unable to download access token for stream', data=urlencode_postdata(filter_dict({
 346                 'bno': broadcast_no,
 347                 'stream_type': 'common',
 348                 'type': 'aid',
 349                 'quality': 'master',
 350                 'pwd': password,
 351             }))), ('CHANNEL', {dict})) or {}
 352         aid = token_info.get('AID')
 353         if not aid:
 354             result = token_info.get('RESULT')
 355             if result == 0:
 356                 raise ExtractorError('This livestream has ended', expected=True)
 357             elif result == -6:
 358                 self.raise_login_required('This livestream is for subscribers only', method='password')
 359             raise ExtractorError('Unable to extract access token')
 360
 361         formats = self._extract_formats(channel_info, broadcast_no, aid)
 362
 363         station_info = traverse_obj(self._download_json(
 364             'https://st.sooplive.co.kr/api/get_station_status.php', broadcast_no,
 365             'Downloading channel metadata', 'Unable to download channel metadata',
 366             query={'szBjId': broadcaster_id}, fatal=False), {dict}) or {}
 367
 368         return {
 369             'id': broadcast_no,
 370             'title': channel_info.get('TITLE') or station_info.get('station_title'),
 371             'uploader': channel_info.get('BJNICK') or station_info.get('station_name'),
 372             'uploader_id': broadcaster_id,
 373             'timestamp': unified_timestamp(station_info.get('broad_start')),
 374             'formats': formats,
 375             'is_live': True,
 376             'http_headers': {'Referer': url},
 377         }
 378
 379
 380 class AfreecaTVUserIE(AfreecaTVBaseIE):
 381     IE_NAME = 'soop:user'
 382     _VALID_URL = r'https?://ch\.(?:sooplive\.co\.kr|afreecatv\.com)/(?P<id>[^/?#]+)/vods/?(?P<slug_type>[^/?#]+)?'
 383     _TESTS = [{
 384         'url': 'https://ch.sooplive.co.kr/ryuryu24/vods/review',
 385         'info_dict': {
 386             '_type': 'playlist',
 387             'id': 'ryuryu24',
 388             'title': 'ryuryu24 - review',
 389         },
 390         'playlist_count': 218,
 391     }, {
 392         'url': 'https://ch.sooplive.co.kr/parang1995/vods/highlight',
 393         'info_dict': {
 394             '_type': 'playlist',
 395             'id': 'parang1995',
 396             'title': 'parang1995 - highlight',
 397         },
 398         'playlist_count': 997,
 399     }, {
 400         'url': 'https://ch.sooplive.co.kr/ryuryu24/vods',
 401         'info_dict': {
 402             '_type': 'playlist',
 403             'id': 'ryuryu24',
 404             'title': 'ryuryu24 - all',
 405         },
 406         'playlist_count': 221,
 407     }, {
 408         'url': 'https://ch.sooplive.co.kr/ryuryu24/vods/balloonclip',
 409         'info_dict': {
 410             '_type': 'playlist',
 411             'id': 'ryuryu24',
 412             'title': 'ryuryu24 - balloonclip',
 413         },
 414         'playlist_count': 0,
 415     }]
 416     _PER_PAGE = 60
 417
 418     def _fetch_page(self, user_id, user_type, page):
 419         page += 1
 420         info = self._download_json(f'https://chapi.sooplive.co.kr/api/{user_id}/vods/{user_type}', user_id,
 421                                    query={'page': page, 'per_page': self._PER_PAGE, 'orderby': 'reg_date'},
 422                                    note=f'Downloading {user_type} video page {page}')
 423         for item in info['data']:
 424             yield self.url_result(
 425                 f'https://vod.sooplive.co.kr/player/{item["title_no"]}/', AfreecaTVIE, item['title_no'])
 426
 427     def _real_extract(self, url):
 428         user_id, user_type = self._match_valid_url(url).group('id', 'slug_type')
 429         user_type = user_type or 'all'
 430         entries = OnDemandPagedList(functools.partial(self._fetch_page, user_id, user_type), self._PER_PAGE)
 431         return self.playlist_result(entries, user_id, f'{user_id} - {user_type}')