yt_dlp/extractor/vlive.py

   1 import itertools
   2 import json
   3
   4 from .naver import NaverBaseIE
   5 from ..compat import (
   6     compat_HTTPError,
   7     compat_str,
   8 )
   9 from ..utils import (
  10     ExtractorError,
  11     int_or_none,
  12     LazyList,
  13     merge_dicts,
  14     str_or_none,
  15     strip_or_none,
  16     traverse_obj,
  17     try_get,
  18     urlencode_postdata,
  19     url_or_none,
  20 )
  21
  22
  23 class VLiveBaseIE(NaverBaseIE):
  24     _NETRC_MACHINE = 'vlive'
  25     _logged_in = False
  26
  27     def _perform_login(self, username, password):
  28         if self._logged_in:
  29             return
  30         LOGIN_URL = 'https://www.vlive.tv/auth/email/login'
  31         self._request_webpage(
  32             LOGIN_URL, None, note='Downloading login cookies')
  33
  34         self._download_webpage(
  35             LOGIN_URL, None, note='Logging in',
  36             data=urlencode_postdata({'email': username, 'pwd': password}),
  37             headers={
  38                 'Referer': LOGIN_URL,
  39                 'Content-Type': 'application/x-www-form-urlencoded'
  40             })
  41
  42         login_info = self._download_json(
  43             'https://www.vlive.tv/auth/loginInfo', None,
  44             note='Checking login status',
  45             headers={'Referer': 'https://www.vlive.tv/home'})
  46
  47         if not try_get(login_info, lambda x: x['message']['login'], bool):
  48             raise ExtractorError('Unable to log in', expected=True)
  49         VLiveBaseIE._logged_in = True
  50
  51     def _call_api(self, path_template, video_id, fields=None, query_add={}, note=None):
  52         if note is None:
  53             note = 'Downloading %s JSON metadata' % path_template.split('/')[-1].split('-')[0]
  54         query = {'appId': '8c6cc7b45d2568fb668be6e05b6e5a3b', 'gcc': 'KR', 'platformType': 'PC'}
  55         if fields:
  56             query['fields'] = fields
  57         if query_add:
  58             query.update(query_add)
  59         try:
  60             return self._download_json(
  61                 'https://www.vlive.tv/globalv-web/vam-web/' + path_template % video_id, video_id,
  62                 note, headers={'Referer': 'https://www.vlive.tv/'}, query=query)
  63         except ExtractorError as e:
  64             if isinstance(e.cause, compat_HTTPError) and e.cause.code == 403:
  65                 self.raise_login_required(json.loads(e.cause.read().decode('utf-8'))['message'])
  66             raise
  67
  68
  69 class VLiveIE(VLiveBaseIE):
  70     IE_NAME = 'vlive'
  71     _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/(?:video|embed)/(?P<id>[0-9]+)'
  72     _TESTS = [{
  73         'url': 'http://www.vlive.tv/video/1326',
  74         'md5': 'cc7314812855ce56de70a06a27314983',
  75         'info_dict': {
  76             'id': '1326',
  77             'ext': 'mp4',
  78             'title': "Girl's Day's Broadcast",
  79             'creator': "Girl's Day",
  80             'view_count': int,
  81             'uploader_id': 'muploader_a',
  82             'upload_date': '20150817',
  83             'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
  84             'timestamp': 1439816449,
  85             'like_count': int,
  86             'channel': 'Girl\'s Day',
  87             'channel_id': 'FDF27',
  88             'comment_count': int,
  89             'release_timestamp': 1439818140,
  90             'release_date': '20150817',
  91             'duration': 1014,
  92         },
  93         'params': {
  94             'skip_download': True,
  95         },
  96     }, {
  97         'url': 'http://www.vlive.tv/video/16937',
  98         'info_dict': {
  99             'id': '16937',
 100             'ext': 'mp4',
 101             'title': '첸백시 걍방',
 102             'creator': 'EXO',
 103             'view_count': int,
 104             'subtitles': 'mincount:12',
 105             'uploader_id': 'muploader_j',
 106             'upload_date': '20161112',
 107             'thumbnail': r're:^https?://.*\.(?:jpg|png)$',
 108             'timestamp': 1478923074,
 109             'like_count': int,
 110             'channel': 'EXO',
 111             'channel_id': 'F94BD',
 112             'comment_count': int,
 113             'release_timestamp': 1478924280,
 114             'release_date': '20161112',
 115             'duration': 906,
