_ Git - youtube-dl/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 ..compat import (
  10     compat_urllib_parse_urlencode,
  11     compat_str,
  12 )
  13 from ..utils import (
  14     dict_get,
  15     ExtractorError,
  16     float_or_none,
  17     int_or_none,
  18     remove_start,
  19     try_get,
  20     urlencode_postdata,
  21 )
  22
  23
  24 class VLiveIE(InfoExtractor):
  25     IE_NAME = 'vlive'
  26     _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<id>[0-9]+)'
  27     _NETRC_MACHINE = 'vlive'
  28     _TESTS = [{
  29         'url': 'http://www.vlive.tv/video/1326',
  30         'md5': 'cc7314812855ce56de70a06a27314983',
  31         'info_dict': {
  32             'id': '1326',
  33             'ext': 'mp4',
  34             'title': "[V LIVE] Girl's Day's Broadcast",
  35             'creator': "Girl's Day",
  36             'view_count': int,
  37         },
  38     }, {
  39         'url': 'http://www.vlive.tv/video/16937',
  40         'info_dict': {
  41             'id': '16937',
  42             'ext': 'mp4',
  43             'title': '[V LIVE] 첸백시 걍방',
  44             'creator': 'EXO',
  45             'view_count': int,
  46             'subtitles': 'mincount:12',
  47         },
  48         'params': {
  49             'skip_download': True,
  50         },
  51     }, {
  52         'url': 'https://www.vlive.tv/video/129100',
  53         'md5': 'ca2569453b79d66e5b919e5d308bff6b',
  54         'info_dict': {
  55             'id': '129100',
  56             'ext': 'mp4',
  57             'title': "[V LIVE] [BTS+] Run BTS! 2019 - EP.71 :: Behind the scene",
  58             'creator': "BTS+",
  59             'view_count': int,
  60             'subtitles': 'mincount:10',
  61         },
  62         'skip': 'This video is only available for CH+ subscribers',
  63     }]
  64
  65     @classmethod
  66     def suitable(cls, url):
  67         return False if VLivePlaylistIE.suitable(url) else super(VLiveIE, cls).suitable(url)
  68
  69     def _real_initialize(self):
  70         self._login()
  71
  72     def _login(self):
  73         email, password = self._get_login_info()
  74         if None in (email, password):
  75             return
  76
  77         def is_logged_in():
  78             login_info = self._download_json(
  79                 'https://www.vlive.tv/auth/loginInfo', None,
  80                 note='Downloading login info',
  81                 headers={'Referer': 'https://www.vlive.tv/home'})
  82
  83             return try_get(login_info,
  84                            lambda x: x['message']['login'], bool) or False
  85
  86         if is_logged_in():
  87             return
  88
  89         LOGIN_URL = 'https://www.vlive.tv/auth/email/login'
  90         self._request_webpage(LOGIN_URL, None,
  91                               note='Downloading login cookies')
  92
  93         self._download_webpage(
  94             LOGIN_URL, None, note='Logging in',
  95             data=urlencode_postdata({'email': email, 'pwd': password}),
  96             headers={
  97                 'Referer': LOGIN_URL,
  98                 'Content-Type': 'application/x-www-form-urlencoded'
  99             })
 100
 101         if not is_logged_in():
 102             raise ExtractorError('Unable to log in', expected=True)
 103
 104     def _real_extract(self, url):
 105         video_id = self._match_id(url)
 106
 107         webpage = self._download_webpage(
 108             'https://www.vlive.tv/video/%s' % video_id, video_id)
 109
 110         VIDEO_PARAMS_RE = r'\bvlive\.video\.init\(([^)]+)'
 111         VIDEO_PARAMS_FIELD = 'video params'
 112
 113         params = self._parse_json(self._search_regex(
 114             VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD, default=''), video_id,
 115             transform_source=lambda s: '[' + s + ']', fatal=False)
 116
 117         if not params or len(params) < 7:
 118             params = self._search_regex(
 119                 VIDEO_PARAMS_RE, webpage, VIDEO_PARAMS_FIELD)
 120             params = [p.strip(r'"') for p in re.split(r'\s*,\s*', params)]
 121
 122         status, long_video_id, key = params[2], params[5], params[6]
 123         status = remove_start(status, 'PRODUCT_')
 124
 125         if status in ('LIVE_ON_AIR', 'BIG_EVENT_ON_AIR'):
 126             return self._live(video_id, webpage)
 127         elif status in ('VOD_ON_AIR', 'BIG_EVENT_INTRO'):
 128             return self._replay(video_id, webpage, long_video_id, key)
 129
 130         if status == 'LIVE_END':
 131             raise ExtractorError('Uploading for replay. Please wait...',
 132                                  expected=True)
 133         elif status == 'COMING_SOON':
 134             raise ExtractorError('Coming soon!', expected=True)
 135         elif status == 'CANCELED':
 136             raise ExtractorError('We are sorry, '
 137                                  'but the live broadcast has been canceled.',
 138                                  expected=True)
 139         elif status == 'ONLY_APP':
 140             raise ExtractorError('Unsupported video type', expected=True)
 141         else:
