yt_dlp/extractor/tudou.py

@@ -20,91 +20,45 @@ class TudouIE(InfoExtractor):
         video_id = self._match_id(url)
         webpage = self._download_webpage(url, video_id)
         data = self._search_json(r'window\.__INITIAL_DATA__\s*=', webpage, 'initial data', video_id)
+        # print('==========')
+        # print(webpage)
+        # print(data)
 
-        # The .json file is much cleaner, thanks for the guide
+        # The json file produced same results as the regex did, but it's much cleaner, thanks for the guide
         video_id = data['data']['data']['data']['extra']['videoId']
+        videoLongId = str(data['data']['data']['data']['extra']['videoLongId'])
         title = data['data']['data']['data']['extra']['videoTitle']
         show_name = data['data']['data']['data']['extra']['showName']
 
-        params1 = {
+        video_url = 'https://play.tudou.com' + data['config']['url']
-            "vid": "XNjAwMjk3Nzk1Mg==",
-            "play_ability": "16782592",
-            "current_showid": "590457",
-            "preferClarity": "2",
-            "extag": "EXT-X-PRIVINF",
-            "master_m3u8": "1",
-            "media_type": "standard,subtitle",
-            "app_ver": "2.1.75",
-            "ccode": "050F",
-            "client_ip": "192.168.1.1",
-            "utid": "bIeZHQHll0oCAXj0ME8oOTqH",
-            "client_ts": "1695807088",
-            "ckey": "140#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",
-            "version": "2.1.75",
-            "vs": "1.0",
-            "pver": "2.1.75",
-            "sver": "2.0",
-            "site": "-1",
-            "aw": "w",
-            "fu": "0",
-            "d": "0",
-            "bt": "pc",
-            "os": "win",
-            "osv": "10",
-            "dq": "auto",
-            "atm": "",
-            "partnerid": "null",
-            "wintype": "interior",
-            "isvert": "0",
-            "vip": "0",
-            "emb": "",
-            "p": "1",
-            "rst": "mp4",
-            "needbf": "2",
-            "avs": "1.0",
-            "callback": "",  # youkuPlayer_call_1695807088862
-            "_t": "08847696931775378"
-        }
-
-        headers1 = {
-            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:109.0) Gecko/20100101 Firefox/117.0",
-            "cookie": "_m_h5_tk=da6bb02219fb5a5f223d967490e4cb67_1695744613439; _m_h5_tk_enc=8c8b786a476dbcc04fd0e19c70a5317d",
-            "Referer": "https://play.tudou.com/"
-        }
-
         # About video_url
         # The video url is not stored in the json file above, instead, the website uses m3u8 scheme
-        # This get.json file contains 4 cdn_url and 4 m3u8_url
+        # With F12 developer tool, I've locked one request.
-        # I'm not sure what cdn_url stands for, but it also delivers the full video
+        # Each time I click the button to play the video, the browser will GET a .m3u8 file which contains urls of all clips of that video, in currently selected resolution (in the webpage player).
-        # I'll put the cdn_url here for now, if needed I can replace it with m3u8_url
+        # In Debugger panel, I also found a get.json file. Can't visit the source url, it'll fail, but can right-click and download the get.json. In get.json file, there're 4 m3u8_url that represent all 4 resolutions available for this video.
-        # I probably need to do a bit more on stuff like cookies, to pass the authentication, I wonder how those values are calculated
+        # These 2 files might be what I should be looking for, guess so.
 
-        # And about params1{} in line 32 (sometimes they call it querystring?)
+        # Tried to copy the link and send the request via PYTHON request module, with headers, fail, 403
-        # IS there a way to know which value is essential for the request process below?
+        # Tried to copy the cUrl and send via Insomnia, fail again, 403 forbidden
-        # Maybe I can just leave it, it works anyway.
+        # Tudou.com is a bit similar to Youku.com(already available in yt-dlp), Tudou.com is acquired by Youku.com many years ago, they're probably sharing some servers and I do find similar domains in these 2 sites
-        # But I still want to simplify it, it's a bit too long.
+        # Therefore I also checked the Youku extractor, but don't know how they get to things like, line 119 'https://log.mmstat.com/eg.js'
+        # I also searched the internet and found another code for Youku.com, in that code there're token settings, appKey, sign, etc..
 
-        data_m3u8 = self._download_json(
+        # So I'm guessing, for Tudou.com, there might be something to do with the token too pass the authentication...
-            'https://ups.youku.com/ups/get.json', video_id,
+        # I'll keep looking into it, but if you can come up with any tips it'll be appreciated.
-            'Downloading JSON metadata',
-            query=params1, headers=headers1)
 
-        list_of_url = []
+        print('==========')
-        for item in data_m3u8['data']['stream']:
+        print(f'videoId = {video_id}')
-            item_of_url = {
+        print(f'videoLongId = {videoLongId}')
-                'width': item['width'],
+        print(f'title = {title}')
-                'height': item['height'],
+        print(f'show_name = {show_name}')
-                'size': item['size'],
+        print(f'video_url = {video_url}')
-                'cdn_url': item['segs'][0]['cdn_url'].replace('http://', 'https://'),
+        print('==========')
-            }
-            list_of_url.append(item_of_url)
-        # Sort urls in descending order, according to size, so the first one in the list will always be the best quality one
-        sorted_list_of_url = sorted(list_of_url, key=lambda d: d['size'], reverse=True)
 
         return {
             'id': video_id,
             'title': title,
             'ext': 'mp4',
-            'url': sorted_list_of_url[0]['cdn_url'],
+            'url': video_url,
             'show_name': show_name,
         }