Merge branch 'release/v0.0.7'

v0.0.6.6

README.md

@@ -7,10 +7,10 @@ pytchat is a python library for fetching youtube live chat.
 pytchat is a python library for fetching youtube live chat
 without using youtube api, Selenium or BeautifulSoup.
 
-pytchatはAPIを使わずにYouTubeチャットを取得するための軽量pythonライブラリです。
+pytchatはAPIを使わずにYouTubeチャットを取得するためのpythonライブラリです。
 
 Other features:
-+ Customizable chat data processors including youtube api compatible one.
++ Customizable [chat data processors](https://github.com/taizan-hokuto/pytchat/wiki/ChatProcessor) including youtube api compatible one.
 + Available on asyncio context. 
 + Quick fetching of initial chat data by generating continuation params
 instead of web scraping.

pytchat/__init__.py

@@ -2,7 +2,7 @@
 pytchat is a python library for fetching youtube live chat without using yt api, Selenium, or BeautifulSoup.
 """
 __copyright__    = 'Copyright (C) 2019 taizan-hokuto'
-__version__      = '0.0.6.5'
+__version__      = '0.0.7'
 __license__      = 'MIT'
 __author__       = 'taizan-hokuto'
 __author_email__ = '55448286+taizan-hokuto@users.noreply.github.com'

pytchat/cli/__init__.py

@@ -22,9 +22,10 @@ def main():
     # Arguments 
     parser = argparse.ArgumentParser(description=f'pytchat v{__version__}')
     parser.add_argument('-v', f'--{Arguments.Name.VIDEO}', type=str,
-         help='Video IDs separated by commas without space')
+         help='Video IDs separated by commas without space.\n'
+         'If ID starts with a hyphen (-), enclose the ID in square brackets.')
     parser.add_argument('-o', f'--{Arguments.Name.OUTPUT}', type=str,
-         help='Output directory (end with "/")', default='./')
+         help='Output directory (end with "/"). default="./"', default='./')
     parser.add_argument(f'--{Arguments.Name.VERSION}', action='store_true',
          help='Settings version')
     Arguments(parser.parse_args().__dict__)
@@ -43,11 +44,17 @@ def main():
                       f" video_id: {video_id}\n"
                       f" channel:  {info.get_channel_name()}\n"
                       f" title:    {info.get_title()}")
+                path = Path(Arguments().output+video_id+'.html')
+                print(f"output path: {path.resolve()}")
                 Extractor(video_id, 
-                  processor = HTMLArchiver(Arguments().output+video_id+'.html')
+                  processor = HTMLArchiver(Arguments().output+video_id+'.html'),
+                  callback = _disp_progress
                 ).extract()
-                print("Extraction end.\n")
+                print("\nExtraction end.\n")
             except (InvalidVideoIdException, NoContentsException) as e:
                 print(e)
         return
     parser.print_help()
+
+def _disp_progress(a,b):
+    print('.',end="",flush=True)

pytchat/core_multithread/livechat.py

@@ -11,8 +11,8 @@ from queue import Queue
 from .buffer import Buffer
 from ..parser.live import Parser
 from .. import config
-from ..exceptions  import ChatParseException,IllegalFunctionCall
+from ..exceptions import ChatParseException, IllegalFunctionCall
-from ..paramgen    import liveparam, arcparam
+from ..paramgen import liveparam, arcparam
 from ..processors.default.processor import DefaultProcessor
 from ..processors.combinator import Combinator
 
@@ -27,7 +27,7 @@ class LiveChat:
     ---------
     video_id : str
         動画ID
-    
+
     seektime : int
         （ライブチャット取得時は無視）
         取得開始するアーカイブ済みチャットの経過時間(秒）
@@ -61,7 +61,7 @@ class LiveChat:
 
     topchat_only : bool
         Trueの場合、上位チャットのみ取得する。
-        
+
     Attributes
     ---------
     _executor : ThreadPoolExecutor
@@ -72,22 +72,20 @@ class LiveChat:
     '''
 
     _setup_finished = False
-    #チャット監視中のListenerのリスト
-    _listeners = []
 
     def __init__(self, video_id,
-                seektime = 0,
+                 seektime=0,
-                processor = DefaultProcessor(),
+                 processor=DefaultProcessor(),
-                buffer = None,
+                 buffer=None,
-                interruptable = True,
+                 interruptable=True,
-                callback = None,
+                 callback=None,
-                done_callback = None,
+                 done_callback=None,
-                direct_mode = False,
+                 direct_mode=False,
-                force_replay = False,
+                 force_replay=False,
-                topchat_only  = False,
+                 topchat_only=False,
-                logger = config.logger(__name__)
+                 logger=config.logger(__name__)
-                ):
+                 ):
-        self.video_id  = video_id
+        self.video_id = video_id
         self.seektime = seektime
         if isinstance(processor, tuple):
             self.processor = Combinator(processor)
@@ -98,57 +96,51 @@ class LiveChat:
         self._done_callback = done_callback
         self._executor = ThreadPoolExecutor(max_workers=2)
         self._direct_mode = direct_mode
-        self._is_alive   = True
+        self._is_alive = True
         self._is_replay = force_replay
-        self._parser = Parser(is_replay = self._is_replay)
+        self._parser = Parser(is_replay=self._is_replay)
         self._pauser = Queue()
         self._pauser.put_nowait(None)
-        self._setup()
         self._first_fetch = True
         self._fetch_url = "live_chat/get_live_chat?continuation="
         self._topchat_only = topchat_only
         self._logger = logger
-        LiveChat._logger = logger
+        if interruptable:
-        if not LiveChat._setup_finished:
+            signal.signal(signal.SIGINT, lambda a, b: self.terminate())
-            LiveChat._setup_finished = True
+        self._setup()
-            if interruptable:
-                signal.signal(signal.SIGINT,  (lambda a, b:  
-                (LiveChat.shutdown(None,signal.SIGINT,b))
-                ))
-        LiveChat._listeners.append(self)
 
