Raphaël G. Git Repositories - youtubedl/blob - youtube_dl/extractor/vlive.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import re
   5 import time
   6 import itertools
   7
   8 from .common import InfoExtractor
   9 from .naver import NaverBaseIE
  10 from ..compat import compat_str
  11 from ..utils import (
  12     ExtractorError,
  13     merge_dicts,
  14     remove_start,
  15     try_get,
  16     urlencode_postdata,
  17 )
  18
  19
  20 class VLiveIE(NaverBaseIE):
  21     IE_NAME = 'vlive'
  22     _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
  23     _NETRC_MACHINE = 'vlive'
  24     _TESTS = [{
  25         'url': 'http://www.vlive.tv/video/1326',
  26         'md5': 'cc7314812855ce56de70a06a27314983',
  27         'info_dict': {
  28             'id': '1326',
  29             'ext': 'mp4',
  30             'title': "[V LIVE] Girl's Day's Broadcast",
  31             'creator': "Girl's Day",
  32             'view_count': int,
  33             'uploader_id': 'muploader_a',
  34         },
  35     }, {
  36         'url': 'http://www.vlive.tv/video/16937',
  37         'info_dict': {
  38             'id': '16937',
  39             'ext': 'mp4',
  40             'title': '[V LIVE] 첸백시 걍방',
  41             'creator': 'EXO',
  42             'view_count': int,
  43             'subtitles': 'mincount:12',
  44             'uploader_id': 'muploader_j',
  45         },
  46         'params': {
  47             'skip_download': True,
  48         },
  49     }, {
  50         'url': 'https://www.vlive.tv/video/129100',
  51         'md5': 'ca2569453b79d66e5b919e5d308bff6b',
  52         'info_dict': {
  53             'id': '129100',
  54             'ext': 'mp4',
  55             'title': '[V LIVE] [BTS+] Run BTS! 2019 - EP.71 :: Behind the scene',
  56             'creator': 'BTS+',
  57             'view_count': int,
  58             'subtitles': 'mincount:10',
  59         },
  60         'skip': 'This video is only available for CH+ subscribers',
  61     }]
  62
  63     @classmethod
  64     def suitable(cls, url):
  65         return False if VLivePlaylistIE.suitable(url) else super(VLiveIE, cls).suitable(url)
  66
  67     def _real_initialize(self):
  68         self._login()
  69
  70     def _login(self):
  71         email, password = self._get_login_info()
  72         if None in (email, password):
  73             return
  74
  75         def is_logged_in():
  76             login_info = self._download_json(
  77                 'https://www.vlive.tv/auth/loginInfo', None,
  78                 note='Downloading login info',
  79                 headers={'Referer': 'https://www.vlive.tv/home'})
  80             return try_get(
  81                 login_info, lambda x: x['message']['login'], bool) or False
  82
  83         LOGIN_URL = 'https://www.vlive.tv/auth/email/login'
  84         self._request_webpage(
  85             LOGIN_URL, None, note='Downloading login cookies')
  86
  87         self._download_webpage(
  88             LOGIN_URL, None, note='Logging in',
  89             data=urlencode_postdata({'email': email, 'pwd': password}),
  90             headers={
  91                 'Referer': LOGIN_URL,
  92                 'Content-Type': 'application/x-www-form-urlencoded'
  93             })
  94
  95         if not is_logged_in():
  96             raise ExtractorError('Unable to log in', expected=True)
  97
  98     def _real_extract(self, url):
  99         video_id = self._match_id(url)
 100
 101         webpage = self._download_webpage(
 102             'https://www.vlive.tv/video/%s' % video_id, video_id)
 103
 104         VIDEO_PARAMS_RE = r'\bvlive\.video\.init\(([^)]+)'
 105         VIDEO_PARAMS_FIELD = 'video params'
 106
 107         params = self._parse_json(self._search_regex(
 108             VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD, default=''), video_id,
 109             transform_source=lambda s: '[' + s + ']', fatal=False)
 110
 111         if not params or len(params) < 7:
 112             params = self._search_regex(
 113                 VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD)
 114             params = [p.strip(r'"') for p in re.split(r'\s*,\s*', params)]
 115
 116         status, long_video_id, key = params[2], params[5], params[6]
 117         status = remove_start(status, 'PRODUCT_')
 118
 119         if status in ('LIVE_ON_AIR', 'BIG_EVENT_ON_AIR'):
 120             return self._live(video_id, webpage)
 121         elif status in ('VOD_ON_AIR', 'BIG_EVENT_INTRO'):
 122             return self._replay(video_id, webpage, long_video_id, key)
 123
 124         if status == 'LIVE_END':
 125             raise ExtractorError('Uploading for replay. Please wait...',
 126                                  expected=True)
 127         elif status == 'COMING_SOON':
 128             raise ExtractorError('Coming soon!', expected=True)
 129         elif status == 'CANCELED':
 130             raise ExtractorError('We are sorry, '
 131                                  'but the live broadcast has been canceled.',
 132                                  expected=True)
 133         elif status == 'ONLY_APP':
 134             raise ExtractorError('Unsupported video type', expected=True)
 135         else:
 136             raise ExtractorError('Unknown status %s' % status)