 116         },
 117         'params': {
 118             'skip_download': True,
 119         },
 120     }, {
 121         'url': 'https://www.vlive.tv/video/129100',
 122         'md5': 'ca2569453b79d66e5b919e5d308bff6b',
 123         'info_dict': {
 124             'id': '129100',
 125             'ext': 'mp4',
 126             'title': '[V LIVE] [BTS+] Run BTS! 2019 - EP.71 :: Behind the scene',
 127             'creator': 'BTS+',
 128             'view_count': int,
 129             'subtitles': 'mincount:10',
 130         },
 131         'skip': 'This video is only available for CH+ subscribers',
 132     }, {
 133         'url': 'https://www.vlive.tv/embed/1326',
 134         'only_matching': True,
 135     }, {
 136         # works only with gcc=KR
 137         'url': 'https://www.vlive.tv/video/225019',
 138         'only_matching': True,
 139     }, {
 140         'url': 'https://www.vlive.tv/video/223906',
 141         'info_dict': {
 142             'id': '58',
 143             'title': 'RUN BTS!'
 144         },
 145         'playlist_mincount': 120
 146     }]
 147
 148     def _real_extract(self, url):
 149         video_id = self._match_id(url)
 150
 151         post = self._call_api(
 152             'post/v1.0/officialVideoPost-%s', video_id,
 153             'author{nickname},channel{channelCode,channelName},officialVideo{commentCount,exposeStatus,likeCount,playCount,playTime,status,title,type,vodId},playlist{playlistSeq,totalCount,name}')
 154
 155         playlist_id = str_or_none(try_get(post, lambda x: x['playlist']['playlistSeq']))
 156         if not self._yes_playlist(playlist_id, video_id):
 157             video = post['officialVideo']
 158             return self._get_vlive_info(post, video, video_id)
 159
 160         playlist_name = str_or_none(try_get(post, lambda x: x['playlist']['name']))
 161         playlist_count = str_or_none(try_get(post, lambda x: x['playlist']['totalCount']))
 162
 163         playlist = self._call_api(
 164             'playlist/v1.0/playlist-%s/posts', playlist_id, 'data', {'limit': playlist_count})
 165
 166         entries = []
 167         for video_data in playlist['data']:
 168             video = video_data.get('officialVideo')
 169             video_id = str_or_none(video.get('videoSeq'))
 170             entries.append(self._get_vlive_info(video_data, video, video_id))
 171
 172         return self.playlist_result(entries, playlist_id, playlist_name)
 173
 174     def _get_vlive_info(self, post, video, video_id):
 175         def get_common_fields():
 176             channel = post.get('channel') or {}
 177             return {
 178                 'title': video.get('title'),
 179                 'creator': post.get('author', {}).get('nickname'),
 180                 'channel': channel.get('channelName'),
 181                 'channel_id': channel.get('channelCode'),
 182                 'duration': int_or_none(video.get('playTime')),
 183                 'view_count': int_or_none(video.get('playCount')),
 184                 'like_count': int_or_none(video.get('likeCount')),
 185                 'comment_count': int_or_none(video.get('commentCount')),
 186                 'timestamp': int_or_none(video.get('createdAt'), scale=1000),
 187                 'release_timestamp': int_or_none(traverse_obj(video, 'onAirStartAt', 'willStartAt'), scale=1000),
 188                 'thumbnail': video.get('thumb'),
 189             }
 190
 191         video_type = video.get('type')
 192         if video_type == 'VOD':
 193             inkey = self._call_api('video/v1.0/vod/%s/inkey', video_id)['inkey']
 194             vod_id = video['vodId']
 195             info_dict = merge_dicts(
 196                 get_common_fields(),
 197                 self._extract_video_info(video_id, vod_id, inkey))
 198             thumbnail = video.get('thumb')
 199             if thumbnail:
 200                 if not info_dict.get('thumbnails') and info_dict.get('thumbnail'):
 201                     info_dict['thumbnails'] = [{'url': info_dict.pop('thumbnail')}]
 202                 info_dict.setdefault('thumbnails', []).append({'url': thumbnail, 'preference': 1})
 203             return info_dict
 204         elif video_type == 'LIVE':
 205             status = video.get('status')
 206             if status == 'ON_AIR':
 207                 stream_url = self._call_api(
 208                     'old/v3/live/%s/playInfo',