     def _setup(self):
-        #direct modeがTrueでcallback未設定の場合例外発生。
+        # direct modeがTrueでcallback未設定の場合例外発生。
         if self._direct_mode:
             if self._callback is None:
                 raise IllegalFunctionCall(
                     "When direct_mode=True, callback parameter is required.")
         else:
-            #direct modeがFalseでbufferが未設定ならばデフォルトのbufferを作成
+            # direct modeがFalseでbufferが未設定ならばデフォルトのbufferを作成
             if self._buffer is None:
-                self._buffer = Buffer(maxsize = 20)
+                self._buffer = Buffer(maxsize=20)
-                #callbackが指定されている場合はcallbackを呼ぶループタスクを作成
+                # callbackが指定されている場合はcallbackを呼ぶループタスクを作成
             if self._callback is None:
-                pass 
+                pass
             else:
-                #callbackを呼ぶループタスクの開始
+                # callbackを呼ぶループタスクの開始
-                self._executor.submit(self._callback_loop,self._callback)
+                self._executor.submit(self._callback_loop, self._callback)
-        #_listenループタスクの開始
+        # _listenループタスクの開始
         listen_task = self._executor.submit(self._startlisten)
-        #add_done_callbackの登録
+        # add_done_callbackの登録
         if self._done_callback is None:
             listen_task.add_done_callback(self.finish)
         else:
             listen_task.add_done_callback(self._done_callback)
 
     def _startlisten(self):
-        time.sleep(0.1)  #sleep shortly to prohibit skipping fetching data
+        time.sleep(0.1)  # sleep shortly to prohibit skipping fetching data
         """Fetch first continuation parameter,
         create and start _listen loop.
         """
-        initial_continuation = liveparam.getparam(self.video_id,3)
+        initial_continuation = liveparam.getparam(self.video_id, 3)
         self._listen(initial_continuation)
-   
+
     def _listen(self, continuation):
         ''' Fetch chat data and store them into buffer,
         get next continuaiton parameter and loop.
@@ -164,33 +156,34 @@ class LiveChat:
                     continuation = self._check_pause(continuation)
                     contents = self._get_contents(
                         continuation, session, headers)
-                    metadata, chatdata =  self._parser.parse(contents)
+                    metadata, chatdata = self._parser.parse(contents)
 
                     timeout = metadata['timeoutMs']/1000
                     chat_component = {
-                        "video_id" : self.video_id,
+                        "video_id": self.video_id,
-                        "timeout"  : timeout,
+                        "timeout": timeout,
-                        "chatdata" : chatdata
+                        "chatdata": chatdata
                     }
-                    time_mark =time.time()
+                    time_mark = time.time()
                     if self._direct_mode:
-                        processed_chat = self.processor.process([chat_component])
+                        processed_chat = self.processor.process(
-                        if isinstance(processed_chat,tuple):
+                            [chat_component])
+                        if isinstance(processed_chat, tuple):
                             self._callback(*processed_chat)
                         else:
                             self._callback(processed_chat)
                     else:
                         self._buffer.put(chat_component)
                     diff_time = timeout - (time.time()-time_mark)
-                    time.sleep(diff_time if diff_time > 0 else 0)        
+                    time.sleep(diff_time if diff_time > 0 else 0)
-                    continuation = metadata.get('continuation')  
+                    continuation = metadata.get('continuation')
         except ChatParseException as e:
             self._logger.debug(f"[{self.video_id}]{str(e)}")
-            return            
+            return
-        except (TypeError , json.JSONDecodeError) :
+        except (TypeError, json.JSONDecodeError):
             self._logger.error(f"{traceback.format_exc(limit = -1)}")
             return
-        
+
         self._logger.debug(f"[{self.video_id}]finished fetching chat.")
 
     def _check_pause(self, continuation):
@@ -202,7 +195,7 @@ class LiveChat:
             '''
             self._pauser.put_nowait(None)
             if not self._is_replay:
-                continuation = liveparam.getparam(self.video_id,3)
+                continuation = liveparam.getparam(self.video_id, 3)
         return continuation
 
     def _get_contents(self, continuation, session, headers):
@@ -214,7 +207,7 @@ class LiveChat:
           -------
             'continuationContents' which includes metadata & chat data.
         '''
-        livechat_json = ( 
+        livechat_json = (
             self._get_livechat_json(continuation, session, headers)
         )
         contents = self._parser.get_contents(livechat_json)
@@ -225,7 +218,7 @@ class LiveChat:
                 self._fetch_url = "live_chat_replay/get_live_chat_replay?continuation="
                 continuation = arcparam.getparam(
                     self.video_id, self.seektime, self._topchat_only)
-                livechat_json = ( self._get_livechat_json(
+                livechat_json = (self._get_livechat_json(
                     continuation, session, headers))
                 reload_continuation = self._parser.reload_continuation(
                     self._parser.get_contents(livechat_json))
@@ -244,26 +237,26 @@ class LiveChat:
         continuation = urllib.parse.quote(continuation)
         livechat_json = None
         status_code = 0
-        url =f"https://www.youtube.com/{self._fetch_url}{continuation}&pbj=1"
+        url = f"https://www.youtube.com/{self._fetch_url}{continuation}&pbj=1"
         for _ in range(MAX_RETRY + 1):
-            with session.get(url ,headers = headers) as resp:
+            with session.get(url, headers=headers) as resp:
                 try:
                     text = resp.text
                     livechat_json = json.loads(text)
                     break
-                except json.JSONDecodeError :
+                except json.JSONDecodeError:
                     time.sleep(1)
                     continue
         else:
             self._logger.error(f"[{self.video_id}]"
-                    f"Exceeded retry count. status_code={status_code}")
+                               f"Exceeded retry count. status_code={status_code}")
             return None
         return livechat_json
-  
+
-    def _callback_loop(self,callback):
+    def _callback_loop(self, callback):
         """ コンストラクタでcallbackを指定している場合、バックグラウンドで
         callbackに指定された関数に一定間隔でチャットデータを投げる。        
-        
+
         Parameter
         ---------
         callback : func
@@ -280,13 +273,13 @@ class LiveChat:
     def get(self):
         """ bufferからデータを取り出し、processorに投げ、
         加工済みのチャットデータを返す。
-        
+
         Returns
              : Processorによって加工されたチャットデータ
         """
         if self._callback is None:
             items = self._buffer.get()
-            return  self.processor.process(items)
+            return self.processor.process(items)
         raise IllegalFunctionCall(
             "既にcallbackを登録済みのため、get()は実行できません。")
 
