yt_dlp/extractor/afreecatv.py

   1 import functools
   2 import re
   3
   4 from .common import InfoExtractor
   5 from ..utils import (
   6     ExtractorError,
   7     OnDemandPagedList,
   8     date_from_str,
   9     determine_ext,
  10     int_or_none,
  11     qualities,
  12     traverse_obj,
  13     unified_strdate,
  14     unified_timestamp,
  15     update_url_query,
  16     url_or_none,
  17     urlencode_postdata,
  18     xpath_text,
  19 )
  20
  21
  22 class AfreecaTVIE(InfoExtractor):
  23     IE_NAME = 'afreecatv'
  24     IE_DESC = 'afreecatv.com'
  25     _VALID_URL = r'''(?x)
  26                     https?://
  27                         (?:
  28                             (?:(?:live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
  29                             (?:
  30                                 /app/(?:index|read_ucc_bbs)\.cgi|
  31                                 /player/[Pp]layer\.(?:swf|html)
  32                             )\?.*?\bnTitleNo=|
  33                             vod\.afreecatv\.com/(PLAYER/STATION|player)/
  34                         )
  35                         (?P<id>\d+)
  36                     '''
  37     _NETRC_MACHINE = 'afreecatv'
  38     _TESTS = [{
  39         'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
  40         'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
  41         'info_dict': {
  42             'id': '36164052',
  43             'ext': 'mp4',
  44             'title': '데일리 에이프릴 요정들의 시상식!',
  45             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  46             'uploader': 'dailyapril',
  47             'uploader_id': 'dailyapril',
  48             'upload_date': '20160503',
  49         },
  50         'skip': 'Video is gone',
  51     }, {
  52         'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
  53         'info_dict': {
  54             'id': '36153164',
  55             'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  56             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  57             'uploader': 'dailyapril',
  58             'uploader_id': 'dailyapril',
  59         },
  60         'playlist_count': 2,
  61         'playlist': [{
  62             'md5': 'd8b7c174568da61d774ef0203159bf97',
  63             'info_dict': {
  64                 'id': '36153164_1',
  65                 'ext': 'mp4',
  66                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  67                 'upload_date': '20160502',
  68             },
  69         }, {
  70             'md5': '58f2ce7f6044e34439ab2d50612ab02b',
  71             'info_dict': {
  72                 'id': '36153164_2',
  73                 'ext': 'mp4',
  74                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  75                 'upload_date': '20160502',
  76             },
  77         }],
  78         'skip': 'Video is gone',
  79     }, {
  80         # non standard key
  81         'url': 'http://vod.afreecatv.com/PLAYER/STATION/20515605',
  82         'info_dict': {
  83             'id': '20170411_BE689A0E_190960999_1_2_h',
  84             'ext': 'mp4',
  85             'title': '혼자사는여자집',
  86             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  87             'uploader': '♥이슬이',
  88             'uploader_id': 'dasl8121',
  89             'upload_date': '20170411',
  90             'duration': 213,
  91         },
  92         'params': {
  93             'skip_download': True,
  94         },
  95     }, {
  96         # adult content
  97         'url': 'https://vod.afreecatv.com/player/97267690',
  98         'info_dict': {
  99             'id': '20180327_27901457_202289533_1',
 100             'ext': 'mp4',
 101             'title': '[생]빨개요♥ (part 1)',
 102             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
 103             'uploader': '[SA]서아',
 104             'uploader_id': 'bjdyrksu',
 105             'upload_date': '20180327',
 106             'duration': 3601,
 107         },
 108         'params': {
 109             'skip_download': True,
 110         },
 111         'skip': 'The VOD does not exist',
 112     }, {
 113         'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
 114         'only_matching': True,
 115     }, {
 116         'url': 'https://vod.afreecatv.com/player/96753363',
 117         'info_dict': {
 118             'id': '20230108_9FF5BEE1_244432674_1',
 119             'ext': 'mp4',
 120             'uploader_id': 'rlantnghks',
 121             'uploader': '페이즈으',
 122             'duration': 10840,
 123             'thumbnail': 'http://videoimg.afreecatv.com/php/SnapshotLoad.php?rowKey=20230108_9FF5BEE1_244432674_1_r',
