yt_dlp/extractor/vlive.py

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