@@ -304,13 +297,13 @@ class LiveChat:
             return
         if self._pauser.empty():
             self._pauser.put_nowait(None)
-        
+
     def is_alive(self):
         return self._is_alive
 
-    def finish(self,sender):
+    def finish(self, sender):
         '''Listener終了時のコールバック'''
-        try: 
+        try:
             self.terminate()
         except CancelledError:
             self._logger.debug(f'[{self.video_id}]cancelled:{sender}')
@@ -319,14 +312,7 @@ class LiveChat:
         '''
         Listenerを終了する。
         '''
-        self._is_alive = False
+        if self.is_alive():
-        if self._direct_mode == False:
+            self._is_alive = False
-            #bufferにダミーオブジェクトを入れてis_alive()を判定させる
+            self._buffer.put({})
-            self._buffer.put({'chatdata':'','timeout':0}) 
+            self._logger.info(f'[{self.video_id}]終了しました')
-        self._logger.info(f'[{self.video_id}]finished.')
-  
-    @classmethod
-    def shutdown(cls, event, sig = None, handler=None):
-        cls._logger.debug("shutdown...")
-        for t in LiveChat._listeners:
-            t._is_alive = False

pytchat/exceptions.py

@@ -1,46 +1,52 @@
 class ChatParseException(Exception):
     '''
-    チャットデータをパースするライブラリが投げる例外の基底クラス
+    Base exception thrown by the parser
     '''
     pass
 
 class NoYtinitialdataException(ChatParseException):
     '''
-    配信ページ内にチャットデータurlが見つからないときに投げる例外
+    Thrown when the video is not found.
     '''
     pass
 
 class ResponseContextError(ChatParseException):
     '''
-    配信ページでチャットデータ無効の時に投げる例外
+    Thrown when chat data is invalid.
     '''
     pass
 
 class NoLivechatRendererException(ChatParseException):
     '''
-    チャットデータのJSON中にlivechatRendererがない時に投げる例外
+    Thrown when livechatRenderer is missing in JSON.
     '''
     pass
 
 
 class NoContentsException(ChatParseException):
     '''
-    チャットデータのJSON中にContinuationContentsがない時に投げる例外
+    Thrown when ContinuationContents is missing in JSON.
     '''
     pass
 
 class NoContinuationsException(ChatParseException):
     '''
-    チャットデータのContinuationContents中にcontinuationがない時に投げる例外
+    Thrown when continuation is missing in ContinuationContents.
     '''
     pass
 
 class IllegalFunctionCall(Exception):
     '''
-    set_callback()を実行済みにもかかわらず
+    Thrown when get () is called even though 
-    get()を呼び出した場合の例外
+    set_callback () has been executed.
     '''
     pass
 
 class InvalidVideoIdException(Exception):
+    '''
+    Thrown when the video_id is not exist (VideoInfo).
+    '''
     pass
+
+class UnknownConnectionError(Exception):
+    pass

pytchat/paramgen/arcparam.py

@@ -12,6 +12,7 @@ Author: taizan-hokuto (2019) @taizan205
 ver 0.0.1 2019.10.05
 '''
 
+
 def _gen_vid(video_id):
     """generate video_id parameter.
     Parameter
@@ -23,7 +24,7 @@ def _gen_vid(video_id):
         bytes : base64 encoded video_id parameter.
     """
     header_magic = b'\x0A\x0F\x1A\x0D\x0A'
-    header_id =  video_id.encode()
+    header_id = video_id.encode()
     header_sep_1 = b'\x1A\x13\xEA\xA8\xDD\xB9\x01\x0D\x0A\x0B'
     header_terminator = b'\x20\x01'
 
@@ -40,42 +41,46 @@ def _gen_vid(video_id):
         b64enc(reduce(lambda x, y: x+y, item)).decode()
     ).encode()
 
+
 def _nval(val):
     """convert value to byte array"""
-    if val<0: raise ValueError
+    if val < 0:
+        raise ValueError
     buf = b''
     while val >> 7:
         m = val & 0xFF | 0x80
-        buf += m.to_bytes(1,'big')
+        buf += m.to_bytes(1, 'big')
         val >>= 7
-    buf += val.to_bytes(1,'big')
+    buf += val.to_bytes(1, 'big')
     return buf
 
+
 def _build(video_id, seektime, topchat_only):
     switch_01 = b'\x04' if topchat_only else b'\x01'
     if seektime < 0:
-        times =_nval(0)
+        times = _nval(0)
-        switch = b'\x04'       
+        switch = b'\x04'
-    elif seektime == 0:  
+    elif seektime == 0:
-        times =_nval(1)
+        times = _nval(1)
-        switch = b'\x03'       
+        switch = b'\x03'
     else:
-        times =_nval(int(seektime*1000000))
+        times = _nval(int(seektime*1000000))
         switch = b'\x03'
     parity = b'\x00'
 
-    header_magic= b'\xA2\x9D\xB0\xD3\x04'
+    header_magic = b'\xA2\x9D\xB0\xD3\x04'
-    sep_0       = b'\x1A'
+    sep_0 = b'\x1A'
-    vid         = _gen_vid(video_id)
+    vid = _gen_vid(video_id)
-    time_tag    = b'\x28'
+    time_tag = b'\x28'
-    timestamp1  = times
+    timestamp1 = times
-    sep_1       = b'\x30\x00\x38\x00\x40\x00\x48'
+    sep_1 = b'\x30\x00\x38\x00\x40\x00\x48'
-    sep_2       = b'\x52\x1C\x08\x00\x10\x00\x18\x00\x20\x00'
+    sep_2 = b'\x52\x1C\x08\x00\x10\x00\x18\x00\x20\x00'
-    chkstr      = b'\x2A\x0E\x73\x74\x61\x74\x69\x63\x63\x68\x65\x63\x6B\x73\x75\x6D\x40'
+    chkstr = b'\x2A\x0E\x73\x74\x61\x74\x69\x63\x63\x68\x65\x63\x6B\x73\x75\x6D\x40'
-    sep_3       = b'\x00\x58\x03\x60'
+    sep_3 = b'\x00\x58\x03\x60'
-    sep_4       = b'\x68' + parity + b'\x72\x04\x08'
+    sep_4 = b'\x68' + parity + b'\x72\x04\x08'
-    sep_5       = b'\x10' + parity + b'\x78\x00'
+    sep_5 = b'\x10' + parity + b'\x78\x00'
-    body = [
+
+    body = b''.join([
         sep_0,
         _nval(len(vid)),
         vid,
@@ -90,18 +95,17 @@ def _build(video_id, seektime, topchat_only):
         sep_4,
         switch_01,
         sep_5
-    ]
+    ])
 
