1# coding: utf-8 2from __future__ import unicode_literals 3 4import itertools 5import json 6 7from .naver import NaverBaseIE 8from ..compat import ( 9 compat_HTTPError, 10 compat_str, 11) 12from ..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 25class 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 77class 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 = post.get('playlist') 150 if not playlist or self.get_param('noplaylist'): 151 if playlist: 152 self.to_screen( 153 'Downloading just video %s because of --no-playlist' 154 % video_id) 155 156 video = post['officialVideo'] 157 return self._get_vlive_info(post, video, video_id) 158 else: 159 playlist_name = playlist.get('name') 160 playlist_id = str_or_none(playlist.get('playlistSeq')) 161 playlist_count = str_or_none(playlist.get('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 'thumbnail': video.get('thumb'), 188 } 189 190 video_type = video.get('type') 191 if video_type == 'VOD': 192 inkey = self._call_api('video/v1.0/vod/%s/inkey', video_id)['inkey'] 193 vod_id = video['vodId'] 194 info_dict = merge_dicts( 195 get_common_fields(), 196 self._extract_video_info(video_id, vod_id, inkey)) 197 thumbnail = video.get('thumb') 198 if thumbnail: 199 if not info_dict.get('thumbnails') and info_dict.get('thumbnail'): 200 info_dict['thumbnails'] = [{'url': info_dict.pop('thumbnail')}] 201 info_dict.setdefault('thumbnails', []).append({'url': thumbnail, 'preference': 1}) 202 return info_dict 203 elif video_type == 'LIVE': 204 status = video.get('status') 205 if status == 'ON_AIR': 206 stream_url = self._call_api( 207 'old/v3/live/%s/playInfo', 208 video_id)['result']['adaptiveStreamUrl'] 209 formats = self._extract_m3u8_formats(stream_url, video_id, 'mp4') 210 self._sort_formats(formats) 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 232class 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 self._sort_formats(formats) 289 entry = { 290 'formats': formats, 291 'id': video_id, 292 'thumbnail': upload_info.get('imageUrl'), 293 } 294 elif upload_type == 'V': 295 vod_id = upload_info.get('videoId') 296 if not vod_id: 297 continue 298 inkey = self._call_api(self._FVIDEO_TMPL % 'inKey', video_id)['inKey'] 299 entry = self._extract_video_info(video_id, vod_id, inkey) 300 if entry: 301 entry['title'] = '%s_part%s' % (title, idx) 302 entries.append(entry) 303 return self.playlist_result( 304 entries, post_id, title, strip_or_none(post.get('plainBody'))) 305 306 307class VLiveChannelIE(VLiveBaseIE): 308 IE_NAME = 'vlive:channel' 309 _VALID_URL = r'https?://(?:channels\.vlive\.tv|(?:(?:www|m)\.)?vlive\.tv/channel)/(?P<channel_id>[0-9A-Z]+)(?:/board/(?P<posts_id>\d+))?' 310 _TESTS = [{ 311 'url': 'http://channels.vlive.tv/FCD4B', 312 'info_dict': { 313 'id': 'FCD4B', 314 'title': 'MAMAMOO', 315 }, 316 'playlist_mincount': 110 317 }, { 318 'url': 'https://www.vlive.tv/channel/FCD4B', 319 'only_matching': True, 320 }, { 321 'url': 'https://www.vlive.tv/channel/FCD4B/board/3546', 322 'info_dict': { 323 'id': 'FCD4B-3546', 324 'title': 'MAMAMOO - Star Board', 325 }, 326 'playlist_mincount': 880 327 }] 328 329 def _entries(self, posts_id, board_name): 330 if board_name: 331 posts_path = 'post/v1.0/board-%s/posts' 332 query_add = {'limit': 100, 'sortType': 'LATEST'} 333 else: 334 posts_path = 'post/v1.0/channel-%s/starPosts' 335 query_add = {'limit': 100} 336 337 for page_num in itertools.count(1): 338 video_list = self._call_api( 339 posts_path, posts_id, 'channel{channelName},contentType,postId,title,url', query_add, 340 note=f'Downloading playlist page {page_num}') 341 342 for video in try_get(video_list, lambda x: x['data'], list) or []: 343 video_id = str(video.get('postId')) 344 video_title = str_or_none(video.get('title')) 345 video_url = url_or_none(video.get('url')) 346 if not all((video_id, video_title, video_url)) or video.get('contentType') != 'VIDEO': 347 continue 348 channel_name = try_get(video, lambda x: x['channel']['channelName'], compat_str) 349 yield self.url_result(video_url, VLivePostIE.ie_key(), video_id, video_title, channel=channel_name) 350 351 after = try_get(video_list, lambda x: x['paging']['nextParams']['after'], compat_str) 352 if not after: 353 break 354 query_add['after'] = after 355 356 def _real_extract(self, url): 357 channel_id, posts_id = self._match_valid_url(url).groups() 358 359 board_name = None 360 if posts_id: 361 board = self._call_api( 362 'board/v1.0/board-%s', posts_id, 'title,boardType') 363 board_name = board.get('title') or 'Unknown' 364 if board.get('boardType') not in ('STAR', 'VLIVE_PLUS'): 365 raise ExtractorError(f'Board {board_name!r} is not supported', expected=True) 366 367 entries = LazyList(self._entries(posts_id or channel_id, board_name)) 368 channel_name = entries[0]['channel'] 369 370 return self.playlist_result( 371 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) 374