 124             'upload_date': '20230108',
 125             'title': '젠지 페이즈',
 126         },
 127         'params': {
 128             'skip_download': True,
 129         },
 130     }]
 131
 132     @staticmethod
 133     def parse_video_key(key):
 134         video_key = {}
 135         m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
 136         if m:
 137             video_key['upload_date'] = m.group('upload_date')
 138             video_key['part'] = int(m.group('part'))
 139         return video_key
 140
 141     def _perform_login(self, username, password):
 142         login_form = {
 143             'szWork': 'login',
 144             'szType': 'json',
 145             'szUid': username,
 146             'szPassword': password,
 147             'isSaveId': 'false',
 148             'szScriptVar': 'oLoginRet',
 149             'szAction': '',
 150         }
 151
 152         response = self._download_json(
 153             'https://login.afreecatv.com/app/LoginAction.php', None,
 154             'Logging in', data=urlencode_postdata(login_form))
 155
 156         _ERRORS = {
 157             -4: 'Your account has been suspended due to a violation of our terms and policies.',
 158             -5: 'https://member.afreecatv.com/app/user_delete_progress.php',
 159             -6: 'https://login.afreecatv.com/membership/changeMember.php',
 160             -8: "Hello! AfreecaTV 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.",
 161             -9: 'https://member.afreecatv.com/app/pop_login_block.php',
 162             -11: 'https://login.afreecatv.com/afreeca/second_login.php',
 163             -12: 'https://member.afreecatv.com/app/user_security.php',
 164             0: 'The username does not exist or you have entered the wrong password.',
 165             -1: 'The username does not exist or you have entered the wrong password.',
 166             -3: 'You have entered your username/password incorrectly.',
 167             -7: 'You cannot use your Global AfreecaTV account to access Korean AfreecaTV.',
 168             -10: 'Sorry for the inconvenience. \nYour account has been blocked due to an unauthorized access. \nPlease contact our Help Center for assistance.',
 169             -32008: 'You have failed to log in. Please contact our Help Center.',
 170         }
 171
 172         result = int_or_none(response.get('RESULT'))
 173         if result != 1:
 174             error = _ERRORS.get(result, 'You have failed to log in.')
 175             raise ExtractorError(
 176                 'Unable to login: %s said: %s' % (self.IE_NAME, error),
 177                 expected=True)
 178
 179     def _real_extract(self, url):
 180         video_id = self._match_id(url)
 181
 182         partial_view = False
 183         adult_view = False
 184         for _ in range(2):
 185             data = self._download_json(
 186                 'https://api.m.afreecatv.com/station/video/a/view',
 187                 video_id, headers={'Referer': url}, data=urlencode_postdata({
 188                     'nTitleNo': video_id,
 189                     'nApiLevel': 10,
 190                 }))['data']
 191             if traverse_obj(data, ('code', {int})) == -6221:
 192                 raise ExtractorError('The VOD does not exist', expected=True)
 193             query = {
 194                 'nTitleNo': video_id,
 195                 'nStationNo': data['station_no'],
 196                 'nBbsNo': data['bbs_no'],
 197             }
 198             if partial_view:
 199                 query['partialView'] = 'SKIP_ADULT'
 200             if adult_view:
 201                 query['adultView'] = 'ADULT_VIEW'
 202             video_xml = self._download_xml(
 203                 'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php',
 204                 video_id, 'Downloading video info XML%s'
 205                 % (' (skipping adult)' if partial_view else ''),
 206                 video_id, headers={
 207                     'Referer': url,
 208                 }, query=query)
 209
 210             flag = xpath_text(video_xml, './track/flag', 'flag', default=None)
 211             if flag and flag == 'SUCCEED':
 212                 break
 213             if flag == 'PARTIAL_ADULT':
 214                 self.report_warning(
 215                     'In accordance with local laws and regulations, underage users are restricted from watching adult content. '
 216                     'Only content suitable for all ages will be downloaded. '
 217                     'Provide account credentials if you wish to download restricted content.')