-    body = reduce(lambda x, y: x+y, body)
+    return urllib.parse.quote(
-    
+        b64enc(header_magic +
-    return  urllib.parse.quote(
+               _nval(len(body)) +
-                b64enc( header_magic +
+               body
-                        _nval(len(body)) +
+               ).decode()
-                        body
+    )
-                ).decode()
-            )
 
-def getparam(video_id, seektime = 0, topchat_only = False):
+
+def getparam(video_id, seektime=0, topchat_only=False):
     '''
     Parameter
     ---------

pytchat/paramgen/liveparam.py

@@ -11,6 +11,8 @@ Author: taizan-hokuto (2019) @taizan205
 
 ver 0.0.1 2019.10.05
 '''
+
+
 def _gen_vid(video_id):
     """generate video_id parameter.
     Parameter
@@ -22,11 +24,11 @@ def _gen_vid(video_id):
     byte[] : base64 encoded video_id parameter.
     """
     header_magic = b'\x0A\x0F\x0A\x0D\x0A'
-    header_id =  video_id.encode()
+    header_id = video_id.encode()
     header_sep_1 = b'\x1A'
     header_sep_2 = b'\x43\xAA\xB9\xC1\xBD\x01\x3D\x0A'
     header_suburl = ('https://www.youtube.com/live_chat?v='
-                    f'{video_id}&is_popout=1').encode()
+                     f'{video_id}&is_popout=1').encode()
     header_terminator = b'\x20\x02'
 
     item = [
@@ -44,62 +46,66 @@ def _gen_vid(video_id):
         b64enc(reduce(lambda x, y: x+y, item)).decode()
     ).encode()
 
-def _tzparity(video_id,times):
+
-    t=0
+def _tzparity(video_id, times):
-    for i,s in enumerate(video_id):
+    t = 0
+    for i, s in enumerate(video_id):
         ss = ord(s)
         if(ss % 2 == 0):
             t += ss*(12-i)
         else:
             t ^= ss*i
 
-    return ((times^t) % 2).to_bytes(1,'big')
+    return ((times ^ t) % 2).to_bytes(1, 'big')
+
 
 def _nval(val):
     """convert value to byte array"""
-    if val<0: raise ValueError
+    if val < 0:
+        raise ValueError
     buf = b''
     while val >> 7:
         m = val & 0xFF | 0x80
-        buf += m.to_bytes(1,'big')
+        buf += m.to_bytes(1, 'big')
         val >>= 7
-    buf += val.to_bytes(1,'big')
+    buf += val.to_bytes(1, 'big')
     return buf
 
+
 def _build(video_id, _ts1, _ts2, _ts3, _ts4, _ts5, topchat_only):
-    #_short_type2
+    # _short_type2
     switch_01 = b'\x04' if topchat_only else b'\x01'
-    parity = _tzparity(video_id, _ts1^_ts2^_ts3^_ts4^_ts5)
+    parity = _tzparity(video_id, _ts1 ^ _ts2 ^ _ts3 ^ _ts4 ^ _ts5)
 
-    header_magic= b'\xD2\x87\xCC\xC8\x03'
+    header_magic = b'\xD2\x87\xCC\xC8\x03'
-    sep_0       = b'\x1A'
+    sep_0 = b'\x1A'
-    vid         = _gen_vid(video_id)
+    vid = _gen_vid(video_id)
-    time_tag    = b'\x28'
+    time_tag = b'\x28'
-    timestamp1  = _nval(_ts1)
+    timestamp1 = _nval(_ts1)
-    sep_1       = b'\x30\x00\x38\x00\x40\x02\x4A'
+    sep_1 = b'\x30\x00\x38\x00\x40\x02\x4A'
-    un_len      = b'\x2B'
+    un_len = b'\x2B'
-    sep_2       = b'\x08'+parity+b'\x10\x00\x18\x00\x20\x00'
+    sep_2 = b'\x08'+parity+b'\x10\x00\x18\x00\x20\x00'
-    chkstr      = b'\x2A\x0E\x73\x74\x61\x74\x69\x63\x63\x68\x65\x63\x6B\x73\x75\x6D'
+    chkstr = b'\x2A\x0E\x73\x74\x61\x74\x69\x63\x63\x68\x65\x63\x6B\x73\x75\x6D'
-    sep_3       = b'\x3A\x00\x40\x00\x4A'
+    sep_3 = b'\x3A\x00\x40\x00\x4A'
-    sep_4_len   = b'\x02'
+    sep_4_len = b'\x02'
-    sep_4       = b'\x08\x01'
+    sep_4 = b'\x08\x01'
-    ts_2_start  = b'\x50'
+    ts_2_start = b'\x50'
-    timestamp2  = _nval(_ts2)
+    timestamp2 = _nval(_ts2)
-    ts_2_end    = b'\x58'
+    ts_2_end = b'\x58'
-    sep_5       = b'\x03'
+    sep_5 = b'\x03'
-    ts_3_start  = b'\x50'
+    ts_3_start = b'\x50'
-    timestamp3  = _nval(_ts3)
+    timestamp3 = _nval(_ts3)
-    ts_3_end    = b'\x58'
+    ts_3_end = b'\x58'
-    timestamp4  = _nval(_ts4)
+    timestamp4 = _nval(_ts4)
-    sep_6       = b'\x68'
+    sep_6 = b'\x68'
-    #switch
+    # switch
-    sep_7       = b'\x82\x01\x04\x08'
+    sep_7 = b'\x82\x01\x04\x08'
-    #switch
+    # switch
-    sep_8       = b'\x10\x00'
+    sep_8 = b'\x10\x00'
-    sep_9       = b'\x88\x01\x00\xA0\x01'
+    sep_9 = b'\x88\x01\x00\xA0\x01'
-    timestamp5  = _nval(_ts5)
+    timestamp5 = _nval(_ts5)
 