 137
 138     def _get_common_fields(self, webpage):
 139         title = self._og_search_title(webpage)
 140         creator = self._html_search_regex(
 141             r'<div[^>]+class="info_area"[^>]*>\s*(?:<em[^>]*>.*?</em\s*>\s*)?<a\s+[^>]*>([^<]+)',
 142             webpage, 'creator', fatal=False)
 143         thumbnail = self._og_search_thumbnail(webpage)
 144         return {
 145             'title': title,
 146             'creator': creator,
 147             'thumbnail': thumbnail,
 148         }
 149
 150     def _live(self, video_id, webpage):
 151         init_page = self._download_init_page(video_id)
 152
 153         live_params = self._search_regex(
 154             r'"liveStreamInfo"\s*:\s*(".*"),',
 155             init_page, 'live stream info')
 156         live_params = self._parse_json(live_params, video_id)
 157         live_params = self._parse_json(live_params, video_id)
 158
 159         formats = []
 160         for vid in live_params.get('resolutions', []):
 161             formats.extend(self._extract_m3u8_formats(
 162                 vid['cdnUrl'], video_id, 'mp4',
 163                 m3u8_id=vid.get('name'),
 164                 fatal=False, live=True))
 165         self._sort_formats(formats)
 166
 167         info = self._get_common_fields(webpage)
 168         info.update({
 169             'title': self._live_title(info['title']),
 170             'id': video_id,
 171             'formats': formats,
 172             'is_live': True,
 173         })
 174         return info
 175
 176     def _replay(self, video_id, webpage, long_video_id, key):
 177         if '' in (long_video_id, key):
 178             init_page = self._download_init_page(video_id)
 179             video_info = self._parse_json(self._search_regex(
 180                 (r'(?s)oVideoStatus\s*=\s*({.+?})\s*</script',
 181                  r'(?s)oVideoStatus\s*=\s*({.+})'), init_page, 'video info'),
 182                 video_id)
 183             if video_info.get('status') == 'NEED_CHANNEL_PLUS':
 184                 self.raise_login_required(
 185                     'This video is only available for CH+ subscribers')
 186             long_video_id, key = video_info['vid'], video_info['inkey']
 187
 188         return merge_dicts(
 189             self._get_common_fields(webpage),
 190             self._extract_video_info(video_id, long_video_id, key))
 191
 192     def _download_init_page(self, video_id):
 193         return self._download_webpage(
 194             'https://www.vlive.tv/video/init/view',
 195             video_id, note='Downloading live webpage',
 196             data=urlencode_postdata({'videoSeq': video_id}),
 197             headers={
 198                 'Referer': 'https://www.vlive.tv/video/%s' % video_id,
 199                 'Content-Type': 'application/x-www-form-urlencoded'
 200             })
 201
 202
 203 class VLiveChannelIE(InfoExtractor):
 204     IE_NAME = 'vlive:channel'
 205     _VALID_URL = r'https?://channels\.vlive\.tv/(?P<id>[0-9A-Z]+)'
 206     _TEST = {
 207         'url': 'http://channels.vlive.tv/FCD4B',
 208         'info_dict': {
 209             'id': 'FCD4B',
 210             'title': 'MAMAMOO',
 211         },
 212         'playlist_mincount': 110
 213     }
 214     _APP_ID = '8c6cc7b45d2568fb668be6e05b6e5a3b'
 215
 216     def _real_extract(self, url):
 217         channel_code = self._match_id(url)
 218
 219         webpage = self._download_webpage(
 220             'http://channels.vlive.tv/%s/video' % channel_code, channel_code)
 221
 222         app_id = None
 223
 224         app_js_url = self._search_regex(
 225             r'<script[^>]+src=(["\'])(?P<url>http.+?/app\.js.*?)\1',
 226             webpage, 'app js', default=None, group='url')
 227
 228         if app_js_url:
 229             app_js = self._download_webpage(
 230                 app_js_url, channel_code, 'Downloading app JS', fatal=False)
 231             if app_js:
 232                 app_id = self._search_regex(
 233                     r'Global\.VFAN_APP_ID\s*=\s*[\'"]([^\'"]+)[\'"]',
 234                     app_js, 'app id', default=None)
 235
 236         app_id = app_id or self._APP_ID
 237
 238         channel_info = self._download_json(
 239             'http://api.vfan.vlive.tv/vproxy/channelplus/decodeChannelCode',
 240             channel_code, note='Downloading decode channel code',
 241             query={
 242                 'app_id': app_id,
 243                 'channelCode': channel_code,
 244                 '_': int(time.time())
 245             })
 246
 247         channel_seq = channel_info['result']['channelSeq']
 248         channel_name = None
 249         entries = []
 250
 251         for page_num in itertools.count(1):
 252             video_list = self._download_json(
 253                 'http://api.vfan.vlive.tv/vproxy/channelplus/getChannelVideoList',
 254                 channel_code, note='Downloading channel list page #%d' % page_num,
 255                 query={
 256                     'app_id': app_id,
 257                     'channelSeq': channel_seq,
 258                     # Large values of maxNumOfRows (~300 or above) may cause
 259                     # empty responses (see [1]), e.g. this happens for [2] that
 260                     # has more than 300 videos.