 218                 partial_view = True
 219                 continue
 220             elif flag == 'ADULT':
 221                 if not adult_view:
 222                     adult_view = True
 223                     continue
 224                 error = 'Only users older than 19 are able to watch this video. Provide account credentials to download this content.'
 225             else:
 226                 error = flag
 227             raise ExtractorError(
 228                 '%s said: %s' % (self.IE_NAME, error), expected=True)
 229         else:
 230             raise ExtractorError('Unable to download video info')
 231
 232         video_element = video_xml.findall('./track/video')[-1]
 233         if video_element is None or video_element.text is None:
 234             raise ExtractorError(
 235                 'Video %s does not exist' % video_id, expected=True)
 236
 237         video_url = video_element.text.strip()
 238
 239         title = xpath_text(video_xml, './track/title', 'title', fatal=True)
 240
 241         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
 242         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
 243         duration = int_or_none(xpath_text(
 244             video_xml, './track/duration', 'duration'))
 245         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
 246
 247         common_entry = {
 248             'uploader': uploader,
 249             'uploader_id': uploader_id,
 250             'thumbnail': thumbnail,
 251         }
 252
 253         info = common_entry.copy()
 254         info.update({
 255             'id': video_id,
 256             'title': title,
 257             'duration': duration,
 258         })
 259
 260         if not video_url:
 261             entries = []
 262             file_elements = video_element.findall('./file')
 263             one = len(file_elements) == 1
 264             for file_num, file_element in enumerate(file_elements, start=1):
 265                 file_url = url_or_none(file_element.text)
 266                 if not file_url:
 267                     continue
 268                 key = file_element.get('key', '')
 269                 upload_date = unified_strdate(self._search_regex(
 270                     r'^(\d{8})_', key, 'upload date', default=None))
 271                 if upload_date is not None:
 272                     # sometimes the upload date isn't included in the file name
 273                     # instead, another random ID is, which may parse as a valid
 274                     # date but be wildly out of a reasonable range
 275                     parsed_date = date_from_str(upload_date)
 276                     if parsed_date.year < 2000 or parsed_date.year >= 2100:
 277                         upload_date = None
 278                 file_duration = int_or_none(file_element.get('duration'))
 279                 format_id = key if key else '%s_%s' % (video_id, file_num)
 280                 if determine_ext(file_url) == 'm3u8':
 281                     formats = self._extract_m3u8_formats(
 282                         file_url, video_id, 'mp4', entry_protocol='m3u8_native',
 283                         m3u8_id='hls',
 284                         note='Downloading part %d m3u8 information' % file_num)
 285                 else:
 286                     formats = [{
 287                         'url': file_url,
 288                         'format_id': 'http',
 289                     }]
 290                 if not formats and not self.get_param('ignore_no_formats'):
 291                     continue
 292                 file_info = common_entry.copy()
 293                 file_info.update({
 294                     'id': format_id,
 295                     'title': title if one else '%s (part %d)' % (title, file_num),
 296                     'upload_date': upload_date,
 297                     'duration': file_duration,
 298                     'formats': formats,
 299                 })
 300                 entries.append(file_info)
 301             entries_info = info.copy()
 302             entries_info.update({
 303                 '_type': 'multi_video',
 304                 'entries': entries,
 305             })
 306             return entries_info
 307
 308         info = {
 309             'id': video_id,
 310             'title': title,
 311             'uploader': uploader,
 312             'uploader_id': uploader_id,
 313             'duration': duration,
 314             'thumbnail': thumbnail,
 315         }
 316
 317         if determine_ext(video_url) == 'm3u8':
 318             info['formats'] = self._extract_m3u8_formats(
 319                 video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 320                 m3u8_id='hls')
 321         else:
 322             app, playpath = video_url.split('mp4:')
 323             info.update({
 324                 'url': app,
 325                 'ext': 'flv',
 326                 'play_path': 'mp4:' + playpath,
 327                 'rtmp_live': True,  # downloading won't end without this
 328             })
 329
 330         return info
 331
 332
 333 class AfreecaTVLiveIE(AfreecaTVIE):  # XXX: Do not subclass from concrete IE
 334
 335     IE_NAME = 'afreecatv:live'
 336     _VALID_URL = r'https?://play\.afreeca(?:tv)?\.com/(?P<id>[^/]+)(?:/(?P<bno>\d+))?'