-    body = [
+    body = b''.join([
         sep_0,
         _nval(len(vid)),
         vid,
@@ -121,37 +127,35 @@ def _build(video_id, _ts1, _ts2, _ts3, _ts4, _ts5, topchat_only):
         ts_3_end,
         timestamp4,
         sep_6,
-        switch_01,#
+        switch_01,
         sep_7,
-        switch_01,#
+        switch_01,
         sep_8,
         sep_9,
         timestamp5
-    ]
+    ])
+
+    return urllib.parse.quote(
+        b64enc(header_magic +
+               _nval(len(body)) +
+               body
+               ).decode()
+    )
 
-    body = reduce(lambda x, y: x+y, body)
-    
-    return  urllib.parse.quote(
-                b64enc( header_magic +
-                        _nval(len(body)) +
-                        body
-                ).decode()
-            )
 
-    
 def _times(past_sec):
-    
+
     n = int(time.time())
-    
+
-    _ts1= n - random.uniform(0,1*3)    
+    _ts1 = n - random.uniform(0, 1*3)
-    _ts2= n - random.uniform(0.01,0.99)    
+    _ts2 = n - random.uniform(0.01, 0.99)
-    _ts3= n - past_sec + random.uniform(0,1)
+    _ts3 = n - past_sec + random.uniform(0, 1)
-    _ts4= n - random.uniform(10*60,60*60)         
+    _ts4 = n - random.uniform(10*60, 60*60)
-    _ts5= n - random.uniform(0.01,0.99)
+    _ts5 = n - random.uniform(0.01, 0.99)
-    return list(map(lambda x:int(x*1000000),[_ts1,_ts2,_ts3,_ts4,_ts5]))
+    return list(map(lambda x: int(x*1000000), [_ts1, _ts2, _ts3, _ts4, _ts5]))
 
 
-def getparam(video_id, past_sec = 0, topchat_only = False):
+def getparam(video_id, past_sec=0, topchat_only=False):
     '''
     Parameter
     ---------
@@ -160,5 +164,4 @@ def getparam(video_id, past_sec = 0, topchat_only = False):
     topchat_only : bool
         if True, fetch only 'top chat'
     '''
-    return _build(video_id,*_times(past_sec),topchat_only)
+    return _build(video_id, *_times(past_sec), topchat_only)
-

pytchat/processors/compatible/processor.py

@@ -4,17 +4,19 @@ from .renderer.textmessage import LiveChatTextMessageRenderer
 from .renderer.paidmessage import LiveChatPaidMessageRenderer
 from .renderer.paidsticker import LiveChatPaidStickerRenderer
 from .renderer.legacypaid import LiveChatLegacyPaidMessageRenderer
+from .renderer.membership import LiveChatMembershipItemRenderer
 from .. chat_processor import ChatProcessor
 from ... import config
 logger = config.logger(__name__)
 
+
 class CompatibleProcessor(ChatProcessor):
 
     def process(self, chat_components: list):
 
         chatlist = []
         timeout = 0
-        ret={}
+        ret = {}
         ret["kind"] = "youtube#liveChatMessageListResponse"
         ret["etag"] = ""
         ret["nextPageToken"] = ""
@@ -23,20 +25,24 @@ class CompatibleProcessor(ChatProcessor):
             for chat_component in chat_components:
                 timeout += chat_component.get('timeout', 0)
                 chatdata = chat_component.get('chatdata')
-             
+
-                if chatdata is None: break
+                if chatdata is None:
+                    break
                 for action in chatdata:
-                    if action is None: continue
+                    if action is None:
-                    if action.get('addChatItemAction') is None: continue
+                        continue
-                    if action['addChatItemAction'].get('item') is None: continue
+                    if action.get('addChatItemAction') is None:
+                        continue
+                    if action['addChatItemAction'].get('item') is None:
+                        continue
 
                     chat = self.parse(action)
                     if chat:
                         chatlist.append(chat)
         ret["pollingIntervalMillis"] = int(timeout*1000)
-        ret["pageInfo"]={
+        ret["pageInfo"] = {
-            "totalResults":len(chatlist),
+            "totalResults": len(chatlist),
-            "resultsPerPage":len(chatlist),
+            "resultsPerPage": len(chatlist),
         }
         ret["items"] = chatlist
 
@@ -47,8 +53,9 @@ class CompatibleProcessor(ChatProcessor):
         action = sitem.get("addChatItemAction")
         if action:
             item = action.get("item")
-        if item is None: return None
+        if item is None:
-        rd={}
+            return None
+        rd = {}
         try:
             renderer = self.get_renderer(item)
             if renderer == None:
@@ -57,25 +64,26 @@ class CompatibleProcessor(ChatProcessor):
             rd["kind"] = "youtube#liveChatMessage"
             rd["etag"] = ""
             rd["id"] = 'LCC.' + renderer.get_id()
-            rd["snippet"]       = renderer.get_snippet()
+            rd["snippet"] = renderer.get_snippet()
             rd["authorDetails"] = renderer.get_authordetails()
-        except (KeyError,TypeError,AttributeError) as e:
+        except (KeyError, TypeError, AttributeError) as e:
             logger.error(f"Error: {str(type(e))}-{str(e)}")
             logger.error(f"item: {sitem}")
             return None
-        
+
-        return rd        
+        return rd
 
     def get_renderer(self, item):
         if item.get("liveChatTextMessageRenderer"):
             renderer = LiveChatTextMessageRenderer(item)
         elif item.get("liveChatPaidMessageRenderer"):
             renderer = LiveChatPaidMessageRenderer(item)
-        elif item.get( "liveChatPaidStickerRenderer"):
+        elif item.get("liveChatPaidStickerRenderer"):
             renderer = LiveChatPaidStickerRenderer(item)
         elif item.get("liveChatLegacyPaidMessageRenderer"):
             renderer = LiveChatLegacyPaidMessageRenderer(item)
+        elif item.get("liveChatMembershipItemRenderer"):
+            renderer = LiveChatMembershipItemRenderer(item)
         else:
             renderer = None
         return renderer
-

