From 9b306bee138dcdec049db011a406a6c01eba639b Mon Sep 17 00:00:00 2001
From: Johannes Wong <doublevanos@gmail.com>
Date: Sat, 4 Mar 2023 17:33:19 -0800
Subject: [PATCH 1/2] Updated bilibili extractor to support new site code
 parsing when using cookies

---
 src/you_get/extractors/bilibili.py | 62 +++++++++++++++++++++++-------
 1 file changed, 48 insertions(+), 14 deletions(-)

diff --git a/src/you_get/extractors/bilibili.py b/src/you_get/extractors/bilibili.py
index 6335e6dd87..4ac82a26f9 100644
--- a/src/you_get/extractors/bilibili.py
+++ b/src/you_get/extractors/bilibili.py
@@ -338,21 +338,47 @@ def prepare(self, **kwargs):
 
         # bangumi
         elif sort == 'bangumi':
+            ep_id = ""
+            avid = ""
+            cid = ""
+
             initial_state_text = match1(html_content, r'__INITIAL_STATE__=(.*?);\(function\(\)')  # FIXME
-            initial_state = json.loads(initial_state_text)
+            if (initial_state_text):
+                initial_state = json.loads(initial_state_text)
+ 
+                # warn if this bangumi has more than 1 video
+                epn = len(initial_state['epList'])
+                if epn > 1 and not kwargs.get('playlist'):
+                    log.w('This bangumi currently has %s videos. (use --playlist to download all videos.)' % epn)
 
-            # warn if this bangumi has more than 1 video
-            epn = len(initial_state['epList'])
-            if epn > 1 and not kwargs.get('playlist'):
-                log.w('This bangumi currently has %s videos. (use --playlist to download all videos.)' % epn)
+                # set video title
+                self.title = initial_state['h1Title']
 
-            # set video title
-            self.title = initial_state['h1Title']
+                # construct playinfos
+                ep_id = initial_state['epInfo']['id']
+                avid = initial_state['epInfo']['aid']
+                cid = initial_state['epInfo']['cid']
+            else:
+                initial_state_text = match1(html_content, r'\"episodes\":(.*?)\,\"user_status')
+                pinitial_state = json.loads(initial_state_text)
+
+                epn = len(pinitial_state)
+                if epn > 1 and not kwargs.get('playlist'):
+                    log.w('This bangumi currently has %s videos. (use --playlist to download all videos.)' % epn)
+
+                initial_state = {}
+                for dic in pinitial_state:
+                    if dic['link'] == self.url.rstrip('/'):
+                        initial_state = dic
+                        break
+
+                self.title = initial_state['share_copy']
+
+                # construct playinfos
+                ep_id = initial_state['id']
+                avid = initial_state['aid']
+                cid = initial_state['cid']
 
-            # construct playinfos
-            ep_id = initial_state['epInfo']['id']
-            avid = initial_state['epInfo']['aid']
-            cid = initial_state['epInfo']['cid']
             playinfos = []
             api_url = self.bilibili_bangumi_api(avid, cid, ep_id)
             api_content = get_content(api_url, headers=self.bilibili_headers(referer=self.url))
@@ -716,10 +742,18 @@ def download_playlist_by_url(self, url, **kwargs):
                                 self.download(**kwargs)
 
         elif sort == 'bangumi':
+            episodes = []
             initial_state_text = match1(html_content, r'__INITIAL_STATE__=(.*?);\(function\(\)')  # FIXME
-            initial_state = json.loads(initial_state_text)
-            epn, i = len(initial_state['epList']), 0
-            for ep in initial_state['epList']:
+            if (initial_state_text):
+                # initial_state = json.loads(initial_state_text)
+                episodes = json.loads(initial_state_text)['epList']
+            else:
+                initial_state_text = match1(html_content, r'\"episodes\":(.*?)\,\"user_status')
+                # initial_state = json.loads(initial_state_text)
+                episodes = json.loads(initial_state_text)
+
+            epn, i = len(episodes), 0
+            for ep in episodes:
                 i += 1; log.w('Extracting %s of %s videos ...' % (i, epn))
                 ep_id = ep['id']
                 epurl = 'https://www.bilibili.com/bangumi/play/ep%s/' % ep_id

From e443642b81479c5c5a34d8b17b9cb9ba683b63ca Mon Sep 17 00:00:00 2001
From: Johannes Wong <doublevanos@gmail.com>
Date: Sun, 12 Mar 2023 09:46:57 -0700
Subject: [PATCH 2/2] Forgot to handle query parameters when parsing new
 bilibili site code

---
 src/you_get/extractors/bilibili.py | 10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

diff --git a/src/you_get/extractors/bilibili.py b/src/you_get/extractors/bilibili.py
index 4ac82a26f9..2d748b70b6 100644
--- a/src/you_get/extractors/bilibili.py
+++ b/src/you_get/extractors/bilibili.py
@@ -368,7 +368,15 @@ def prepare(self, **kwargs):
 
                 initial_state = {}
                 for dic in pinitial_state:
-                    if dic['link'] == self.url.rstrip('/'):
+                    query_start = self.url.find('?')
+                    url = ""
+
+                    if query_start > 0:
+                       url = self.url[0:query_start].rstrip('/')
+                    else:
+                       url = self.url.rstrip('/')
+
+                    if dic['link'] == url:
                         initial_state = dic
                         break