youtube_dl/extractor/bilibili.py

   1 # coding: utf-8
   2 from __future__ import unicode_literals
   3
   4 import hashlib
   5 import re
   6
   7 from .common import InfoExtractor
   8 from ..compat import (
   9     compat_parse_qs,
  10     compat_urlparse,
  11 )
  12 from ..utils import (
  13     ExtractorError,
  14     int_or_none,
  15     float_or_none,
  16     parse_iso8601,
  17     smuggle_url,
  18     str_or_none,
  19     strip_jsonp,
  20     unified_timestamp,
  21     unsmuggle_url,
  22     urlencode_postdata,
  23 )
  24
  25
  26 class BiliBiliIE(InfoExtractor):
  27     _VALID_URL = r'''(?x)
  28                     https?://
  29                         (?:(?:www|bangumi)\.)?
  30                         bilibili\.(?:tv|com)/
  31                         (?:
  32                             (?:
  33                                 video/[aA][vV]|
  34                                 anime/(?P<anime_id>\d+)/play\#
  35                             )(?P<id_bv>\d+)|
  36                             video/[bB][vV](?P<id>[^/?#&]+)
  37                         )
  38                     '''
  39
  40     _TESTS = [{
  41         'url': 'http://www.bilibili.tv/video/av1074402/',
  42         'md5': '5f7d29e1a2872f3df0cf76b1f87d3788',
  43         'info_dict': {
  44             'id': '1074402',
  45             'ext': 'flv',
  46             'title': '【金坷垃】金泡沫',
  47             'description': 'md5:ce18c2a2d2193f0df2917d270f2e5923',
  48             'duration': 308.067,
  49             'timestamp': 1398012678,
  50             'upload_date': '20140420',
  51             'thumbnail': r're:^https?://.+\.jpg',
  52             'uploader': '菊子桑',
  53             'uploader_id': '156160',
  54         },
  55     }, {
  56         # Tested in BiliBiliBangumiIE
  57         'url': 'http://bangumi.bilibili.com/anime/1869/play#40062',
  58         'only_matching': True,
  59     }, {
  60         'url': 'http://bangumi.bilibili.com/anime/5802/play#100643',
  61         'md5': '3f721ad1e75030cc06faf73587cfec57',
  62         'info_dict': {
  63             'id': '100643',
  64             'ext': 'mp4',
  65             'title': 'CHAOS;CHILD',
  66             'description': '如果你是神明，并且能够让妄想成为现实。那你会进行怎么样的妄想？是淫靡的世界？独裁社会？毁灭性的制裁？还是……2015年，涩谷。从6年前发生的大灾害“涩谷地震”之后复兴了的这个街区里新设立的私立高中...',
  67         },
  68         'skip': 'Geo-restricted to China',
  69     }, {
  70         # Title with double quotes
  71         'url': 'http://www.bilibili.com/video/av8903802/',
  72         'info_dict': {
  73             'id': '8903802',
  74             'title': '阿滴英文｜英文歌分享#6 "Closer',
  75             'description': '滴妹今天唱Closer給你聽! 有史以来，被推最多次也是最久的歌曲，其实歌词跟我原本想像差蛮多的，不过还是好听！ 微博@阿滴英文',
  76         },
  77         'playlist': [{
  78             'info_dict': {
  79                 'id': '8903802_part1',
  80                 'ext': 'flv',
  81                 'title': '阿滴英文｜英文歌分享#6 "Closer',
  82                 'description': 'md5:3b1b9e25b78da4ef87e9b548b88ee76a',
  83                 'uploader': '阿滴英文',
  84                 'uploader_id': '65880958',
  85                 'timestamp': 1488382634,
  86                 'upload_date': '20170301',
  87             },
  88             'params': {
  89                 'skip_download': True,  # Test metadata only
  90             },
  91         }, {
  92             'info_dict': {
  93                 'id': '8903802_part2',
  94                 'ext': 'flv',
  95                 'title': '阿滴英文｜英文歌分享#6 "Closer',
  96                 'description': 'md5:3b1b9e25b78da4ef87e9b548b88ee76a',
  97                 'uploader': '阿滴英文',
  98                 'uploader_id': '65880958',
  99                 'timestamp': 1488382634,
 100                 'upload_date': '20170301',
 101             },
 102             'params': {
 103                 'skip_download': True,  # Test metadata only
 104             },
 105         }]
 106     }, {
 107         # new BV video id format
 108         'url': 'https://www.bilibili.com/video/BV1JE411F741',
 109         'only_matching': True,
 110     }]
 111
 112     _APP_KEY = 'iVGUTjsxvpLeuDCf'
 113     _BILIBILI_KEY = 'aHRmhWMLkdeMuILqORnYZocwMBpMEOdt'
 114
 115     def _report_error(self, result):
 116         if 'message' in result:
 117             raise ExtractorError('%s said: %s' % (self.IE_NAME, result['message']), expected=True)
 118         elif 'code' in result:
 119             raise ExtractorError('%s returns error %d' % (self.IE_NAME, result['code']), expected=True)
 120         else:
 121             raise ExtractorError('Can\'t extract Bangumi episode ID')
 122
 123     def _real_extract(self, url):
 124         url, smuggled_data = unsmuggle_url(url, {})
 125
 126         mobj = re.match(self._VALID_URL, url)
 127         video_id = mobj.group('id') or mobj.group('id_bv')
 128         anime_id = mobj.group('anime_id')
 129         webpage = self._download_webpage(url, video_id)
 130
 131         if 'anime/' not in url:
 132             cid = self._search_regex(
 133                 r'\bcid(?:["\']:|=)(\d+)', webpage, 'cid',
 134                 default=None
 135             ) or compat_parse_qs(self._search_regex(
 136                 [r'EmbedPlayer\([^)]+,\s*"([^"]+)"\)',
 137                  r'EmbedPlayer\([^)]+,\s*\\"([^"]+)\\"\)',
 138                  r'<iframe[^>]+src="https://secure\.bilibili\.com/secure,([^"]+)"'],
 139                 webpage, 'player parameters'))['cid'][0]
 140         else:
 141             if 'no_bangumi_tip' not in smuggled_data:
 142                 self.to_screen('Downloading episode %s. To download all videos in anime %s, re-run youtube-dl with %s' % (
 143                     video_id, anime_id, compat_urlparse.urljoin(url, '//bangumi.bilibili.com/anime/%s' % anime_id)))
 144             headers = {
 145                 'Content-Type': 'application/x-www-form-urlencoded; charset=UTF-8',
 146                 'Referer': url
 147             }
 148             headers.update(self.geo_verification_headers())
 149
 150             js = self._download_json(
 151                 'http://bangumi.bilibili.com/web_api/get_source', video_id,
 152                 data=urlencode_postdata({'episode_id': video_id}),
 153                 headers=headers)
 154             if 'result' not in js:
 155                 self._report_error(js)
 156             cid = js['result']['cid']
 157
 158         headers = {
 159             'Accept': 'application/json',
 160             'Referer': url
 161         }
 162         headers.update(self.geo_verification_headers())
 163
 164         entries = []
 165
 166         RENDITIONS = ('qn=80&quality=80&type=', 'quality=2&type=mp4')
 167         for num, rendition in enumerate(RENDITIONS, start=1):
 168             payload = 'appkey=%s&cid=%s&otype=json&%s' % (self._APP_KEY, cid, rendition)
 169             sign = hashlib.md5((payload + self._BILIBILI_KEY).encode('utf-8')).hexdigest()
 170
 171             video_info = self._download_json(
 172                 'http://interface.bilibili.com/v2/playurl?%s&sign=%s' % (payload, sign),
 173                 video_id, note='Downloading video info page',
 174                 headers=headers, fatal=num == len(RENDITIONS))
 175
 176             if not video_info:
 177                 continue
 178
 179             if 'durl' not in video_info:
 180                 if num < len(RENDITIONS):
 181                     continue
 182                 self._report_error(video_info)
 183
 184             for idx, durl in enumerate(video_info['durl']):
 185                 formats = [{
 186                     'url': durl['url'],
 187                     'filesize': int_or_none(durl['size']),
 188                 }]
 189                 for backup_url in durl.get('backup_url', []):
 190                     formats.append({
 191                         'url': backup_url,
 192                         # backup URLs have lower priorities
 193                         'preference': -2 if 'hd.mp4' in backup_url else -3,
 194                     })
 195
 196                 for a_format in formats:
 197                     a_format.setdefault('http_headers', {}).update({
 198                         'Referer': url,
 199                     })
 200
 201                 self._sort_formats(formats)
 202
 203                 entries.append({
 204                     'id': '%s_part%s' % (video_id, idx),
 205                     'duration': float_or_none(durl.get('length'), 1000),
 206                     'formats': formats,
 207                 })
 208             break
 209
 210         title = self._html_search_regex(
 211             ('<h1[^>]+\btitle=(["\'])(?P<title>(?:(?!\1).)+)\1',
 212              '(?s)<h1[^>]*>(?P<title>.+?)</h1>'), webpage, 'title',
 213             group='title')
 214         description = self._html_search_meta('description', webpage)
 215         timestamp = unified_timestamp(self._html_search_regex(
 216             r'<time[^>]+datetime="([^"]+)"', webpage, 'upload time',
 217             default=None) or self._html_search_meta(