pytchat/processors/compatible/renderer/membership.py

@@ -0,0 +1,40 @@
+from .base import BaseRenderer
+
+
+class LiveChatMembershipItemRenderer(BaseRenderer):
+    def __init__(self, item):
+        super().__init__(item, "newSponsorEvent")
+
+    def get_snippet(self):
+        message = self.get_message(self.renderer)
+        return {
+            "type": self.chattype,
+            "liveChatId": "",
+            "authorChannelId": self.renderer.get("authorExternalChannelId"),
+            "publishedAt": self.get_publishedat(self.renderer.get("timestampUsec", 0)),
+            "hasDisplayContent": True,
+            "displayMessage": message,
+
+        }
+
+    def get_authordetails(self):
+        authorExternalChannelId = self.renderer.get("authorExternalChannelId")
+        # parse subscriber type
+        isVerified, isChatOwner, _, isChatModerator = (
+            self.get_badges(self.renderer)
+        )
+        return {
+            "channelId": authorExternalChannelId,
+            "channelUrl": "http://www.youtube.com/channel/"+authorExternalChannelId,
+            "displayName": self.renderer["authorName"]["simpleText"],
+            "profileImageUrl": self.renderer["authorPhoto"]["thumbnails"][1]["url"],
+            "isVerified": isVerified,
+            "isChatOwner": isChatOwner,
+            "isChatSponsor": True,
+            "isChatModerator": isChatModerator
+        }
+
+    def get_message(self, renderer):
+        message = (renderer["headerSubtext"]["runs"][0]["text"]
+                   )+' / '+(renderer["authorName"]["simpleText"])
+        return message

pytchat/processors/default/processor.py

@@ -4,15 +4,18 @@ from .renderer.textmessage import LiveChatTextMessageRenderer
 from .renderer.paidmessage import LiveChatPaidMessageRenderer
 from .renderer.paidsticker import LiveChatPaidStickerRenderer
 from .renderer.legacypaid import LiveChatLegacyPaidMessageRenderer
+from .renderer.membership import LiveChatMembershipItemRenderer
 from .. chat_processor import ChatProcessor
 from ... import config
+
 logger = config.logger(__name__)
 
+
 class Chatdata:
-    def __init__(self,chatlist:list, timeout:float):
+    def __init__(self, chatlist: list, timeout: float):
         self.items = chatlist
         self.interval = timeout
-    
+
     def tick(self):
         if self.interval == 0:
             time.sleep(1)
@@ -25,6 +28,7 @@ class Chatdata:
             return
         await asyncio.sleep(self.interval/len(self.items))
 
+
 class DefaultProcessor(ChatProcessor):
     def process(self, chat_components: list):
 
@@ -35,24 +39,27 @@ class DefaultProcessor(ChatProcessor):
             for component in chat_components:
                 timeout += component.get('timeout', 0)
                 chatdata = component.get('chatdata')
-                if chatdata is None: continue
+                if chatdata is None:
+                    continue
                 for action in chatdata:
-                    if action is None: continue
+                    if action is None:
-                    if action.get('addChatItemAction') is None: continue
+                        continue
-                    if action['addChatItemAction'].get('item') is None: continue
+                    if action.get('addChatItemAction') is None:
+                        continue
+                    if action['addChatItemAction'].get('item') is None:
+                        continue
 
                     chat = self._parse(action)
                     if chat:
                         chatlist.append(chat)
         return Chatdata(chatlist, float(timeout))
-  
 
     def _parse(self, sitem):
-
         action = sitem.get("addChatItemAction")
         if action:
             item = action.get("item")
-        if item is None: return None
+        if item is None:
+            return None
         try:
             renderer = self._get_renderer(item)
             if renderer == None:
@@ -60,20 +67,22 @@ class DefaultProcessor(ChatProcessor):
 
             renderer.get_snippet()
             renderer.get_authordetails()
-        except (KeyError,TypeError) as e:
+        except (KeyError, TypeError) as e:
             logger.error(f"{str(type(e))}-{str(e)} sitem:{str(sitem)}")
             return None
-        return renderer        
+        return renderer
 
     def _get_renderer(self, item):
         if item.get("liveChatTextMessageRenderer"):
             renderer = LiveChatTextMessageRenderer(item)
         elif item.get("liveChatPaidMessageRenderer"):
             renderer = LiveChatPaidMessageRenderer(item)
-        elif item.get( "liveChatPaidStickerRenderer"):
+        elif item.get("liveChatPaidStickerRenderer"):
             renderer = LiveChatPaidStickerRenderer(item)
         elif item.get("liveChatLegacyPaidMessageRenderer"):
             renderer = LiveChatLegacyPaidMessageRenderer(item)
+        elif item.get("liveChatMembershipItemRenderer"):
+            renderer = LiveChatMembershipItemRenderer(item)
         else:
             renderer = None
-        return renderer
+        return renderer

pytchat/processors/default/renderer/membership.py

@@ -0,0 +1,15 @@
+from .base import BaseRenderer
+
+
+class LiveChatMembershipItemRenderer(BaseRenderer):
+    def __init__(self, item):
+        super().__init__(item, "newSponsor")
+
+    def get_authordetails(self):
+        super().get_authordetails()
+        self.author.isChatSponsor = True
+
+    def get_message(self, renderer):
+        message = (renderer["headerSubtext"]["runs"][0]["text"]
+                   )+' / '+(renderer["authorName"]["simpleText"])
+        return message

pytchat/processors/html_archiver.py

@@ -8,6 +8,11 @@ PATTERN  = re.compile(r"(.*)\(([0-9]+)\)$")
 fmt_headers = ['datetime','elapsed','authorName','message','superchat'
      ,'type','authorChannel']
 
+HEADER_HTML = '''
+<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" "http://www.w3.org/TR/html4/strict.dtd">
+<meta http-equiv="Content-Type" content="text/html;charset=UTF-8">
+'''
+
 class HTMLArchiver(ChatProcessor):
     '''
     HtmlArchiver saves chat data as HTML table format.
@@ -17,6 +22,7 @@ class HTMLArchiver(ChatProcessor):
         super().__init__()
         self.save_path = self._checkpath(save_path)
         with open(self.save_path, mode='a', encoding = 'utf-8') as f:
+            f.write(HEADER_HTML)
             f.write('<table border="1" style="border-collapse: collapse">')
             f.writelines(self._parse_html_header(fmt_headers))
         self.processor = DefaultProcessor()

pytchat/tool/extract/asyncdl.py

@@ -7,12 +7,15 @@ from . worker import ExtractWorker
 from . patch import Patch
 from ... import config 
 from ... paramgen import arcparam
+from ... exceptions import UnknownConnectionError
 from concurrent.futures import CancelledError
+from json import JSONDecodeError
 from urllib.parse import quote
 
