yt_dlp/extractor/afreecatv.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import functools
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import compat_xpath
   9 from ..utils import (
  10     ExtractorError,
  11     OnDemandPagedList,
  12     date_from_str,
  13     determine_ext,
  14     int_or_none,
  15     qualities,
  16     traverse_obj,
  17     unified_strdate,
  18     unified_timestamp,
  19     update_url_query,
  20     url_or_none,
  21     urlencode_postdata,
  22     xpath_text,
  23 )
  24
  25
  26 class AfreecaTVIE(InfoExtractor):
  27     IE_NAME = 'afreecatv'
  28     IE_DESC = 'afreecatv.com'
  29     _VALID_URL = r'''(?x)
  30                     https?://
  31                         (?:
  32                             (?:(?:live|afbbs|www)\.)?afreeca(?:tv)?\.com(?::\d+)?
  33                             (?:
  34                                 /app/(?:index|read_ucc_bbs)\.cgi|
  35                                 /player/[Pp]layer\.(?:swf|html)
  36                             )\?.*?\bnTitleNo=|
  37                             vod\.afreecatv\.com/(PLAYER/STATION|player)/
  38                         )
  39                         (?P<id>\d+)
  40                     '''
  41     _NETRC_MACHINE = 'afreecatv'
  42     _TESTS = [{
  43         'url': 'http://live.afreecatv.com:8079/app/index.cgi?szType=read_ucc_bbs&szBjId=dailyapril&nStationNo=16711924&nBbsNo=18605867&nTitleNo=36164052&szSkin=',
  44         'md5': 'f72c89fe7ecc14c1b5ce506c4996046e',
  45         'info_dict': {
  46             'id': '36164052',
  47             'ext': 'mp4',
  48             'title': '데일리 에이프릴 요정들의 시상식!',
  49             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  50             'uploader': 'dailyapril',
  51             'uploader_id': 'dailyapril',
  52             'upload_date': '20160503',
  53         },
  54         'skip': 'Video is gone',
  55     }, {
  56         'url': 'http://afbbs.afreecatv.com:8080/app/read_ucc_bbs.cgi?nStationNo=16711924&nTitleNo=36153164&szBjId=dailyapril&nBbsNo=18605867',
  57         'info_dict': {
  58             'id': '36153164',
  59             'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  60             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
  61             'uploader': 'dailyapril',
  62             'uploader_id': 'dailyapril',
  63         },
  64         'playlist_count': 2,
  65         'playlist': [{
  66             'md5': 'd8b7c174568da61d774ef0203159bf97',
  67             'info_dict': {
  68                 'id': '36153164_1',
  69                 'ext': 'mp4',
  70                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  71                 'upload_date': '20160502',
  72             },
  73         }, {
  74             'md5': '58f2ce7f6044e34439ab2d50612ab02b',
  75             'info_dict': {
  76                 'id': '36153164_2',
  77                 'ext': 'mp4',
  78                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
  79                 'upload_date': '20160502',
  80             },
  81         }],
  82         'skip': 'Video is gone',
  83     }, {
  84         'url': 'http://vod.afreecatv.com/PLAYER/STATION/18650793',
  85         'info_dict': {
  86             'id': '18650793',
  87             'ext': 'mp4',
  88             'title': '오늘은 다르다! 쏘님의 우월한 위아래~ 댄스리액션!',
  89             'thumbnail': r're:^https?://.*\.jpg$',
  90             'uploader': '윈아디',
  91             'uploader_id': 'badkids',
  92             'duration': 107,
  93         },
  94         'params': {
  95             'skip_download': True,
  96         },
  97     }, {
  98         'url': 'http://vod.afreecatv.com/PLAYER/STATION/10481652',
  99         'info_dict': {
 100             'id': '10481652',
 101             'title': "BJ유트루와 함께하는 '팅커벨 메이크업!'",
 102             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
 103             'uploader': 'dailyapril',
 104             'uploader_id': 'dailyapril',
 105             'duration': 6492,
 106         },
 107         'playlist_count': 2,
 108         'playlist': [{
 109             'md5': 'd8b7c174568da61d774ef0203159bf97',
 110             'info_dict': {
 111                 'id': '20160502_c4c62b9d_174361386_1',
 112                 'ext': 'mp4',
 113                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 1)",
 114                 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
 115                 'uploader': 'dailyapril',
 116                 'uploader_id': 'dailyapril',
 117                 'upload_date': '20160502',
 118                 'duration': 3601,
 119             },