 142             raise ExtractorError('Unknown status %s' % status)
 143
 144     def _get_common_fields(self, webpage):
 145         title = self._og_search_title(webpage)
 146         creator = self._html_search_regex(
 147             r'<div[^>]+class="info_area"[^>]*>\s*(?:<em[^>]*>.*</em\s*>\s*)?<a\s+[^>]*>([^<]+)',
 148             webpage, 'creator', fatal=False)
 149         thumbnail = self._og_search_thumbnail(webpage)
 150         return {
 151             'title': title,
 152             'creator': creator,
 153             'thumbnail': thumbnail,
 154         }
 155
 156     def _live(self, video_id, webpage):
 157         init_page = self._download_init_page(video_id)
 158
 159         live_params = self._search_regex(
 160             r'"liveStreamInfo"\s*:\s*(".*"),',
 161             init_page, 'live stream info')
 162         live_params = self._parse_json(live_params, video_id)
 163         live_params = self._parse_json(live_params, video_id)
 164
 165         formats = []
 166         for vid in live_params.get('resolutions', []):
 167             formats.extend(self._extract_m3u8_formats(
 168                 vid['cdnUrl'], video_id, 'mp4',
 169                 m3u8_id=vid.get('name'),
 170                 fatal=False, live=True))
 171         self._sort_formats(formats)
 172
 173         info = self._get_common_fields(webpage)
 174         info.update({
 175             'title': self._live_title(info['title']),
 176             'id': video_id,
 177             'formats': formats,
 178             'is_live': True,
 179         })
 180         return info
 181
 182     def _replay(self, video_id, webpage, long_video_id, key):
 183         if '' in (long_video_id, key):
 184             init_page = self._download_init_page(video_id)
 185             video_info = self._parse_json(self._search_regex(
 186                 r'(?s)oVideoStatus\s*=\s*({.*})', init_page, 'video info'),
 187                 video_id)
 188             if video_info['status'] == 'NEED_CHANNEL_PLUS':
 189                 self.raise_login_required(
 190                     'This video is only available for CH+ subscribers')
 191             long_video_id, key = video_info['vid'], video_info['inkey']
 192
 193         playinfo = self._download_json(
 194             'http://global.apis.naver.com/rmcnmv/rmcnmv/vod_play_videoInfo.json?%s'
 195             % compat_urllib_parse_urlencode({
 196                 'videoId': long_video_id,
 197                 'key': key,
 198                 'ptc': 'http',
 199                 'doct': 'json',  # document type (xml or json)
 200                 'cpt': 'vtt',  # captions type (vtt or ttml)
 201             }), video_id)
 202
 203         formats = [{
 204             'url': vid['source'],
 205             'format_id': vid.get('encodingOption', {}).get('name'),
 206             'abr': float_or_none(vid.get('bitrate', {}).get('audio')),
 207             'vbr': float_or_none(vid.get('bitrate', {}).get('video')),
 208             'width': int_or_none(vid.get('encodingOption', {}).get('width')),
 209             'height': int_or_none(vid.get('encodingOption', {}).get('height')),
 210             'filesize': int_or_none(vid.get('size')),
 211         } for vid in playinfo.get('videos', {}).get('list', []) if vid.get('source')]
 212         self._sort_formats(formats)
 213
 214         view_count = int_or_none(playinfo.get('meta', {}).get('count'))
 215
 216         subtitles = {}
 217         for caption in playinfo.get('captions', {}).get('list', []):
 218             lang = dict_get(caption, ('locale', 'language', 'country', 'label'))
 219             if lang and caption.get('source'):
 220                 subtitles[lang] = [{
 221                     'ext': 'vtt',
 222                     'url': caption['source']}]
 223
 224         info = self._get_common_fields(webpage)
 225         info.update({
 226             'id': video_id,
 227             'formats': formats,
 228             'view_count': view_count,
 229             'subtitles': subtitles,
 230         })
 231         return info
 232
 233     def _download_init_page(self, video_id):
 234         return self._download_webpage(
 235             'https://www.vlive.tv/video/init/view',
 236             video_id, note='Downloading live webpage',
 237             data=urlencode_postdata({'videoSeq': video_id}),
 238             headers={
 239                 'Referer': 'https://www.vlive.tv/video/%s' % video_id,
 240                 'Content-Type': 'application/x-www-form-urlencoded'
 241             })
 242
 243
 244 class VLiveChannelIE(InfoExtractor):
 245     IE_NAME = 'vlive:channel'
 246     _VALID_URL = r'https?://channels\.vlive\.tv/(?P<id>[0-9A-Z]+)'
 247     _TEST = {
 248         'url': 'http://channels.vlive.tv/FCD4B',
 249         'info_dict': {
 250             'id': 'FCD4B',
 251             'title': 'MAMAMOO',
 252         },
 253         'playlist_mincount': 110
 254     }
 255     _APP_ID = '8c6cc7b45d2568fb668be6e05b6e5a3b'
 256
 257     def _real_extract(self, url):
 258         channel_code = self._match_id(url)
 259
 260         webpage = self._download_webpage(
 261             'http://channels.vlive.tv/%s/video' % channel_code, channel_code)
