Merge branch 'hotfix/fix'

Add GitHub actions

.github/workflows/run_test.yml

@@ -0,0 +1,27 @@
+name: Run All UnitTest
+
+on: [push, pull_request]
+
+jobs:
+  build:
+
+    runs-on: ubuntu-latest
+    strategy:
+      max-parallel: 4
+      matrix:
+        python-version: [3.7, 3.8]
+
+    steps:
+      - uses: actions/checkout@v2
+      - name: Set up Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -r requirements.txt -r requirements_test.txt
+      - name: Test with pytest
+        run: |
+          export PYTHONPATH=./
+          pytest --verbose --color=yes

README.md

@@ -5,9 +5,7 @@ pytchat is a python library for fetching youtube live chat.
 
 ## Description
 pytchat is a python library for fetching youtube live chat
-without using youtube api, Selenium or BeautifulSoup.
+without using Selenium or BeautifulSoup.
-
-pytchatは、YouTubeチャットを閲覧するためのpythonライブラリです。
 
 Other features:
 + Customizable [chat data processors](https://github.com/taizan-hokuto/pytchat/wiki/ChatProcessor) including youtube api compatible one.
@@ -16,7 +14,7 @@ Other features:
 instead of web scraping.
 
 For more detailed information, see [wiki](https://github.com/taizan-hokuto/pytchat/wiki). <br>
-より詳細な解説は[wiki](https://github.com/taizan-hokuto/pytchat/wiki/Home_jp)を参照してください。
+[wiki (Japanese)](https://github.com/taizan-hokuto/pytchat/wiki/Home_jp)
 
 ## Install
 ```python
@@ -26,145 +24,61 @@ pip install pytchat
 
 ### CLI
 
-One-liner command.
++ One-liner command.
-Save chat data to html, with embedded custom emojis.
+
++ Save chat data to html with embedded custom emojis.
+
++ Show chat stream (--echo option).
 
 ```bash
-$ pytchat -v https://www.youtube.com/watch?v=ZJ6Q4U_Vg6s -o "c:/temp/"
+$ pytchat -v uIx8l2xlYVY -o "c:/temp/"
 # options:
 #  -v : Video ID or URL that includes ID
 #  -o : output directory (default path: './')
+#  --echo : Show chats.
 # saved filename is [video_id].html
 ```
 
 
-### on-demand mode
+### Fetch chat data (see [wiki](https://github.com/taizan-hokuto/pytchat/wiki/PytchatCore))
 ```python
-from pytchat import LiveChat
+import pytchat
-livechat = LiveChat(video_id = "Zvp1pJpie4I")
+chat = pytchat.create(video_id="uIx8l2xlYVY")
-# It is also possible to specify a URL that includes the video ID:
+while chat.is_alive():
-# livechat = LiveChat("https://www.youtube.com/watch?v=Zvp1pJpie4I")
+    for c in chat.get().sync_items():
-while livechat.is_alive():
-  try:
-    chatdata = livechat.get()
-    for c in chatdata.items:
         print(f"{c.datetime} [{c.author.name}]- {c.message}")
-        chatdata.tick()
-  except KeyboardInterrupt:
-    livechat.terminate()
-    break
 ```
 
-### callback mode
+
+### Output JSON format string (feature of [DefaultProcessor](https://github.com/taizan-hokuto/pytchat/wiki/DefaultProcessor))
 ```python
-from pytchat import LiveChat
+import pytchat
 import time
 
-def main():
+chat = pytchat.create(video_id="uIx8l2xlYVY")
-  livechat = LiveChat(video_id = "Zvp1pJpie4I", callback = disp)
-  while livechat.is_alive():
-    #other background operation.
-    time.sleep(1)
-  livechat.terminate()
-
-#callback function (automatically called)
-def disp(chatdata):
-    for c in chatdata.items:
-        print(f"{c.datetime} [{c.author.name}]- {c.message}")
-        chatdata.tick()
-
-if __name__ == '__main__':
-  main()
-
-```
-
-### asyncio context:
-```python
-from pytchat import LiveChatAsync
-from concurrent.futures import CancelledError
-import asyncio
-
-async def main():
-  livechat = LiveChatAsync("Zvp1pJpie4I", callback = func)
-  while livechat.is_alive():
-    #other background operation.
-    await asyncio.sleep(3)
-
-#callback function is automatically called.