 120         }, {
 121             'md5': '58f2ce7f6044e34439ab2d50612ab02b',
 122             'info_dict': {
 123                 'id': '20160502_39e739bb_174361386_2',
 124                 'ext': 'mp4',
 125                 'title': "BJ유트루와 함께하는 '팅커벨 메이크업!' (part 2)",
 126                 'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
 127                 'uploader': 'dailyapril',
 128                 'uploader_id': 'dailyapril',
 129                 'upload_date': '20160502',
 130                 'duration': 2891,
 131             },
 132         }],
 133         'params': {
 134             'skip_download': True,
 135         },
 136     }, {
 137         # non standard key
 138         'url': 'http://vod.afreecatv.com/PLAYER/STATION/20515605',
 139         'info_dict': {
 140             'id': '20170411_BE689A0E_190960999_1_2_h',
 141             'ext': 'mp4',
 142             'title': '혼자사는여자집',
 143             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
 144             'uploader': '♥이슬이',
 145             'uploader_id': 'dasl8121',
 146             'upload_date': '20170411',
 147             'duration': 213,
 148         },
 149         'params': {
 150             'skip_download': True,
 151         },
 152     }, {
 153         # PARTIAL_ADULT
 154         'url': 'http://vod.afreecatv.com/PLAYER/STATION/32028439',
 155         'info_dict': {
 156             'id': '20180327_27901457_202289533_1',
 157             'ext': 'mp4',
 158             'title': '[생]빨개요♥ (part 1)',
 159             'thumbnail': 're:^https?://(?:video|st)img.afreecatv.com/.*$',
 160             'uploader': '[SA]서아',
 161             'uploader_id': 'bjdyrksu',
 162             'upload_date': '20180327',
 163             'duration': 3601,
 164         },
 165         'params': {
 166             'skip_download': True,
 167         },
 168         'expected_warnings': ['adult content'],
 169     }, {
 170         'url': 'http://www.afreecatv.com/player/Player.swf?szType=szBjId=djleegoon&nStationNo=11273158&nBbsNo=13161095&nTitleNo=36327652',
 171         'only_matching': True,
 172     }, {
 173         'url': 'http://vod.afreecatv.com/PLAYER/STATION/15055030',
 174         'only_matching': True,
 175     }, {
 176         'url': 'http://vod.afreecatv.com/player/15055030',
 177         'only_matching': True,
 178     }]
 179
 180     @staticmethod
 181     def parse_video_key(key):
 182         video_key = {}
 183         m = re.match(r'^(?P<upload_date>\d{8})_\w+_(?P<part>\d+)$', key)
 184         if m:
 185             video_key['upload_date'] = m.group('upload_date')
 186             video_key['part'] = int(m.group('part'))
 187         return video_key
 188
 189     def _perform_login(self, username, password):
 190         login_form = {
 191             'szWork': 'login',
 192             'szType': 'json',
 193             'szUid': username,
 194             'szPassword': password,
 195             'isSaveId': 'false',
 196             'szScriptVar': 'oLoginRet',
 197             'szAction': '',
 198         }
 199
 200         response = self._download_json(
 201             'https://login.afreecatv.com/app/LoginAction.php', None,
 202             'Logging in', data=urlencode_postdata(login_form))
 203
 204         _ERRORS = {
 205             -4: 'Your account has been suspended due to a violation of our terms and policies.',
 206             -5: 'https://member.afreecatv.com/app/user_delete_progress.php',
 207             -6: 'https://login.afreecatv.com/membership/changeMember.php',
 208             -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.",
 209             -9: 'https://member.afreecatv.com/app/pop_login_block.php',
 210             -11: 'https://login.afreecatv.com/afreeca/second_login.php',
 211             -12: 'https://member.afreecatv.com/app/user_security.php',
 212             0: 'The username does not exist or you have entered the wrong password.',
 213             -1: 'The username does not exist or you have entered the wrong password.',
 214             -3: 'You have entered your username/password incorrectly.',
 215             -7: 'You cannot use your Global AfreecaTV account to access Korean AfreecaTV.',
 216             -10: 'Sorry for the inconvenience. \nYour account has been blocked due to an unauthorized access. \nPlease contact our Help Center for assistance.',
 217             -32008: 'You have failed to log in. Please contact our Help Center.',
 218         }
 219
 220         result = int_or_none(response.get('RESULT'))
 221         if result != 1:
 222             error = _ERRORS.get(result, 'You have failed to log in.')