 337     _TESTS = [{
 338         'url': 'https://play.afreecatv.com/pyh3646/237852185',
 339         'info_dict': {
 340             'id': '237852185',
 341             'ext': 'mp4',
 342             'title': '【 우루과이 오늘은 무슨일이? 】',
 343             'uploader': '박진우[JINU]',
 344             'uploader_id': 'pyh3646',
 345             'timestamp': 1640661495,
 346             'is_live': True,
 347         },
 348         'skip': 'Livestream has ended',
 349     }, {
 350         'url': 'http://play.afreeca.com/pyh3646/237852185',
 351         'only_matching': True,
 352     }, {
 353         'url': 'http://play.afreeca.com/pyh3646',
 354         'only_matching': True,
 355     }]
 356
 357     _LIVE_API_URL = 'https://live.afreecatv.com/afreeca/player_live_api.php'
 358
 359     _QUALITIES = ('sd', 'hd', 'hd2k', 'original')
 360
 361     def _real_extract(self, url):
 362         broadcaster_id, broadcast_no = self._match_valid_url(url).group('id', 'bno')
 363         password = self.get_param('videopassword')
 364
 365         info = self._download_json(self._LIVE_API_URL, broadcaster_id, fatal=False,
 366                                    data=urlencode_postdata({'bid': broadcaster_id})) or {}
 367         channel_info = info.get('CHANNEL') or {}
 368         broadcaster_id = channel_info.get('BJID') or broadcaster_id
 369         broadcast_no = channel_info.get('BNO') or broadcast_no
 370         password_protected = channel_info.get('BPWD')
 371         if not broadcast_no:
 372             raise ExtractorError(f'Unable to extract broadcast number ({broadcaster_id} may not be live)', expected=True)
 373         if password_protected == 'Y' and password is None:
 374             raise ExtractorError(
 375                 'This livestream is protected by a password, use the --video-password option',
 376                 expected=True)
 377
 378         formats = []
 379         quality_key = qualities(self._QUALITIES)
 380         for quality_str in self._QUALITIES:
 381             params = {
 382                 'bno': broadcast_no,
 383                 'stream_type': 'common',
 384                 'type': 'aid',
 385                 'quality': quality_str,
 386             }
 387             if password is not None:
 388                 params['pwd'] = password
 389             aid_response = self._download_json(
 390                 self._LIVE_API_URL, broadcast_no, fatal=False,
 391                 data=urlencode_postdata(params),
 392                 note=f'Downloading access token for {quality_str} stream',
 393                 errnote=f'Unable to download access token for {quality_str} stream')
 394             aid = traverse_obj(aid_response, ('CHANNEL', 'AID'))
 395             if not aid:
 396                 continue
 397
 398             stream_base_url = channel_info.get('RMD') or 'https://livestream-manager.afreecatv.com'
 399             stream_info = self._download_json(
 400                 f'{stream_base_url}/broad_stream_assign.html', broadcast_no, fatal=False,
 401                 query={
 402                     'return_type': channel_info.get('CDN', 'gcp_cdn'),
 403                     'broad_key': f'{broadcast_no}-common-{quality_str}-hls',
 404                 },
 405                 note=f'Downloading metadata for {quality_str} stream',
 406                 errnote=f'Unable to download metadata for {quality_str} stream') or {}
 407
 408             if stream_info.get('view_url'):