-async def func(chatdata):
-  for c in chatdata.items:
-    print(f"{c.datetime} [{c.author.name}]-{c.message} {c.amountString}")
-    await chatdata.tick_async()
-
-if __name__ == '__main__':
-  try:
-    loop = asyncio.get_event_loop()
-    loop.run_until_complete(main())
-  except CancelledError:
-    pass
-```
-
-
-### youtube api compatible processor:
-```python
-from pytchat import LiveChat, CompatibleProcessor
-import time
-
-chat = LiveChat("Zvp1pJpie4I", 
-  processor = CompatibleProcessor() )
-
 while chat.is_alive():
-  try:
+    print(chat.get().json())
-    data = chat.get()
+    time.sleep(5)
-    polling = data['pollingIntervalMillis']/1000
+    '''
-    for c in data['items']:
+    # Each chat item can also be output in JSON format.
-      if c.get('snippet'):
+    for c in chat.get().items:
-        print(f"[{c['authorDetails']['displayName']}]"
+        print(c.json())
-              f"-{c['snippet']['displayMessage']}")
+    '''     
-        time.sleep(polling/len(data['items']))
-  except KeyboardInterrupt:
-    chat.terminate()
 ```
-### replay: 
-If specified video is not live,
-automatically try to fetch archived chat data.
 
-```python
-from pytchat import LiveChat
 
-def main():
+### other
-  #seektime (seconds): start position of chat.
++ Fetch chat with a buffer ([LiveChat](https://github.com/taizan-hokuto/pytchat/wiki/LiveChat))
-  chat = LiveChat("ojes5ULOqhc", seektime = 60*30)
-  print('Replay from 30:00')
-  try:
-    while chat.is_alive():
-      data = chat.get()
-      for c in data.items:
-        print(f"{c.elapsedTime} [{c.author.name}]-{c.message} {c.amountString}")
-        data.tick()
-  except KeyboardInterrupt:
-    chat.terminate()
 
-if __name__ == '__main__':
++ Use with asyncio ([LiveChatAsync](https://github.com/taizan-hokuto/pytchat/wiki/LiveChatAsync))
-  main()
-```
-### Extract archived chat data as [HTML](https://github.com/taizan-hokuto/pytchat/wiki/HTMLArchiver) or [tab separated values](https://github.com/taizan-hokuto/pytchat/wiki/TSVArchiver).
-```python
-from pytchat import HTMLArchiver, Extractor
 
-video_id = "*******"
++ YT API compatible chat processor ([CompatibleProcessor](https://github.com/taizan-hokuto/pytchat/wiki/CompatibleProcessor))
-ex = Extractor(
+
-    video_id,
++ Extract archived chat data ([Extractor](https://github.com/taizan-hokuto/pytchat/wiki/Extractor))
-    div=10,
-    processor=HTMLArchiver("c:/test.html")
-)
 
-ex.extract()
-print("finished.")
-```
 
 ## Structure of Default Processor
-Each item can be got with `items` function.
+Each item can be got with `sync_items()` function.
 <table>
   <tr>
     <th>name</th>
@@ -298,6 +212,9 @@ Most of source code of CLI refer to:
 
 [PetterKraabol / Twitch-Chat-Downloader](https://github.com/PetterKraabol/Twitch-Chat-Downloader)
 
+Progress bar in CLI is based on:
+
+[vladignatyev/progress.py](https://gist.github.com/vladignatyev/06860ec2040cb497f0f3)
 
 ## Author
 

pytchat/__init__.py

@@ -1,14 +1,29 @@
 """
 pytchat is a lightweight python library to browse youtube livechat without Selenium or BeautifulSoup.