 headers = config.headers
 REPLAY_URL = "https://www.youtube.com/live_chat_replay/" \
              "get_live_chat_replay?continuation="
+MAX_RETRY_COUNT = 3
 
 def _split(start, end, count, min_interval_sec = 120):
     """
@@ -53,13 +56,22 @@ def ready_blocks(video_id, duration, div, callback):
             tasks = [_create_block(session, video_id,  seektime, callback)
                 for  seektime in _split(-1, duration, div)]
             return await asyncio.gather(*tasks)
-
+    
     async def _create_block(session, video_id, seektime, callback):
         continuation = arcparam.getparam(video_id, seektime = seektime)
         url = f"{REPLAY_URL}{quote(continuation)}&pbj=1"
-        async with session.get(url, headers = headers) as resp:
+        for _ in range(MAX_RETRY_COUNT):
-            text = await resp.text()
+            try :
-        next_continuation, actions = parser.parse(json.loads(text))
+                async with session.get(url, headers = headers) as resp:
+                    text = await resp.text()
+                next_continuation, actions = parser.parse(json.loads(text))
+                break
+            except JSONDecodeError:
+                await asyncio.sleep(3)
+        else:
+            cancel()
+            raise UnknownConnectionError("Abort: Unknown connection error.")
+
         if actions:
             first = parser.get_offset(actions[0])
             last = parser.get_offset(actions[-1])
@@ -71,6 +83,7 @@ def ready_blocks(video_id, duration, div, callback):
                 first = first,
                 last = last
             )
+            
     """
     fetch initial blocks.
     """  
@@ -95,9 +108,18 @@ def fetch_patch(callback, blocks, video_id):
 
     async def _fetch(continuation,session) -> Patch:
         url = f"{REPLAY_URL}{quote(continuation)}&pbj=1"
-        async with session.get(url,headers = config.headers) as resp:
+        for _ in range(MAX_RETRY_COUNT):
-            chat_json = await resp.text()
+            try:
-        continuation, actions = parser.parse(json.loads(chat_json))
+                async with session.get(url,headers = config.headers) as resp:
+                    chat_json = await resp.text()
+                continuation, actions = parser.parse(json.loads(chat_json))
+                break
+            except JSONDecodeError:
+                await asyncio.sleep(3)
+        else:
+            cancel()
+            raise UnknownConnectionError("Abort: Unknown connection error.")
+
         if actions:
             last = parser.get_offset(actions[-1])
             first = parser.get_offset(actions[0])
@@ -105,6 +127,7 @@ def fetch_patch(callback, blocks, video_id):
                 callback(actions, last - first)
             return Patch(actions, continuation, first, last)
         return Patch(continuation = continuation)
+
     """
     allocate workers and assign blocks.
     """   

tests/test_calculator_parse.py

@@ -36,7 +36,7 @@ def test_process_0():
     chat_component = {
         'video_id':'',
         'timeout':10,
-        'chatdata':load_chatdata(r"tests\testdata\calculator\superchat_0.json")
+        'chatdata':load_chatdata(r"tests/testdata/calculator/superchat_0.json")
     }
     assert SuperchatCalculator().process([chat_component])=={'￥': 6800.0, '€': 2.0}
 
@@ -47,7 +47,7 @@ def test_process_1():
     chat_component = {
         'video_id':'',
         'timeout':10,
-        'chatdata':load_chatdata(r"tests\testdata\calculator\text_only.json")
+        'chatdata':load_chatdata(r"tests/testdata/calculator/text_only.json")
     }
     assert SuperchatCalculator().process([chat_component])=={}
 
@@ -59,7 +59,7 @@ def test_process_2():
         chat_component = {
             'video_id':'',
             'timeout':10,
-            'chatdata':load_chatdata(r"tests\testdata\calculator\replay_end.json")
+            'chatdata':load_chatdata(r"tests/testdata/calculator/replay_end.json")
         }
         assert False
         SuperchatCalculator().process([chat_component])

tests/test_compatible_processor.py

@@ -1,10 +1,11 @@
 import json
 import pytest
-import asyncio,aiohttp
+import asyncio
+import aiohttp
 from pytchat.parser.live import Parser
 from pytchat.processors.compatible.processor import CompatibleProcessor
 from pytchat.exceptions import (
-    NoLivechatRendererException,NoYtinitialdataException,
+    NoLivechatRendererException, NoYtinitialdataException,
     ResponseContextError, NoContentsException)
 
 from pytchat.processors.compatible.renderer.textmessage import LiveChatTextMessageRenderer
@@ -14,6 +15,7 @@ from pytchat.processors.compatible.renderer.legacypaid import LiveChatLegacyPaid
 
 parser = Parser(is_replay=False)
 
+
 def test_textmessage(mocker):
     '''api互換processorのテスト：通常テキストメッセージ'''
     processor = CompatibleProcessor()
@@ -22,16 +24,16 @@ def test_textmessage(mocker):
 
     _, chatdata = parser.parse(parser.get_contents(json.loads(_json)))
     data = {
-        "video_id" : "",
+        "video_id": "",
-        "timeout" : 7,
+        "timeout": 7,
-        "chatdata" : chatdata
+        "chatdata": chatdata
     }
     ret = processor.process([data])
 
-    assert ret["kind"]== "youtube#liveChatMessageListResponse"
+    assert ret["kind"] == "youtube#liveChatMessageListResponse"
-    assert ret["pollingIntervalMillis"]==data["timeout"]*1000
+    assert ret["pollingIntervalMillis"] == data["timeout"]*1000
     assert ret.keys() == {
-        "kind",   "etag",   "pageInfo",   "nextPageToken","pollingIntervalMillis","items"
+        "kind",   "etag",   "pageInfo",   "nextPageToken", "pollingIntervalMillis", "items"
     }
     assert ret["pageInfo"].keys() == {
         "totalResults",   "resultsPerPage"
@@ -48,8 +50,9 @@ def test_textmessage(mocker):
     assert ret["items"][0]["snippet"]["textMessageDetails"].keys() == {
         'messageText'
     }
-    assert "LCC." in ret["items"][0]["id"] 
+    assert "LCC." in ret["items"][0]["id"]
-    assert ret["items"][0]["snippet"]["type"]=="textMessageEvent"
+    assert ret["items"][0]["snippet"]["type"] == "textMessageEvent"
+
 
 def test_newsponcer(mocker):
     '''api互換processorのテスト：メンバ新規登録'''
@@ -59,22 +62,22 @@ def test_newsponcer(mocker):
 