 209                     video_id)['result']['adaptiveStreamUrl']
 210                 formats = self._extract_m3u8_formats(stream_url, video_id, 'mp4')
 211                 info = get_common_fields()
 212                 info.update({
 213                     'title': video['title'],
 214                     'id': video_id,
 215                     'formats': formats,
 216                     'is_live': True,
 217                 })
 218                 return info
 219             elif status == 'ENDED':
 220                 raise ExtractorError(
 221                     'Uploading for replay. Please wait...', expected=True)
 222             elif status == 'RESERVED':
 223                 raise ExtractorError('Coming soon!', expected=True)
 224             elif video.get('exposeStatus') == 'CANCEL':
 225                 raise ExtractorError(
 226                     'We are sorry, but the live broadcast has been canceled.',
 227                     expected=True)
 228             else:
 229                 raise ExtractorError('Unknown status ' + status)
 230
 231
 232 class VLivePostIE(VLiveBaseIE):
 233     IE_NAME = 'vlive:post'
 234     _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/post/(?P<id>\d-\d+)'
 235     _TESTS = [{
 236         # uploadType = SOS
 237         'url': 'https://www.vlive.tv/post/1-20088044',
 238         'info_dict': {
 239             'id': '1-20088044',
 240             'title': 'Hola estrellitas la tierra les dice hola (si era así no?) Ha...',
 241             'description': 'md5:fab8a1e50e6e51608907f46c7fa4b407',
 242         },
 243         'playlist_count': 3,
 244     }, {
 245         # uploadType = V
 246         'url': 'https://www.vlive.tv/post/1-20087926',
 247         'info_dict': {
 248             'id': '1-20087926',
 249             'title': 'James Corden: And so, the baby becamos the Papa💜😭💪😭',
 250         },
 251         'playlist_count': 1,
 252     }]
 253     _FVIDEO_TMPL = 'fvideo/v1.0/fvideo-%%s/%s'
 254
 255     def _real_extract(self, url):
 256         post_id = self._match_id(url)
 257
 258         post = self._call_api(
 259             'post/v1.0/post-%s', post_id,
 260             'attachments{video},officialVideo{videoSeq},plainBody,title')
 261
 262         video_seq = str_or_none(try_get(
 263             post, lambda x: x['officialVideo']['videoSeq']))
 264         if video_seq:
 265             return self.url_result(
 266                 'http://www.vlive.tv/video/' + video_seq,
 267                 VLiveIE.ie_key(), video_seq)
 268
 269         title = post['title']
 270         entries = []
 271         for idx, video in enumerate(post['attachments']['video'].values()):
 272             video_id = video.get('videoId')
 273             if not video_id:
 274                 continue
 275             upload_type = video.get('uploadType')
 276             upload_info = video.get('uploadInfo') or {}
 277             entry = None
 278             if upload_type == 'SOS':
 279                 download = self._call_api(
 280                     self._FVIDEO_TMPL % 'sosPlayInfo', video_id)['videoUrl']['download']
 281                 formats = []
 282                 for f_id, f_url in download.items():
 283                     formats.append({
 284                         'format_id': f_id,
 285                         'url': f_url,
 286                         'height': int_or_none(f_id[:-1]),
 287                     })
 288                 entry = {
 289                     'formats': formats,
 290                     'id': video_id,
 291                     'thumbnail': upload_info.get('imageUrl'),
 292                 }
 293             elif upload_type == 'V':
 294                 vod_id = upload_info.get('videoId')
 295                 if not vod_id:
 296                     continue
 297                 inkey = self._call_api(self._FVIDEO_TMPL % 'inKey', video_id)['inKey']
 298                 entry = self._extract_video_info(video_id, vod_id, inkey)
 299             if entry:
 300                 entry['title'] = '%s_part%s' % (title, idx)
 301                 entries.append(entry)
 302         return self.playlist_result(
 303             entries, post_id, title, strip_or_none(post.get('plainBody')))
 304
 305
 306 class VLiveChannelIE(VLiveBaseIE):
 307     IE_NAME = 'vlive:channel'
 308     _VALID_URL = r'https?://(?:channels\.vlive\.tv|(?:(?:www|m)\.)?vlive\.tv/channel)/(?P<channel_id>[0-9A-Z]+)(?:/board/(?P<posts_id>\d+))?'