 """
-__copyright__    = 'Copyright (C) 2019 taizan-hokuto'
+__copyright__    = 'Copyright (C) 2019, 2020 taizan-hokuto'
-__version__      = '0.2.7'
+__version__      = '0.4.4'
 __license__      = 'MIT'
 __author__       = 'taizan-hokuto'
 __author_email__ = '55448286+taizan-hokuto@users.noreply.github.com'
 __url__          = 'https://github.com/taizan-hokuto/pytchat'
 
-__all__ = ["core_async","core_multithread","processors"]
+
+from .exceptions import (
+    ChatParseException,
+    ResponseContextError,
+    NoContents,
+    NoContinuation,
+    IllegalFunctionCall,
+    InvalidVideoIdException,
+    UnknownConnectionError,
+    RetryExceedMaxCount,
+    ChatDataFinished,
+    ReceivedUnknownContinuation,
+    FailedExtractContinuation,
+    VideoInfoParseError,
+    PatternUnmatchError
+)
 
 from .api import (
     cli,
@@ -26,7 +41,7 @@ from .api import (
     SimpleDisplayProcessor,
     SpeedCalculator,
     SuperchatCalculator,
-    VideoInfo
+    VideoInfo,
+    create
 )
-
 # flake8: noqa

pytchat/api.py

@@ -1,5 +1,6 @@
 from . import cli
 from . import config
+from .core import create
 from .core_multithread.livechat import LiveChat
 from .core_async.livechat import LiveChatAsync
 from .processors.chat_processor import ChatProcessor
@@ -15,4 +16,24 @@ from .processors.superchat.calculator import SuperchatCalculator
 from .tool.extract.extractor import Extractor
 from .tool.videoinfo import VideoInfo
 
+__all__ = [
+    cli,
+    config,
+    LiveChat,
+    LiveChatAsync,
+    ChatProcessor,
+    CompatibleProcessor,
+    DummyProcessor,
+    DefaultProcessor,
+    Extractor, 
+    HTMLArchiver,
+    TSVArchiver,
+    JsonfileArchiver,
+    SimpleDisplayProcessor,
+    SpeedCalculator,
+    SuperchatCalculator,
+    VideoInfo,
+    create
+]
+
 # flake8: noqa

pytchat/cli/__init__.py

@@ -1,12 +1,16 @@
 import argparse
-
+import asyncio
+try:
+    from asyncio import CancelledError
+except ImportError:
+    from asyncio.futures import CancelledError
 import os
 import signal
-import time
 from json.decoder import JSONDecodeError
 from pathlib import Path
 from httpcore import ReadTimeout as HCReadTimeout, NetworkError as HCNetworkError
 from .arguments import Arguments
+from .echo import Echo
 from .progressbar import ProgressBar
 from .. exceptions import InvalidVideoIdException, NoContents, PatternUnmatchError, UnknownConnectionError
 from .. processors.html_archiver import HTMLArchiver
@@ -38,69 +42,92 @@ def main():
                         help='Save error data when error occurs(".dat" file)')
     parser.add_argument(f'--{Arguments.Name.VERSION}', action='store_true',
                         help='Show version')
+    parser.add_argument(f'--{Arguments.Name.ECHO}', action='store_true',
+        help='Show chats of specified video')
+
     Arguments(parser.parse_args().__dict__)
 
     if Arguments().print_version:
-        print(f'pytchat v{__version__}     © 2019 taizan-hokuto')
+        print(f'pytchat v{__version__}     © 2019,2020 taizan-hokuto')
         return
 
     # Extractor
     if not Arguments().video_ids:
         parser.print_help()
         return
+
+    # Echo
+    if Arguments().echo:
+        if len(Arguments().video_ids) > 1:
+            print("You can specify only one video ID.")
+            return
+        try:
+            Echo(Arguments().video_ids[0]).run()
+        except InvalidVideoIdException as e:
+            print("Invalid video id:", str(e))
+        except Exception as e:
+            print(type(e), str(e))
+        finally:
+            return
+
+    if not os.path.exists(Arguments().output):
+        print("\nThe specified directory does not exist.:{}\n".format(Arguments().output))
+        return
+    try:
+        Runner().run()
+    except CancelledError as e:
+        print(str(e))
+
+
+class Runner:
+    
+    def run(self) -> None:
+        ex = None
+        pbar = None
         for counter, video_id in enumerate(Arguments().video_ids):
-        if '[' in video_id:
-            video_id = video_id.replace('[', '').replace(']', '')
             if len(Arguments().video_ids) > 1:
                 print(f"\n{'-' * 10} video:{counter + 1} of {len(Arguments().video_ids)} {'-' * 10}")
 
             try:
                 video_id = extract_video_id(video_id)
-            if os.path.exists(Arguments().output):
+                separated_path = str(Path(Arguments().output)) + os.path.sep
-                if Arguments().output[-1] != "/" or Arguments().output[-1] != "\\":
+                path = util.checkpath(separated_path + video_id + '.html')
-                    Arguments().output = '/'.join([Arguments().output, os.path.sep])