 223             raise ExtractorError(
 224                 'Unable to login: %s said: %s' % (self.IE_NAME, error),
 225                 expected=True)
 226
 227     def _real_extract(self, url):
 228         video_id = self._match_id(url)
 229
 230         webpage = self._download_webpage(url, video_id)
 231
 232         if re.search(r'alert\(["\']This video has been deleted', webpage):
 233             raise ExtractorError(
 234                 'Video %s has been deleted' % video_id, expected=True)
 235
 236         station_id = self._search_regex(
 237             r'nStationNo\s*=\s*(\d+)', webpage, 'station')
 238         bbs_id = self._search_regex(
 239             r'nBbsNo\s*=\s*(\d+)', webpage, 'bbs')
 240         video_id = self._search_regex(
 241             r'nTitleNo\s*=\s*(\d+)', webpage, 'title', default=video_id)
 242
 243         partial_view = False
 244         adult_view = False
 245         for _ in range(2):
 246             query = {
 247                 'nTitleNo': video_id,
 248                 'nStationNo': station_id,
 249                 'nBbsNo': bbs_id,
 250             }
 251             if partial_view:
 252                 query['partialView'] = 'SKIP_ADULT'
 253             if adult_view:
 254                 query['adultView'] = 'ADULT_VIEW'
 255             video_xml = self._download_xml(
 256                 'http://afbbs.afreecatv.com:8080/api/video/get_video_info.php',
 257                 video_id, 'Downloading video info XML%s'
 258                 % (' (skipping adult)' if partial_view else ''),
 259                 video_id, headers={
 260                     'Referer': url,
 261                 }, query=query)
 262
 263             flag = xpath_text(video_xml, './track/flag', 'flag', default=None)
 264             if flag and flag == 'SUCCEED':
 265                 break
 266             if flag == 'PARTIAL_ADULT':
 267                 self.report_warning(
 268                     'In accordance with local laws and regulations, underage users are restricted from watching adult content. '
 269                     'Only content suitable for all ages will be downloaded. '
 270                     'Provide account credentials if you wish to download restricted content.')
 271                 partial_view = True
 272                 continue
 273             elif flag == 'ADULT':
 274                 if not adult_view:
 275                     adult_view = True
 276                     continue
 277                 error = 'Only users older than 19 are able to watch this video. Provide account credentials to download this content.'
 278             else:
 279                 error = flag
 280             raise ExtractorError(
 281                 '%s said: %s' % (self.IE_NAME, error), expected=True)
 282         else:
 283             raise ExtractorError('Unable to download video info')
 284
 285         video_element = video_xml.findall(compat_xpath('./track/video'))[-1]
 286         if video_element is None or video_element.text is None:
 287             raise ExtractorError(
 288                 'Video %s does not exist' % video_id, expected=True)
 289
 290         video_url = video_element.text.strip()
 291
 292         title = xpath_text(video_xml, './track/title', 'title', fatal=True)
 293
 294         uploader = xpath_text(video_xml, './track/nickname', 'uploader')
 295         uploader_id = xpath_text(video_xml, './track/bj_id', 'uploader id')
 296         duration = int_or_none(xpath_text(
 297             video_xml, './track/duration', 'duration'))
 298         thumbnail = xpath_text(video_xml, './track/titleImage', 'thumbnail')
 299
 300         common_entry = {
 301             'uploader': uploader,
 302             'uploader_id': uploader_id,
 303             'thumbnail': thumbnail,
 304         }
 305
 306         info = common_entry.copy()
 307         info.update({
 308             'id': video_id,
 309             'title': title,
 310             'duration': duration,
 311         })
 312
 313         if not video_url:
 314             entries = []
 315             file_elements = video_element.findall(compat_xpath('./file'))
 316             one = len(file_elements) == 1
 317             for file_num, file_element in enumerate(file_elements, start=1):
 318                 file_url = url_or_none(file_element.text)
 319                 if not file_url:
 320                     continue
 321                 key = file_element.get('key', '')
 322                 upload_date = unified_strdate(self._search_regex(
 323                     r'^(\d{8})_', key, 'upload date', default=None))
 324                 if upload_date is not None:
 325                     # sometimes the upload date isn't included in the file name
 326                     # instead, another random ID is, which may parse as a valid
 327                     # date but be wildly out of a reasonable range