 309     _TESTS = [{
 310         'url': 'http://channels.vlive.tv/FCD4B',
 311         'info_dict': {
 312             'id': 'FCD4B',
 313             'title': 'MAMAMOO',
 314         },
 315         'playlist_mincount': 110
 316     }, {
 317         'url': 'https://www.vlive.tv/channel/FCD4B',
 318         'only_matching': True,
 319     }, {
 320         'url': 'https://www.vlive.tv/channel/FCD4B/board/3546',
 321         'info_dict': {
 322             'id': 'FCD4B-3546',
 323             'title': 'MAMAMOO - Star Board',
 324         },
 325         'playlist_mincount': 880
 326     }]
 327
 328     def _entries(self, posts_id, board_name):
 329         if board_name:
 330             posts_path = 'post/v1.0/board-%s/posts'
 331             query_add = {'limit': 100, 'sortType': 'LATEST'}
 332         else:
 333             posts_path = 'post/v1.0/channel-%s/starPosts'
 334             query_add = {'limit': 100}
 335
 336         for page_num in itertools.count(1):
 337             video_list = self._call_api(
 338                 posts_path, posts_id, 'channel{channelName},contentType,postId,title,url', query_add,
 339                 note=f'Downloading playlist page {page_num}')
 340
 341             for video in try_get(video_list, lambda x: x['data'], list) or []:
 342                 video_id = str(video.get('postId'))
 343                 video_title = str_or_none(video.get('title'))
 344                 video_url = url_or_none(video.get('url'))
 345                 if not all((video_id, video_title, video_url)) or video.get('contentType') != 'VIDEO':
 346                     continue
 347                 channel_name = try_get(video, lambda x: x['channel']['channelName'], compat_str)
 348                 yield self.url_result(video_url, VLivePostIE.ie_key(), video_id, video_title, channel=channel_name)
 349
 350             after = try_get(video_list, lambda x: x['paging']['nextParams']['after'], compat_str)
 351             if not after:
 352                 break
 353             query_add['after'] = after
 354
 355     def _real_extract(self, url):
 356         channel_id, posts_id = self._match_valid_url(url).groups()
 357
 358         board_name = None
 359         if posts_id:
 360             board = self._call_api(
 361                 'board/v1.0/board-%s', posts_id, 'title,boardType')
 362             board_name = board.get('title') or 'Unknown'
 363             if board.get('boardType') not in ('STAR', 'VLIVE_PLUS'):
 364                 raise ExtractorError(f'Board {board_name!r} is not supported', expected=True)
 365
 366         entries = LazyList(self._entries(posts_id or channel_id, board_name))
 367         channel_name = entries[0]['channel']
 368
 369         return self.playlist_result(
 370             entries,
 371             f'{channel_id}-{posts_id}' if posts_id else channel_id,
 372             f'{channel_name} - {board_name}' if channel_name and board_name else channel_name)