-                path = util.checkpath(Path.resolve(Path(Arguments().output + video_id + '.html')))
-            else:
-                raise FileNotFoundError
-            err = None
-            for _ in range(3):  # retry 3 times
                 try:
                     info = VideoInfo(video_id)
-                    break
+                except (PatternUnmatchError, JSONDecodeError) as e:
-                except (PatternUnmatchError, JSONDecodeError, InvalidVideoIdException) as e:
+                    print("Cannot parse video information.:{} {}".format(video_id, type(e)))
-                    err = e
-                    time.sleep(2)
-                    continue
-            else:
-                print("Cannot parse video information.:{}".format(video_id))
                     if Arguments().save_error_data:
-                    util.save(err.doc, "ERR", ".dat")
+                        util.save(str(e.doc), "ERR", ".dat")
+                    continue
+                except Exception as e:
+                    print("Cannot parse video information.:{} {}".format(video_id, type(e)))
                     continue
 
                 print(f"\n"
                     f" video_id: {video_id}\n"
                     f" channel:  {info.get_channel_name()}\n"
-                  f" title:    {info.get_title()}")
+                    f" title:    {info.get_title()}\n"
+                    f" output path: {path}")
 
-            print(f" output path: {path}")
                 duration = info.get_duration()
-            pbar = ProgressBar(total=(duration * 1000), status="Extracting")
+                pbar = ProgressBar(total=(duration * 1000), status_txt="Extracting")
                 ex = Extractor(video_id,
-                    callback=pbar._disp,
+                        callback=pbar.disp,
                         div=10)
-            signal.signal(signal.SIGINT, (lambda a, b: cancel(ex, pbar)))
+                signal.signal(signal.SIGINT, (lambda a, b: self.cancel(ex, pbar)))
+
                 data = ex.extract()
                 if data == []:
-                return False
+                    continue
-            pbar.reset("#", "=", total=len(data), status="Rendering  ")
+                pbar.reset("#", "=", total=len(data), status_txt="Rendering  ")
-            processor = HTMLArchiver(Arguments().output + video_id + '.html', callback=pbar._disp)
+                processor = HTMLArchiver(path, callback=pbar.disp)
                 processor.process(
                     [{'video_id': None,
                     'timeout': 1,
                     'chatdata': (action["replayChatItemAction"]["actions"][0] for action in data)}]
                 )
                 processor.finalize()
-            pbar.reset('#', '#', status='Completed   ')
+                pbar.reset('#', '#', status_txt='Completed   ')
                 pbar.close()
                 print()
                 if pbar.is_cancelled():
@@ -108,24 +135,43 @@ def main():
             except InvalidVideoIdException:
                 print("Invalid Video ID or URL:", video_id)
             except NoContents as e:
-            print(e)
+                print(f"Abort:{str(e)}:[{video_id}]")
-        except FileNotFoundError:
+            except (JSONDecodeError, PatternUnmatchError) as e:
-            print("The specified directory does not exist.:{}".format(Arguments().output))
+                print("{}:{}".format(e.msg, video_id))
-        except JSONDecodeError as e:
-            print(e.msg)
-            print("JSONDecodeError.:{}".format(video_id))
                 if Arguments().save_error_data:
-                util.save(e.doc, "ERR_JSON_DECODE", ".dat")
+                    util.save(e.doc, "ERR_", ".dat")
             except (UnknownConnectionError, HCNetworkError, HCReadTimeout) as e:
                 print(f"An unknown network error occurred during the processing of [{video_id}]. : " + str(e))
-        except PatternUnmatchError:
-            print(f"PatternUnmatchError [{video_id}]. ")
             except Exception as e:
-            print(type(e), str(e))
+                print(f"Abort:{str(type(e))} {str(e)[:80]}")
+            finally:
+                clear_tasks()
 
         return
 
-
+    def cancel(self, ex=None, pbar=None) -> None:
-def cancel(ex, pbar):
+        '''Called when keyboard interrupted has occurred.
+        '''
+        print("\nKeyboard interrupted.\n")
+        if ex and pbar:
             ex.cancel()
             pbar.cancel()
+
+
+def clear_tasks():
+    '''
+    Clear remained tasks.
+    Called when internal exception has occurred or
+    after each extraction process is completed.
