like lua bilibili

src/you_get/extractors/bilibili.py

@@ -33,8 +33,7 @@ class Bilibili(VideoExtractor):
             {'id': 'flv'},
             {'id': 'hdmp4'},
             {'id': 'mp4'},
-            {'id': 'live'},
+            {'id': 'live'}
-            {'id': 'vc'}
     ]
     fmt2qlt = dict(hdflv=4, flv=3, hdmp4=2, mp4=1)
 
@@ -128,8 +127,6 @@ class Bilibili(VideoExtractor):
             self.bangumi_entry(**kwargs)
         elif 'live.bilibili.com' in self.url:
             self.live_entry(**kwargs)
-        elif 'vc.bilibili.com' in self.url:
-            self.vc_entry(**kwargs)
         else:
             self.entry(**kwargs)
 
@@ -185,48 +182,37 @@ class Bilibili(VideoExtractor):
         self.streams['live']['container'] = 'flv'
         self.streams['live']['size'] = 0
 
-    def vc_entry(self, **kwargs):
-        vc_id = re.search(r'video/(\d+)', self.url)
-        if not vc_id:
-            vc_id = re.search(r'vcdetail\?vc=(\d+)', self.url)
-            if not vc_id:
-                log.wtf('Unknown url pattern')
-        endpoint = 'http://api.vc.bilibili.com/clip/v1/video/detail?video_id={}&need_playurl=1'.format(vc_id.group(1))
-        vc_meta = json.loads(get_content(endpoint, headers=fake_headers))
-        if vc_meta['code'] != 0:
-            log.wtf('{}\n{}'.format(vc_meta['msg'], vc_meta['message']))
-        item = vc_meta['data']['item']
-        self.title = item['description']
-
-        self.streams['vc'] = {}
-        self.streams['vc']['src'] = [item['video_playurl']]
-        self.streams['vc']['container'] = 'mp4'
-        self.streams['vc']['size'] = int(item['video_size'])
-
     def bangumi_entry(self, **kwargs):
         bangumi_id = re.search(r'(\d+)', self.url).group(1)
+        tmp_bangumi_id = re.search(r"varseason_id=\"(\d+)\"",re.sub("\s","",self.page)).group(1)
+
+        if int(bangumi_id) != int(tmp_bangumi_id):
+            bangumi_id = tmp_bangumi_id
+
         bangumi_data = get_bangumi_info(bangumi_id)
         bangumi_payment = bangumi_data.get('payment')
         if bangumi_payment and bangumi_payment['price'] != '0':
             log.w("It's a paid item")
-        # ep_ids = collect_bangumi_epids(bangumi_data)
+        ep_ids = collect_bangumi_epids(bangumi_data)
 
         frag = urllib.parse.urlparse(self.url).fragment
         if frag:
             episode_id = frag
         else:
             episode_id = re.search(r'first_ep_id\s*=\s*"(\d+)"', self.page)
-        # cont = post_content('http://bangumi.bilibili.com/web_api/get_source', post_data=dict(episode_id=episode_id))
+        cont = post_content('http://bangumi.bilibili.com/web_api/get_source', post_data=dict(episode_id=episode_id))
-        # cid = json.loads(cont)['result']['cid']
+        cid = json.loads(cont)['result']['cid']
         cont = get_content('http://bangumi.bilibili.com/web_api/episode/{}.json'.format(episode_id))
         ep_info = json.loads(cont)['result']['currentEpisode']
 
-        index_title = ep_info['indexTitle']
+        long_title = ep_info['longTitle']
-        long_title = ep_info['longTitle'].strip()
+        aid = ep_info['avId']
-        cid = ep_info['danmaku']
 
-        self.title = '{} [{} {}]'.format(self.title, index_title, long_title)
+        idx = 0
-        print(self.title)
+        while ep_ids[idx] != episode_id:
+            idx += 1
+
+        self.title = '{} [{} {}]'.format(self.title, idx+1, long_title)
         self.download_by_vid(cid, bangumi=True, **kwargs)
 
 
@@ -263,8 +249,12 @@ def fetch_sid(cid, aid):
     raise
 
 def collect_bangumi_epids(json_data):
-    eps = json_data['episodes'][::-1]
+    eps = json_data['result']['episodes']
-    return [ep['episode_id'] for ep in eps]
+    eps = sorted(eps, key=lambda item: float(item['index']))
+    result = []
+    for ep in eps:
+        result.append(ep['episode_id'])
+    return result
 
 def get_bangumi_info(bangumi_id):
     BASE_URL = 'http://bangumi.bilibili.com/jsonp/seasoninfo/'
@@ -274,7 +264,7 @@ def get_bangumi_info(bangumi_id):
     season_data = season_data[len('seasonListCallback('):]
     season_data = season_data[: -1 * len(');')]
     json_data = json.loads(season_data)
-    return json_data['result']
+    return json_data
 
 def get_danmuku_xml(cid):
     return get_content('http://comment.bilibili.com/{}.xml'.format(cid))