 218             'uploadDate', webpage, 'timestamp', default=None))
 219         thumbnail = self._html_search_meta(['og:image', 'thumbnailUrl'], webpage)
 220
 221         # TODO 'view_count' requires deobfuscating Javascript
 222         info = {
 223             'id': video_id,
 224             'title': title,
 225             'description': description,
 226             'timestamp': timestamp,
 227             'thumbnail': thumbnail,
 228             'duration': float_or_none(video_info.get('timelength'), scale=1000),
 229         }
 230
 231         uploader_mobj = re.search(
 232             r'<a[^>]+href="(?:https?:)?//space\.bilibili\.com/(?P<id>\d+)"[^>]*>(?P<name>[^<]+)',
 233             webpage)
 234         if uploader_mobj:
 235             info.update({
 236                 'uploader': uploader_mobj.group('name').strip(),
 237                 'uploader_id': uploader_mobj.group('id'),
 238             })
 239         if not info.get('uploader'):
 240             info['uploader'] = self._html_search_meta(
 241                 'author', webpage, 'uploader', default=None)
 242
 243         for entry in entries:
 244             entry.update(info)
 245
 246         if len(entries) == 1:
 247             return entries[0]
 248         else:
 249             for idx, entry in enumerate(entries):
 250                 entry['id'] = '%s_part%d' % (video_id, (idx + 1))
 251
 252             return {
 253                 '_type': 'multi_video',
 254                 'id': video_id,
 255                 'title': title,
 256                 'description': description,
 257                 'entries': entries,
 258             }
 259
 260
 261 class BiliBiliBangumiIE(InfoExtractor):
 262     _VALID_URL = r'https?://bangumi\.bilibili\.com/anime/(?P<id>\d+)'
 263
 264     IE_NAME = 'bangumi.bilibili.com'
 265     IE_DESC = 'BiliBili番剧'
 266
 267     _TESTS = [{
 268         'url': 'http://bangumi.bilibili.com/anime/1869',
 269         'info_dict': {
 270             'id': '1869',
 271             'title': '混沌武士',
 272             'description': 'md5:6a9622b911565794c11f25f81d6a97d2',
 273         },
 274         'playlist_count': 26,
 275     }, {
 276         'url': 'http://bangumi.bilibili.com/anime/1869',
 277         'info_dict': {
 278             'id': '1869',
 279             'title': '混沌武士',
 280             'description': 'md5:6a9622b911565794c11f25f81d6a97d2',
 281         },
 282         'playlist': [{
 283             'md5': '91da8621454dd58316851c27c68b0c13',
 284             'info_dict': {
 285                 'id': '40062',
 286                 'ext': 'mp4',
 287                 'title': '混沌武士',
 288                 'description': '故事发生在日本的江户时代。风是一个小酒馆的打工女。一日，酒馆里来了一群恶霸，虽然他们的举动令风十分不满，但是毕竟风只是一届女流，无法对他们采取什么行动，只能在心里嘟哝。这时，酒家里又进来了个“不良份子...',
 289                 'timestamp': 1414538739,
 290                 'upload_date': '20141028',
 291                 'episode': '疾风怒涛 Tempestuous Temperaments',
 292                 'episode_number': 1,
 293             },
 294         }],
 295         'params': {
 296             'playlist_items': '1',
 297         },
 298     }]
 299
 300     @classmethod
 301     def suitable(cls, url):
 302         return False if BiliBiliIE.suitable(url) else super(BiliBiliBangumiIE, cls).suitable(url)
 303
 304     def _real_extract(self, url):
 305         bangumi_id = self._match_id(url)
 306
 307         # Sometimes this API returns a JSONP response
 308         season_info = self._download_json(
 309             'http://bangumi.bilibili.com/jsonp/seasoninfo/%s.ver' % bangumi_id,
 310             bangumi_id, transform_source=strip_jsonp)['result']
 311
 312         entries = [{
 313             '_type': 'url_transparent',
 314             'url': smuggle_url(episode['webplay_url'], {'no_bangumi_tip': 1}),
 315             'ie_key': BiliBiliIE.ie_key(),
 316             'timestamp': parse_iso8601(episode.get('update_time'), delimiter=' '),
 317             'episode': episode.get('index_title'),
 318             'episode_number': int_or_none(episode.get('index')),
 319         } for episode in season_info['episodes']]
 320
 321         entries = sorted(entries, key=lambda entry: entry.get('episode_number'))
 322
 323         return self.playlist_result(
 324             entries, bangumi_id,
 325             season_info.get('bangumi_title'), season_info.get('evaluate'))
 326
 327
 328 class BilibiliAudioBaseIE(InfoExtractor):
 329     def _call_api(self, path, sid, query=None):
 330         if not query:
 331             query = {'sid': sid}
 332         return self._download_json(
 333             'https://www.bilibili.com/audio/music-service-c/web/' + path,
 334             sid, query=query)['data']
 335
 336
 337 class BilibiliAudioIE(BilibiliAudioBaseIE):
 338     _VALID_URL = r'https?://(?:www\.)?bilibili\.com/audio/au(?P<id>\d+)'
 339     _TEST = {
 340         'url': 'https://www.bilibili.com/audio/au1003142',
 341         'md5': 'fec4987014ec94ef9e666d4d158ad03b',
 342         'info_dict': {
 343             'id': '1003142',
 344             'ext': 'm4a',
 345             'title': '【tsukimi】YELLOW / 神山羊',
 346             'artist': 'tsukimi',
 347             'comment_count': int,
 348             'description': 'YELLOW的mp3版！',
 349             'duration': 183,
 350             'subtitles': {
 351                 'origin': [{
 352                     'ext': 'lrc',
 353                 }],
 354             },
 355             'thumbnail': r're:^https?://.+\.jpg',
 356             'timestamp': 1564836614,
 357             'upload_date': '20190803',
 358             'uploader': 'tsukimi-つきみぐー',
 359             'view_count': int,
 360         },
 361     }
 362
 363     def _real_extract(self, url):
 364         au_id = self._match_id(url)
 365
 366         play_data = self._call_api('url', au_id)
 367         formats = [{
 368             'url': play_data['cdns'][0],
 369             'filesize': int_or_none(play_data.get('size')),
 370         }]
 371
 372         song = self._call_api('song/info', au_id)
 373         title = song['title']
 374         statistic = song.get('statistic') or {}
 375
 376         subtitles = None
 377         lyric = song.get('lyric')
 378         if lyric:
 379             subtitles = {
 380                 'origin': [{
 381                     'url': lyric,
 382                 }]
 383             }
 384
 385         return {
 386             'id': au_id,
 387             'title': title,
 388             'formats': formats,
 389             'artist': song.get('author'),
 390             'comment_count': int_or_none(statistic.get('comment')),
 391             'description': song.get('intro'),
 392             'duration': int_or_none(song.get('duration')),
 393             'subtitles': subtitles,
 394             'thumbnail': song.get('cover'),
 395             'timestamp': int_or_none(song.get('passtime')),
 396             'uploader': song.get('uname'),
 397             'view_count': int_or_none(statistic.get('play')),
 398         }
 399
 400
 401 class BilibiliAudioAlbumIE(BilibiliAudioBaseIE):
 402     _VALID_URL = r'https?://(?:www\.)?bilibili\.com/audio/am(?P<id>\d+)'
 403     _TEST = {
 404         'url': 'https://www.bilibili.com/audio/am10624',
 405         'info_dict': {
 406             'id': '10624',
 407             'title': '每日新曲推荐（每日11:00更新）',
 408             'description': '每天11:00更新，为你推送最新音乐',
 409         },
 410         'playlist_count': 19,
 411     }
 412
 413     def _real_extract(self, url):
 414         am_id = self._match_id(url)
 415
 416         songs = self._call_api(
 417             'song/of-menu', am_id, {'sid': am_id, 'pn': 1, 'ps': 100})['data']
 418
 419         entries = []
 420         for song in songs:
 421             sid = str_or_none(song.get('id'))
 422             if not sid:
 423                 continue
 424             entries.append(self.url_result(
 425                 'https://www.bilibili.com/audio/au' + sid,
 426                 BilibiliAudioIE.ie_key(), sid))
 427
 428         if entries:
 429             album_data = self._call_api('menu/info', am_id) or {}
 430             album_title = album_data.get('title')
 431             if album_title:
 432                 for entry in entries:
 433                     entry['album'] = album_title
 434                 return self.playlist_result(
 435                     entries, am_id, album_title, album_data.get('intro'))
 436
 437         return self.playlist_result(entries, am_id)
 438
 439
 440 class BiliBiliPlayerIE(InfoExtractor):
 441     _VALID_URL = r'https?://player\.bilibili\.com/player\.html\?.*?\baid=(?P<id>\d+)'
 442     _TEST = {
 443         'url': 'http://player.bilibili.com/player.html?aid=92494333&cid=157926707&page=1',
 444         'only_matching': True,
 445     }
 446
 447     def _real_extract(self, url):
 448         video_id = self._match_id(url)
 449         return self.url_result(
 450             'http://www.bilibili.tv/video/av%s/' % video_id,
 451             ie=BiliBiliIE.ie_key(), video_id=video_id)