+    '''
+    async def _shutdown():
+        tasks = [t for t in asyncio.all_tasks()
+                if t is not asyncio.current_task()]
+        for task in tasks:
+            task.cancel()
+            
+    try:
+        loop = asyncio.get_event_loop()
+        loop.run_until_complete(_shutdown())
+    except Exception as e:
+        print(e)

pytchat/cli/arguments.py

@@ -19,6 +19,7 @@ class Arguments(metaclass=Singleton):
         OUTPUT: str = 'output_dir'
         VIDEO_IDS: str = 'video_id'
         SAVE_ERROR_DATA: bool = 'save_error_data'
+        ECHO: bool = 'echo'
 
     def __init__(self,
                  arguments: Optional[Dict[str, Union[str, bool, int]]] = None):
@@ -36,8 +37,9 @@ class Arguments(metaclass=Singleton):
         self.output: str = arguments[Arguments.Name.OUTPUT]
         self.video_ids: List[int] = []
         self.save_error_data: bool = arguments[Arguments.Name.SAVE_ERROR_DATA]
-
+        self.echo: bool = arguments[Arguments.Name.ECHO]
         # Videos
+
         if arguments[Arguments.Name.VIDEO_IDS]:
             self.video_ids = [video_id
                               for video_id in arguments[Arguments.Name.VIDEO_IDS].split(',')]

pytchat/cli/echo.py

@@ -0,0 +1,22 @@
+import pytchat
+from ..exceptions import ChatDataFinished, NoContents
+from ..util.extract_video_id import extract_video_id
+
+
+class Echo:
+    def __init__(self, video_id):
+        self.video_id = extract_video_id(video_id)
+    
+    def run(self):
+        livechat = pytchat.create(self.video_id)
+        while livechat.is_alive():
+            chatdata = livechat.get()
+            for c in chatdata.sync_items():
+                print(f"{c.datetime} [{c.author.name}] {c.message} {c.amountString}")
+        
+        try:
+            livechat.raise_for_status()
+        except (ChatDataFinished, NoContents):
+            print("Chat finished.")
+        except Exception as e:
+            print(type(e), str(e))

pytchat/cli/progressbar.py

@@ -1,5 +1,5 @@
 '''
-This code for this progress bar is based on
+This code is based on
 vladignatyev/progress.py
 https://gist.github.com/vladignatyev/06860ec2040cb497f0f3
 (MIT License)
@@ -9,21 +9,20 @@ import sys
 
 
 class ProgressBar:
-    def __init__(self, total, status):
+    def __init__(self, total, status_txt):
         self._bar_len = 60
         self._cancelled = False
-        self.reset(total=total, status=status)
+        self.reset(total=total, status_txt=status_txt)
-        self._blinker = 0
         
-    def reset(self, symbol_done="=", symbol_space=" ", total=100, status=''):
+    def reset(self, symbol_done="=", symbol_space=" ", total=100, status_txt=''):
-        self.con_width = shutil.get_terminal_size(fallback=(80, 24)).columns
+        self._console_width = shutil.get_terminal_size(fallback=(80, 24)).columns
         self._symbol_done = symbol_done
         self._symbol_space = symbol_space
         self._total = total
-        self._status = status
+        self._status_txt = status_txt
         self._count = 0
 
-    def _disp(self, _, fetched):
+    def disp(self, _, fetched):
         self._progress(fetched, self._total)
 
     def _progress(self, fillin, total):
@@ -39,11 +38,10 @@ class ProgressBar:
             
         bar = self._symbol_done * filled_len + \
               self._symbol_space * (self._bar_len - filled_len)
-        disp = f" [{bar}] {percents:>5.1f}% ...{self._status} "[:self.con_width - 1] + '\r'
+        disp = f" [{bar}] {percents:>5.1f}% ...{self._status_txt} "[:self._console_width - 1] + '\r'
 
         sys.stdout.write(disp)
         sys.stdout.flush()
-        self._blinker += 1
 
     def close(self):
         if not self._cancelled:

pytchat/config/__init__.py

@@ -1,4 +1,4 @@
-import logging
+import logging  # noqa
 from . import mylogger
 headers = {
     'user-agent': 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/84.0.4147.135 Safari/537.36',

pytchat/core/__init__.py

@@ -0,0 +1,7 @@
+from .pytchat import PytchatCore
+from .. util.extract_video_id import extract_video_id
+
+
+def create(video_id: str, **kwargs):
+    _vid = extract_video_id(video_id)
+    return PytchatCore(_vid, **kwargs)

pytchat/core/pytchat.py

@@ -0,0 +1,207 @@
+import httpx
+import json
+import signal
+import time
+import traceback
+import urllib.parse
+from ..parser.live import Parser
+from .. import config
+from .. import exceptions
+from ..paramgen import liveparam, arcparam
+from ..processors.default.processor import DefaultProcessor
+from ..processors.combinator import Combinator
+from ..util.extract_video_id import extract_video_id
+
+headers = config.headers
+MAX_RETRY = 10
+
+
+class PytchatCore:
+    '''
+
+    Parameter
+    ---------
+    video_id : str
+
+    seektime : int
+        start position of fetching chat (seconds).