 328                     parsed_date = date_from_str(upload_date)
 329                     if parsed_date.year < 2000 or parsed_date.year >= 2100:
 330                         upload_date = None
 331                 file_duration = int_or_none(file_element.get('duration'))
 332                 format_id = key if key else '%s_%s' % (video_id, file_num)
 333                 if determine_ext(file_url) == 'm3u8':
 334                     formats = self._extract_m3u8_formats(
 335                         file_url, video_id, 'mp4', entry_protocol='m3u8_native',
 336                         m3u8_id='hls',
 337                         note='Downloading part %d m3u8 information' % file_num)
 338                 else:
 339                     formats = [{
 340                         'url': file_url,
 341                         'format_id': 'http',
 342                     }]
 343                 if not formats and not self.get_param('ignore_no_formats'):
 344                     continue
 345                 self._sort_formats(formats)
 346                 file_info = common_entry.copy()
 347                 file_info.update({
 348                     'id': format_id,
 349                     'title': title if one else '%s (part %d)' % (title, file_num),
 350                     'upload_date': upload_date,
 351                     'duration': file_duration,
 352                     'formats': formats,
 353                 })
 354                 entries.append(file_info)
 355             entries_info = info.copy()
 356             entries_info.update({
 357                 '_type': 'multi_video',
 358                 'entries': entries,
 359             })
 360             return entries_info
 361
 362         info = {
 363             'id': video_id,
 364             'title': title,
 365             'uploader': uploader,
 366             'uploader_id': uploader_id,
 367             'duration': duration,
 368             'thumbnail': thumbnail,
 369         }
 370
 371         if determine_ext(video_url) == 'm3u8':
 372             info['formats'] = self._extract_m3u8_formats(
 373                 video_url, video_id, 'mp4', entry_protocol='m3u8_native',
 374                 m3u8_id='hls')
 375         else:
 376             app, playpath = video_url.split('mp4:')
 377             info.update({
 378                 'url': app,
 379                 'ext': 'flv',
 380                 'play_path': 'mp4:' + playpath,
 381                 'rtmp_live': True,  # downloading won't end without this
 382             })
 383
 384         return info
 385
 386
 387 class AfreecaTVLiveIE(AfreecaTVIE):
 388
 389     IE_NAME = 'afreecatv:live'
 390     _VALID_URL = r'https?://play\.afreeca(?:tv)?\.com/(?P<id>[^/]+)(?:/(?P<bno>\d+))?'
 391     _TESTS = [{
 392         'url': 'https://play.afreecatv.com/pyh3646/237852185',
 393         'info_dict': {
 394             'id': '237852185',
 395             'ext': 'mp4',
 396             'title': '【 우루과이 오늘은 무슨일이? 】',
 397             'uploader': '박진우[JINU]',
 398             'uploader_id': 'pyh3646',
 399             'timestamp': 1640661495,
 400             'is_live': True,
 401         },
 402         'skip': 'Livestream has ended',
 403     }, {
 404         'url': 'http://play.afreeca.com/pyh3646/237852185',
 405         'only_matching': True,
 406     }, {
 407         'url': 'http://play.afreeca.com/pyh3646',
 408         'only_matching': True,
 409     }]
 410
 411     _LIVE_API_URL = 'https://live.afreecatv.com/afreeca/player_live_api.php'
 412
 413     _QUALITIES = ('sd', 'hd', 'hd2k', 'original')
 414
 415     def _real_extract(self, url):
 416         broadcaster_id, broadcast_no = self._match_valid_url(url).group('id', 'bno')
 417         password = self.get_param('videopassword')
 418
 419         info = self._download_json(self._LIVE_API_URL, broadcaster_id, fatal=False,
 420                                    data=urlencode_postdata({'bid': broadcaster_id})) or {}
 421         channel_info = info.get('CHANNEL') or {}
 422         broadcaster_id = channel_info.get('BJID') or broadcaster_id
 423         broadcast_no = channel_info.get('BNO') or broadcast_no
 424         password_protected = channel_info.get('BPWD')
 425         if not broadcast_no:
 426             raise ExtractorError(f'Unable to extract broadcast number ({broadcaster_id} may not be live)', expected=True)
 427         if password_protected == 'Y' and password is None:
 428             raise ExtractorError(
 429                 'This livestream is protected by a password, use the --video-password option',
 430                 expected=True)
 431
 432         formats = []
 433         quality_key = qualities(self._QUALITIES)
 434         for quality_str in self._QUALITIES:
 435             params = {
 436                 'bno': broadcast_no,
