Merge a9eedee3d9 into 48c5663c5f

youtube_dl/extractor/bilibili.py

@@ -20,8 +20,11 @@ from ..utils import (
     unified_timestamp,
     unsmuggle_url,
     urlencode_postdata,
+    try_get,
 )
 
+md5 = lambda s: hashlib.md5(s.encode('utf-8')).hexdigest()
+
 
 class BiliBiliIE(InfoExtractor):
     _VALID_URL = r'''(?x)
@@ -257,6 +260,7 @@ class BiliBiliIE(InfoExtractor):
             }
 
 
+# this extractor doesn't work now
 class BiliBiliBangumiIE(InfoExtractor):
     _VALID_URL = r'https?://bangumi\.bilibili\.com/anime/(?P<id>\d+)'
 
@@ -436,6 +440,71 @@ class BilibiliAudioAlbumIE(BilibiliAudioBaseIE):
         return self.playlist_result(entries, am_id)
 
 
+class BilibiliNewBangumiIE(InfoExtractor):
+    _VALID_URL = r'(?:https?://www\.)?bilibili\.com/bangumi/play/ep(?P<id>\d+)'
+    _TESTS = [{
+        'url': 'https://www.bilibili.com/bangumi/play/ep307448',
+        'md5': '084ae96f913cf13ab626326d86190ddf',
+        'info_dict': {
+            'id': '307448',
+            'ext': 'flv',
+            'title': '异度侵入 ID:INVADED：第3话 SNIPED 瀑布世界',
+            # 'thumbnail': r're:^https?://.*\.jpg$',
+        }
+    }]
+    _APP_KEY = 'iVGUTjsxvpLeuDCf'
+    _BILIBILI_KEY = 'aHRmhWMLkdeMuILqORnYZocwMBpMEOdt'
+    # Don't forget to add ? after url
+    _BILIBILI_API = 'https://bangumi.bilibili.com/player/web_api/v2/playurl'
+    _BILIBILI_TOKEN_API = 'https://api.bilibili.com/x/player/playurl/token'
+
+    def _real_extract(self, url):
+        url, _ = unsmuggle_url(url)
+        video_id = self._search_regex(self._VALID_URL, url, 'video_id', group='id')
+        webpage = self._download_webpage(url, video_id)
+
+        jsondata = self._search_regex(r'window.__INITIAL_STATE__=(.+?);', webpage, 'jsondata', group=1)
+        bgmdata = self._parse_json(jsondata, video_id)  # may throw exception
+
+        try:
+            cid = bgmdata['epInfo']['cid']
+            # aid = bgmdata['epInfo']['aid']
+        except KeyError as e:
+            raise ExtractorError('Failed to extract cid', cause=e, video_id=video_id)
+
+        title = bgmdata.get('h1Title') or self._og_search_title(webpage)
+
+        season = self._search_regex(r'"ssType":(\d+)', webpage, 'season', group=1)
+
+        '''
+        tokens = self._download_json(f'{self._BILIBILI_TOKEN_API}?aid={aid}&cid={cid}', video_id)
+        try:
+           token = tokens['data']['token']
+        except KeyError as e:
+           raise ExtractorError(f'{video_id}: Failed to read JSON', cause = e)
+        '''
+
+        # quality is hardcore to 80
+        # params = f'appkey={self._APP_KEY}&cid={cid}&module=bangumi&otype=json&qn=80&quality=80&season_type={season}&type='
+        params = 'appkey=%s&cid=%s&module=bangumi&otype=json&qn=80&quality=80&season_type=%s&type=' % (self._APP_KEY, cid, season)
+        # may throw exception
+        # urls = self._download_json(f'{self._BILIBILI_API}?{params}&sign={md5(params + self._BILIBILI_KEY)}', video_id)
+        urls = self._download_json('%s?%s&sign=%s' % (self._BILIBILI_API, params, md5(params + self._BILIBILI_KEY)), video_id)
+
+        real_url = try_get(urls, lambda x: x['durl'][0]['url'])
+
+        return {
+            'title': title,
+            'id': video_id,
+            'url': real_url,
+            'ext': urls.get('format'),
+            'size': try_get(urls, lambda x: x['durl'][0]['size']),
+            'http_headers': {
+                'Referer': url,
+            }
+        }
+
+
 class BiliBiliPlayerIE(InfoExtractor):
     _VALID_URL = r'https?://player\.bilibili\.com/player\.html\?.*?\baid=(?P<id>\d+)'
     _TEST = {

youtube_dl/extractor/extractors.py

@@ -105,6 +105,7 @@ from .bilibili import (
     BiliBiliBangumiIE,
     BilibiliAudioIE,
     BilibiliAudioAlbumIE,
+    BilibiliNewBangumiIE,
     BiliBiliPlayerIE,
 )
 from .biobiochiletv import BioBioChileTVIE