+        This option is valid for archived chat only.
+        If negative value, chat data posted before the start of the broadcast
+        will be retrieved as well.
+
+    processor : ChatProcessor
+
+    interruptable : bool
+        Allows keyboard interrupts.
+        Set this parameter to False if your own threading program causes
+        the problem.
+
+    force_replay : bool
+        force to fetch archived chat data, even if specified video is live.
+
+    topchat_only : bool
+        If True, get only top chat.
+
+    hold_exception : bool [default:True]
+        If True, when exceptions occur, the exception is held internally,
+        and can be raised by raise_for_status().
+
+    Attributes
+    ---------
+    _is_alive : bool
+        Flag to stop getting chat.
+    '''
+
+    _setup_finished = False
+
+    def __init__(self, video_id,
+                 seektime=-1,
+                 processor=DefaultProcessor(),
+                 interruptable=True,
+                 force_replay=False,
+                 topchat_only=False,
+                 hold_exception=True,
+                 logger=config.logger(__name__),
+                 ):
+        self._video_id = extract_video_id(video_id)
+        self.seektime = seektime
+        if isinstance(processor, tuple):
+            self.processor = Combinator(processor)
+        else:
+            self.processor = processor
+        self._is_alive = True
+        self._is_replay = force_replay
+        self._hold_exception = hold_exception
+        self._exception_holder = None
+        self._parser = Parser(
+            is_replay=self._is_replay,
+            exception_holder=self._exception_holder
+        )
+        self._first_fetch = True
+        self._fetch_url = "live_chat/get_live_chat?continuation="
+        self._topchat_only = topchat_only
+        self._logger = logger
+        if interruptable:
+            signal.signal(signal.SIGINT, lambda a, b: self.terminate())
+        self._setup()
+
+    def _setup(self):
+        time.sleep(0.1)  # sleep shortly to prohibit skipping fetching data
+        """Fetch first continuation parameter,
+        create and start _listen loop.
+        """
+        self.continuation = liveparam.getparam(self._video_id, 3)
+
+    def _get_chat_component(self):
+        
+        ''' Fetch chat data and store them into buffer,
+        get next continuaiton parameter and loop.
+
+        Parameter
+        ---------
+        continuation : str
+            parameter for next chat data
+        '''
+        try:
+            with httpx.Client(http2=True) as client:
+                if self.continuation and self._is_alive:
+                    contents = self._get_contents(self.continuation, client, headers)
+                    metadata, chatdata = self._parser.parse(contents)
+                    timeout = metadata['timeoutMs'] / 1000
+                    chat_component = {
+                        "video_id": self._video_id,
+                        "timeout": timeout,
+                        "chatdata": chatdata
+                    }
+                    self.continuation = metadata.get('continuation')
+                    return chat_component
+        except exceptions.ChatParseException as e:
+            self._logger.debug(f"[{self._video_id}]{str(e)}")
+            self._raise_exception(e)
+        except (TypeError, json.JSONDecodeError) as e:
+            self._logger.error(f"{traceback.format_exc(limit=-1)}")
+            self._raise_exception(e)
+
+        self._logger.debug(f"[{self._video_id}]finished fetching chat.")
+        self._raise_exception(exceptions.ChatDataFinished)
+
+    def _get_contents(self, continuation, client, headers):
+        '''Get 'continuationContents' from livechat json.
+           If contents is None at first fetching,
+           try to fetch archive chat data.
+
+          Return:
+          -------
+            'continuationContents' which includes metadata & chat data.
+        '''
+        livechat_json = (
+            self._get_livechat_json(continuation, client, headers)
+        )
+        contents = self._parser.get_contents(livechat_json)
+        if self._first_fetch:
+            if contents is None or self._is_replay:
+                '''Try to fetch archive chat data.'''
+                self._parser.is_replay = True
+                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(continuation, client, headers))
+                reload_continuation = self._parser.reload_continuation(
+                    self._parser.get_contents(livechat_json))
+                if reload_continuation:
+                    livechat_json = (self._get_livechat_json(
+                        reload_continuation, client, headers))
+                contents = self._parser.get_contents(livechat_json)
+                self._is_replay = True
+            self._first_fetch = False
+        return contents
+
+    def _get_livechat_json(self, continuation, client, headers):
+        '''
+        Get json which includes chat data.