 261                     # 1. https://github.com/ytdl-org/youtube-dl/issues/13830
 262                     # 2. http://channels.vlive.tv/EDBF.
 263                     'maxNumOfRows': 100,
 264                     '_': int(time.time()),
 265                     'pageNo': page_num
 266                 }
 267             )
 268
 269             if not channel_name:
 270                 channel_name = try_get(
 271                     video_list,
 272                     lambda x: x['result']['channelInfo']['channelName'],
 273                     compat_str)
 274
 275             videos = try_get(
 276                 video_list, lambda x: x['result']['videoList'], list)
 277             if not videos:
 278                 break
 279
 280             for video in videos:
 281                 video_id = video.get('videoSeq')
 282                 if not video_id:
 283                     continue
 284                 video_id = compat_str(video_id)
 285                 entries.append(
 286                     self.url_result(
 287                         'http://www.vlive.tv/video/%s' % video_id,
 288                         ie=VLiveIE.ie_key(), video_id=video_id))
 289
 290         return self.playlist_result(
 291             entries, channel_code, channel_name)
 292
 293
 294 class VLivePlaylistIE(InfoExtractor):
 295     IE_NAME = 'vlive:playlist'
 296     _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<video_id>[0-9]+)/playlist/(?P<id>[0-9]+)'
 297     _VIDEO_URL_TEMPLATE = 'http://www.vlive.tv/video/%s'
 298     _TESTS = [{
 299         # regular working playlist
 300         'url': 'https://www.vlive.tv/video/117956/playlist/117963',
 301         'info_dict': {
 302             'id': '117963',
 303             'title': '아이돌룸(IDOL ROOM) 41회 - (여자)아이들'
 304         },
 305         'playlist_mincount': 10
 306     }, {
 307         # playlist with no playlistVideoSeqs
 308         'url': 'http://www.vlive.tv/video/22867/playlist/22912',
 309         'info_dict': {
 310             'id': '22867',
 311             'ext': 'mp4',
 312             'title': '[V LIVE] Valentine Day Message from MINA',
 313             'creator': 'TWICE',
 314             'view_count': int
 315         },
 316         'params': {
 317             'skip_download': True,
 318         }
 319     }]
 320
 321     def _build_video_result(self, video_id, message):
 322         self.to_screen(message)
 323         return self.url_result(
 324             self._VIDEO_URL_TEMPLATE % video_id,
 325             ie=VLiveIE.ie_key(), video_id=video_id)
 326
 327     def _real_extract(self, url):
 328         mobj = re.match(self._VALID_URL, url)
 329         video_id, playlist_id = mobj.group('video_id', 'id')
 330
 331         if self._downloader.params.get('noplaylist'):
 332             return self._build_video_result(
 333                 video_id,
 334                 'Downloading just video %s because of --no-playlist'
 335                 % video_id)
 336
 337         self.to_screen(
 338             'Downloading playlist %s - add --no-playlist to just download video'
 339             % playlist_id)
 340
 341         webpage = self._download_webpage(
 342             'http://www.vlive.tv/video/%s/playlist/%s'
 343             % (video_id, playlist_id), playlist_id)
 344
 345         raw_item_ids = self._search_regex(
 346             r'playlistVideoSeqs\s*=\s*(\[[^]]+\])', webpage,
 347             'playlist video seqs', default=None, fatal=False)
 348
 349         if not raw_item_ids:
 350             return self._build_video_result(
 351                 video_id,
 352                 'Downloading just video %s because no playlist was found'
 353                 % video_id)
 354
 355         item_ids = self._parse_json(raw_item_ids, playlist_id)
 356
 357         entries = [
 358             self.url_result(
 359                 self._VIDEO_URL_TEMPLATE % item_id, ie=VLiveIE.ie_key(),
 360                 video_id=compat_str(item_id))
 361             for item_id in item_ids]
 362
 363         playlist_name = self._html_search_regex(
 364             r'<div[^>]+class="[^"]*multicam_playlist[^>]*>\s*<h3[^>]+>([^<]+)',
 365             webpage, 'playlist title', fatal=False)
 366
 367         return self.playlist_result(entries, playlist_id, playlist_name)