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