+        '''
+        continuation = urllib.parse.quote(continuation)
+        livechat_json = None
+        err = None
+        url = f"https://www.youtube.com/{self._fetch_url}{continuation}&pbj=1"
+        for _ in range(MAX_RETRY + 1):
+            with client:
+                try:
+                    livechat_json = client.get(url, headers=headers).json()
+                    break
+                except (json.JSONDecodeError, httpx.ConnectTimeout, httpx.ReadTimeout, httpx.ConnectError) as e:
+                    err = e
+                    time.sleep(2)
+                    continue
+        else:
+            self._logger.error(f"[{self._video_id}]"
+                               f"Exceeded retry count. Last error: {str(err)}")
+            self._raise_exception(exceptions.RetryExceedMaxCount())
+        return livechat_json
+    
+    def get(self):
+        if self.is_alive():
+            chat_component = self._get_chat_component()
+            return self.processor.process([chat_component])
+        else:
+            return []
+
+    def is_replay(self):
+        return self._is_replay
+
+    def is_alive(self):
+        return self._is_alive
+
+    def terminate(self):
+        self._is_alive = False
+        self.processor.finalize()
+
+    def raise_for_status(self):
+        if self._exception_holder is not None:
+            raise self._exception_holder
+
+    def _raise_exception(self, exception: Exception = None):
+        self._is_alive = False
+        if self._hold_exception is False:
+            raise exception
+        self._exception_holder = exception

pytchat/core_async/buffer.py

@@ -4,13 +4,13 @@ import asyncio
 
 class Buffer(asyncio.Queue):
     '''
-    チャットデータを格納するバッファの役割を持つFIFOキュー
+    Buffer for storing chat data.
 
     Parameter
     ---------
     maxsize : int
-        格納するチャットブロックの最大個数。0の場合は無限。
+        Maximum number of chat blocks to be stored.
-        最大値を超える場合は古いチャットブロックから破棄される。
+        If it exceeds the maximum, the oldest chat block will be discarded.
     '''
 
     def __init__(self, maxsize=0):

pytchat/core_async/livechat.py

@@ -22,54 +22,51 @@ MAX_RETRY = 10
 
 
 class LiveChatAsync:
-    '''asyncioを利用してYouTubeのライブ配信のチャットデータを取得する。
+    '''LiveChatAsync object fetches chat data and stores them
+    in a buffer with asyncio.
 
     Parameter
     ---------
     video_id : str
-        動画ID
 
     seektime : int
-        （ライブチャット取得時は無視）
+        start position of fetching chat (seconds).
-        取得開始するアーカイブ済みチャットの経過時間(秒）
+        This option is valid for archived chat only.
-        マイナス値を指定した場合は、配信開始前のチャットも取得する。
+        If negative value, chat data posted before the start of the broadcast
+        will be retrieved as well.
 
     processor : ChatProcessor
-        チャットデータを加工するオブジェクト
 
-    buffer : Buffer(maxsize:20[default])
+    buffer : Buffer
-        チャットデータchat_componentを格納するバッファ。
+        buffer of chat data fetched background.
-        maxsize : 格納できるchat_componentの個数
-        default値20個。1個で約5~10秒分。
 
     interruptable : bool
-        Ctrl+Cによる処理中断を行うかどうか。
+        Allows keyboard interrupts.
+        Set this parameter to False if your own threading program causes
+        the problem.
 
     callback : func
-        _listen()関数から一定間隔で自動的に呼びだす関数。
+        function called periodically from _listen().
 
     done_callback : func
-        listener終了時に呼び出すコールバック。
+        function called when listener ends.
 
     exception_handler : func
-        例外を処理する関数
 
     direct_mode : bool
-        Trueの場合、bufferを使わずにcallbackを呼ぶ。
+        If True, invoke specified callback function without using buffer.
-        Trueの場合、callbackの設定が必須
+        callback is required. If not, IllegalFunctionCall will be raised.
-        (設定していない場合IllegalFunctionCall例外を発生させる）
 
     force_replay : bool
-        Trueの場合、ライブチャットが取得できる場合であっても
+        force to fetch archived chat data, even if specified video is live.
-        強制的にアーカイブ済みチャットを取得する。
 
     topchat_only : bool
-        Trueの場合、上位チャットのみ取得する。
+        If True, get only top chat.
 
     Attributes
     ---------
     _is_alive : bool
-        チャット取得を停止するためのフラグ
+        Flag to stop getting chat.