 437                 'stream_type': 'common',
 438                 'type': 'aid',
 439                 'quality': quality_str,
 440             }
 441             if password is not None:
 442                 params['pwd'] = password
 443             aid_response = self._download_json(
 444                 self._LIVE_API_URL, broadcast_no, fatal=False,
 445                 data=urlencode_postdata(params),
 446                 note=f'Downloading access token for {quality_str} stream',
 447                 errnote=f'Unable to download access token for {quality_str} stream')
 448             aid = traverse_obj(aid_response, ('CHANNEL', 'AID'))
 449             if not aid:
 450                 continue
 451
 452             stream_base_url = channel_info.get('RMD') or 'https://livestream-manager.afreecatv.com'
 453             stream_info = self._download_json(
 454                 f'{stream_base_url}/broad_stream_assign.html', broadcast_no, fatal=False,
 455                 query={
 456                     'return_type': channel_info.get('CDN', 'gcp_cdn'),
 457                     'broad_key': f'{broadcast_no}-common-{quality_str}-hls',
 458                 },
 459                 note=f'Downloading metadata for {quality_str} stream',
 460                 errnote=f'Unable to download metadata for {quality_str} stream') or {}
 461
 462             if stream_info.get('view_url'):
 463                 formats.append({
 464                     'format_id': quality_str,
 465                     'url': update_url_query(stream_info['view_url'], {'aid': aid}),
 466                     'ext': 'mp4',
 467                     'protocol': 'm3u8',
 468                     'quality': quality_key(quality_str),
 469                 })
 470
 471         self._sort_formats(formats)
 472
 473         station_info = self._download_json(
 474             'https://st.afreecatv.com/api/get_station_status.php', broadcast_no,
 475             query={'szBjId': broadcaster_id}, fatal=False,
 476             note='Downloading channel metadata', errnote='Unable to download channel metadata') or {}
 477
 478         return {
 479             'id': broadcast_no,
 480             'title': channel_info.get('TITLE') or station_info.get('station_title'),
 481             'uploader': channel_info.get('BJNICK') or station_info.get('station_name'),
 482             'uploader_id': broadcaster_id,
 483             'timestamp': unified_timestamp(station_info.get('broad_start')),
 484             'formats': formats,
 485             'is_live': True,
 486         }
 487
 488
 489 class AfreecaTVUserIE(InfoExtractor):
 490     IE_NAME = 'afreecatv:user'
 491     _VALID_URL = r'https?://bj\.afreeca(?:tv)?\.com/(?P<id>[^/]+)/vods/?(?P<slug_type>[^/]+)?'
 492     _TESTS = [{
 493         'url': 'https://bj.afreecatv.com/ryuryu24/vods/review',
 494         'info_dict': {
 495             '_type': 'playlist',
 496             'id': 'ryuryu24',
 497             'title': 'ryuryu24 - review',
 498         },
 499         'playlist_count': 218,
 500     }, {
 501         'url': 'https://bj.afreecatv.com/parang1995/vods/highlight',
 502         'info_dict': {
 503             '_type': 'playlist',
 504             'id': 'parang1995',
 505             'title': 'parang1995 - highlight',
 506         },
 507         'playlist_count': 997,
 508     }, {
 509         'url': 'https://bj.afreecatv.com/ryuryu24/vods',
 510         'info_dict': {
 511             '_type': 'playlist',
 512             'id': 'ryuryu24',
 513             'title': 'ryuryu24 - all',
 514         },
 515         'playlist_count': 221,
 516     }, {
 517         'url': 'https://bj.afreecatv.com/ryuryu24/vods/balloonclip',
 518         'info_dict': {
 519             '_type': 'playlist',
 520             'id': 'ryuryu24',
 521             'title': 'ryuryu24 - balloonclip',
 522         },
 523         'playlist_count': 0,
 524     }]
 525     _PER_PAGE = 60
 526
 527     def _fetch_page(self, user_id, user_type, page):
 528         page += 1
 529         info = self._download_json(f'https://bjapi.afreecatv.com/api/{user_id}/vods/{user_type}', user_id,
 530                                    query={'page': page, 'per_page': self._PER_PAGE, 'orderby': 'reg_date'},
 531                                    note=f'Downloading {user_type} video page {page}')
 532         for item in info['data']:
 533             yield self.url_result(
 534                 f'https://vod.afreecatv.com/player/{item["title_no"]}/', AfreecaTVIE, item['title_no'])
 535
 536     def _real_extract(self, url):
 537         user_id, user_type = self._match_valid_url(url).group('id', 'slug_type')
 538         user_type = user_type or 'all'
 539         entries = OnDemandPagedList(functools.partial(self._fetch_page, user_id, user_type), self._PER_PAGE)
 540         return self.playlist_result(entries, user_id, f'{user_id} - {user_type}')