 262
 263         app_id = None
 264
 265         app_js_url = self._search_regex(
 266             r'<script[^>]+src=(["\'])(?P<url>http.+?/app\.js.*?)\1',
 267             webpage, 'app js', default=None, group='url')
 268
 269         if app_js_url:
 270             app_js = self._download_webpage(
 271                 app_js_url, channel_code, 'Downloading app JS', fatal=False)
 272             if app_js:
 273                 app_id = self._search_regex(
 274                     r'Global\.VFAN_APP_ID\s*=\s*[\'"]([^\'"]+)[\'"]',
 275                     app_js, 'app id', default=None)
 276
 277         app_id = app_id or self._APP_ID
 278
 279         channel_info = self._download_json(
 280             'http://api.vfan.vlive.tv/vproxy/channelplus/decodeChannelCode',
 281             channel_code, note='Downloading decode channel code',
 282             query={
 283                 'app_id': app_id,
 284                 'channelCode': channel_code,
 285                 '_': int(time.time())
 286             })
 287
 288         channel_seq = channel_info['result']['channelSeq']
 289         channel_name = None
 290         entries = []
 291
 292         for page_num in itertools.count(1):
 293             video_list = self._download_json(
 294                 'http://api.vfan.vlive.tv/vproxy/channelplus/getChannelVideoList',
 295                 channel_code, note='Downloading channel list page #%d' % page_num,
 296                 query={
 297                     'app_id': app_id,
 298                     'channelSeq': channel_seq,
 299                     # Large values of maxNumOfRows (~300 or above) may cause
 300                     # empty responses (see [1]), e.g. this happens for [2] that
 301                     # has more than 300 videos.
 302                     # 1. https://github.com/ytdl-org/youtube-dl/issues/13830
 303                     # 2. http://channels.vlive.tv/EDBF.
 304                     'maxNumOfRows': 100,
 305                     '_': int(time.time()),
 306                     'pageNo': page_num
 307                 }
 308             )
 309
 310             if not channel_name:
 311                 channel_name = try_get(
 312                     video_list,
 313                     lambda x: x['result']['channelInfo']['channelName'],
 314                     compat_str)
 315
 316             videos = try_get(
 317                 video_list, lambda x: x['result']['videoList'], list)
 318             if not videos:
 319                 break
 320
 321             for video in videos:
 322                 video_id = video.get('videoSeq')
 323                 if not video_id:
 324                     continue
 325                 video_id = compat_str(video_id)
 326                 entries.append(
 327                     self.url_result(
 328                         'http://www.vlive.tv/video/%s' % video_id,
 329                         ie=VLiveIE.ie_key(), video_id=video_id))
 330
 331         return self.playlist_result(
 332             entries, channel_code, channel_name)
 333
 334
 335 class VLivePlaylistIE(InfoExtractor):
 336     IE_NAME = 'vlive:playlist'
 337     _VALID_URL = r'https?://(?:(?:www|m)\.)?vlive\.tv/video/(?P<video_id>[0-9]+)/playlist/(?P<id>[0-9]+)'
 338     _TEST = {
 339         'url': 'http://www.vlive.tv/video/22867/playlist/22912',
 340         'info_dict': {
 341             'id': '22912',
 342             'title': 'Valentine Day Message from TWICE'
 343         },
 344         'playlist_mincount': 9
 345     }
 346
 347     def _real_extract(self, url):
 348         mobj = re.match(self._VALID_URL, url)
 349         video_id, playlist_id = mobj.group('video_id', 'id')
 350
 351         VIDEO_URL_TEMPLATE = 'http://www.vlive.tv/video/%s'
 352         if self._downloader.params.get('noplaylist'):
 353             self.to_screen(
 354                 'Downloading just video %s because of --no-playlist' % video_id)
 355             return self.url_result(
 356                 VIDEO_URL_TEMPLATE % video_id,
 357                 ie=VLiveIE.ie_key(), video_id=video_id)
 358
 359         self.to_screen(
 360             'Downloading playlist %s - add --no-playlist to just download video'
 361             % playlist_id)
 362
 363         webpage = self._download_webpage(
 364             'http://www.vlive.tv/video/%s/playlist/%s'
 365             % (video_id, playlist_id), playlist_id)
 366
 367         item_ids = self._parse_json(
 368             self._search_regex(
 369                 r'playlistVideoSeqs\s*=\s*(\[[^]]+\])', webpage,
 370                 'playlist video seqs'),
 371             playlist_id)
 372
 373         entries = [
 374             self.url_result(
 375                 VIDEO_URL_TEMPLATE % item_id, ie=VLiveIE.ie_key(),
 376                 video_id=compat_str(item_id))
 377             for item_id in item_ids]
 378
 379         playlist_name = self._html_search_regex(
 380             r'<div[^>]+class="[^"]*multicam_playlist[^>]*>\s*<h3[^>]+>([^<]+)',
 381             webpage, 'playlist title', fatal=False)
 382
 383         return self.playlist_result(entries, playlist_id, playlist_name)