 409                 formats.append({
 410                     'format_id': quality_str,
 411                     'url': update_url_query(stream_info['view_url'], {'aid': aid}),
 412                     'ext': 'mp4',
 413                     'protocol': 'm3u8',
 414                     'quality': quality_key(quality_str),
 415                 })
 416
 417         station_info = self._download_json(
 418             'https://st.afreecatv.com/api/get_station_status.php', broadcast_no,
 419             query={'szBjId': broadcaster_id}, fatal=False,
 420             note='Downloading channel metadata', errnote='Unable to download channel metadata') or {}
 421
 422         return {
 423             'id': broadcast_no,
 424             'title': channel_info.get('TITLE') or station_info.get('station_title'),
 425             'uploader': channel_info.get('BJNICK') or station_info.get('station_name'),
 426             'uploader_id': broadcaster_id,
 427             'timestamp': unified_timestamp(station_info.get('broad_start')),
 428             'formats': formats,
 429             'is_live': True,
 430         }
 431
 432
 433 class AfreecaTVUserIE(InfoExtractor):
 434     IE_NAME = 'afreecatv:user'
 435     _VALID_URL = r'https?://bj\.afreeca(?:tv)?\.com/(?P<id>[^/]+)/vods/?(?P<slug_type>[^/]+)?'
 436     _TESTS = [{
 437         'url': 'https://bj.afreecatv.com/ryuryu24/vods/review',
 438         'info_dict': {
 439             '_type': 'playlist',
 440             'id': 'ryuryu24',
 441             'title': 'ryuryu24 - review',
 442         },
 443         'playlist_count': 218,
 444     }, {
 445         'url': 'https://bj.afreecatv.com/parang1995/vods/highlight',
 446         'info_dict': {
 447             '_type': 'playlist',
 448             'id': 'parang1995',
 449             'title': 'parang1995 - highlight',
 450         },
 451         'playlist_count': 997,
 452     }, {
 453         'url': 'https://bj.afreecatv.com/ryuryu24/vods',
 454         'info_dict': {
 455             '_type': 'playlist',
 456             'id': 'ryuryu24',
 457             'title': 'ryuryu24 - all',
 458         },
 459         'playlist_count': 221,
 460     }, {
 461         'url': 'https://bj.afreecatv.com/ryuryu24/vods/balloonclip',
 462         'info_dict': {
 463             '_type': 'playlist',
 464             'id': 'ryuryu24',
 465             'title': 'ryuryu24 - balloonclip',
 466         },
 467         'playlist_count': 0,
 468     }]
 469     _PER_PAGE = 60
 470
 471     def _fetch_page(self, user_id, user_type, page):
 472         page += 1
 473         info = self._download_json(f'https://bjapi.afreecatv.com/api/{user_id}/vods/{user_type}', user_id,
 474                                    query={'page': page, 'per_page': self._PER_PAGE, 'orderby': 'reg_date'},
 475                                    note=f'Downloading {user_type} video page {page}')
 476         for item in info['data']:
 477             yield self.url_result(
 478                 f'https://vod.afreecatv.com/player/{item["title_no"]}/', AfreecaTVIE, item['title_no'])
 479
 480     def _real_extract(self, url):
 481         user_id, user_type = self._match_valid_url(url).group('id', 'slug_type')
 482         user_type = user_type or 'all'
 483         entries = OnDemandPagedList(functools.partial(self._fetch_page, user_id, user_type), self._PER_PAGE)
 484         return self.playlist_result(entries, user_id, f'{user_id} - {user_type}')