     _, chatdata = parser.parse(parser.get_contents(json.loads(_json)))
     data = {
-        "video_id" : "",
+        "video_id": "",
-        "timeout" : 7,
+        "timeout": 7,
-        "chatdata" : chatdata
+        "chatdata": chatdata
     }
     ret = processor.process([data])
 
-    assert ret["kind"]== "youtube#liveChatMessageListResponse"
+    assert ret["kind"] == "youtube#liveChatMessageListResponse"
-    assert ret["pollingIntervalMillis"]==data["timeout"]*1000
+    assert ret["pollingIntervalMillis"] == data["timeout"]*1000
     assert ret.keys() == {
-        "kind",   "etag",   "pageInfo",   "nextPageToken","pollingIntervalMillis","items"
+        "kind",   "etag",   "pageInfo",   "nextPageToken", "pollingIntervalMillis", "items"
     }
     assert ret["pageInfo"].keys() == {
         "totalResults",   "resultsPerPage"
     }
     assert ret["items"][0].keys() == {
-        "kind",   "etag",   "id",   "snippet","authorDetails"
+        "kind",   "etag",   "id",   "snippet", "authorDetails"
     }
     assert ret["items"][0]["snippet"].keys() == {
         'type', 'liveChatId', 'authorChannelId', 'publishedAt', 'hasDisplayContent', 'displayMessage'
@@ -83,8 +86,44 @@ def test_newsponcer(mocker):
     assert ret["items"][0]["authorDetails"].keys() == {
         'channelId', 'channelUrl', 'displayName', 'profileImageUrl', 'isVerified', 'isChatOwner', 'isChatSponsor', 'isChatModerator'
     }
-    assert "LCC." in ret["items"][0]["id"] 
+    assert "LCC." in ret["items"][0]["id"]
-    assert ret["items"][0]["snippet"]["type"]=="newSponsorEvent"
+    assert ret["items"][0]["snippet"]["type"] == "newSponsorEvent"
+
+
+def test_newsponcer_rev(mocker):
+    '''api互換processorのテスト：メンバ新規登録'''
+    processor = CompatibleProcessor()
+
+    _json = _open_file("tests/testdata/compatible/newSponsor_rev.json")
+
+    _, chatdata = parser.parse(parser.get_contents(json.loads(_json)))
+    data = {
+        "video_id": "",
+        "timeout": 7,
+        "chatdata": chatdata
+    }
+    ret = processor.process([data])
+
+    assert ret["kind"] == "youtube#liveChatMessageListResponse"
+    assert ret["pollingIntervalMillis"] == data["timeout"]*1000
+    assert ret.keys() == {
+        "kind",   "etag",   "pageInfo",   "nextPageToken", "pollingIntervalMillis", "items"
+    }
+    assert ret["pageInfo"].keys() == {
+        "totalResults",   "resultsPerPage"
+    }
+    assert ret["items"][0].keys() == {
+        "kind",   "etag",   "id",   "snippet", "authorDetails"
+    }
+    assert ret["items"][0]["snippet"].keys() == {
+        'type', 'liveChatId', 'authorChannelId', 'publishedAt', 'hasDisplayContent', 'displayMessage'
+
+    }
+    assert ret["items"][0]["authorDetails"].keys() == {
+        'channelId', 'channelUrl', 'displayName', 'profileImageUrl', 'isVerified', 'isChatOwner', 'isChatSponsor', 'isChatModerator'
+    }
+    assert "LCC." in ret["items"][0]["id"]
+    assert ret["items"][0]["snippet"]["type"] == "newSponsorEvent"
 
 
 def test_superchat(mocker):
@@ -95,16 +134,16 @@ def test_superchat(mocker):
 
     _, chatdata = parser.parse(parser.get_contents(json.loads(_json)))
     data = {
-        "video_id" : "",
+        "video_id": "",
-        "timeout" : 7,
+        "timeout": 7,
-        "chatdata" : chatdata
+        "chatdata": chatdata
     }
     ret = processor.process([data])
 
-    assert ret["kind"]== "youtube#liveChatMessageListResponse"
+    assert ret["kind"] == "youtube#liveChatMessageListResponse"
-    assert ret["pollingIntervalMillis"]==data["timeout"]*1000
+    assert ret["pollingIntervalMillis"] == data["timeout"]*1000
     assert ret.keys() == {
-        "kind",   "etag",   "pageInfo",   "nextPageToken","pollingIntervalMillis","items"
+        "kind",   "etag",   "pageInfo",   "nextPageToken", "pollingIntervalMillis", "items"
     }
     assert ret["pageInfo"].keys() == {
         "totalResults",   "resultsPerPage"
@@ -121,8 +160,9 @@ def test_superchat(mocker):
     assert ret["items"][0]["snippet"]["superChatDetails"].keys() == {
         'amountMicros', 'currency', 'amountDisplayString', 'tier', 'backgroundColor'
     }
-    assert "LCC." in ret["items"][0]["id"] 
+    assert "LCC." in ret["items"][0]["id"]
-    assert ret["items"][0]["snippet"]["type"]=="superChatEvent"
+    assert ret["items"][0]["snippet"]["type"] == "superChatEvent"
+
 
 def test_unregistered_currency(mocker):
     processor = CompatibleProcessor()
@@ -132,14 +172,14 @@ def test_unregistered_currency(mocker):
     _, chatdata = parser.parse(parser.get_contents(json.loads(_json)))
 
     data = {
-        "video_id" : "",
+        "video_id": "",
-        "timeout" : 7,
+        "timeout": 7,
-        "chatdata" : chatdata
+        "chatdata": chatdata
     }
     ret = processor.process([data])
     assert ret["items"][0]["snippet"]["superChatDetails"]["currency"] == "[UNREGISTERD]"
 
 
 def _open_file(path):
-    with open(path,mode ='r',encoding = 'utf-8') as f:
+    with open(path, mode='r', encoding='utf-8') as f:
         return f.read()