     '''
 
     _setup_finished = False
@@ -114,31 +111,30 @@ class LiveChatAsync:
             self._set_exception_handler(exception_handler)
         if interruptable:
             signal.signal(signal.SIGINT,
-                (lambda a, b: asyncio.create_task(
+                (lambda a, b: self._keyboard_interrupt()))
-                 LiveChatAsync.shutdown(None, signal.SIGINT, b))))
         self._setup()
 
     def _setup(self):
-        # direct modeがTrueでcallback未設定の場合例外発生。
+        # An exception is raised when direct mode is true and no callback is set.
         if self._direct_mode:
             if self._callback is None:
                 raise exceptions.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
+            # Create a default buffer if `direct_mode` is False and buffer is not set.
             if self._buffer is None:
                 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
+                # Create a loop task to call callback if the `callback` param is specified.
             if self._callback is None:
                 pass
             else:
-                # callbackを呼ぶループタスクの開始
+                # Create a loop task to call callback if the `callback` param is specified.
                 loop = asyncio.get_event_loop()
                 loop.create_task(self._callback_loop(self._callback))
-        # _listenループタスクの開始
+        # Start a loop task for _listen()
         loop = asyncio.get_event_loop()
         self.listen_task = loop.create_task(self._startlisten())
-        # add_done_callbackの登録
+        # Register add_done_callback
         if self._done_callback is None:
             self.listen_task.add_done_callback(self._finish)
         else:
@@ -246,30 +242,30 @@ class LiveChatAsync:
         '''
         continuation = urllib.parse.quote(continuation)
         livechat_json = None
-        status_code = 0
         url = f"https://www.youtube.com/{self._fetch_url}{continuation}&pbj=1"
         for _ in range(MAX_RETRY + 1):
             try:
                 resp = await client.get(url, headers=headers)
                 livechat_json = resp.json()
                 break
-            except (httpx.HTTPError, json.JSONDecodeError):
+            except (json.JSONDecodeError, httpx.HTTPError):
                 await asyncio.sleep(1)
                 continue
         else:
             self._logger.error(f"[{self._video_id}]"
-                               f"Exceeded retry count. status_code={status_code}")
+                               f"Exceeded retry count.")
             return None
         return livechat_json
 
     async 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
+        """ If a callback is specified in the constructor,
-        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。
+        it throws chat data at regular intervals to the
+        function specified in the callback in the backgroun
 
         Parameter
         ---------
         callback : func
-            加工済みのチャットデータを渡す先の関数。
+            function to which the processed chat data is passed.
         """
         while self.is_alive():
             items = await self._buffer.get()
@@ -280,11 +276,13 @@ class LiveChatAsync:
                 await self._callback(processed_chat)
 
     async 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。
+        Retrieves data from the buffer,
+        throws it to the processor,
+        and returns the processed chat data.
 
         Returns
-             : Processorによって加工されたチャットデータ
+             : Chat data processed by the Processor
         """
         if self._callback is None:
             if self.is_alive():
@@ -293,7 +291,7 @@ class LiveChatAsync:
             else:
                 return []
         raise exceptions.IllegalFunctionCall(
-            "既にcallbackを登録済みのため、get()は実行できません。")
+            "Callback parameter is already set, so get() cannot be performed.")
 
     def is_replay(self):
         return self._is_replay
@@ -314,7 +312,7 @@ class LiveChatAsync:
         return self._is_alive
 
     def _finish(self, sender):
-        '''Listener終了時のコールバック'''
+        '''Called when the _listen() task finished.'''
         try:
             self._task_finished()
         except CancelledError:
@@ -327,9 +325,13 @@ class LiveChatAsync:
         self._buffer.put_nowait({})
         self.processor.finalize()
     
+    def _keyboard_interrupt(self):
+        self.exception = exceptions.ChatDataFinished()
+        self.terminate()
+
     def _task_finished(self):
         '''
-        Listenerを終了する。
+        Terminate fetching chats.
         '''
         if self.is_alive():
             self.terminate()
@@ -339,7 +341,7 @@ class LiveChatAsync:
             self.exception = e
             if not isinstance(e, exceptions.ChatParseException):
                 self._logger.error(f'Internal exception - {type(e)}{str(e)}')
-        self._logger.info(f'[{self._video_id}]終了しました')
+        self._logger.info(f'[{self._video_id}] finished.')
 
     def raise_for_status(self):
         if self.exception is not None:
@@ -349,15 +351,3 @@ class LiveChatAsync:
     def _set_exception_handler(cls, handler):
         loop = asyncio.get_event_loop()
         loop.set_exception_handler(handler)
-
-    @classmethod
-    async def shutdown(cls, event, sig=None, handler=None):
-        cls._logger.debug("shutdown...")
-        tasks = [t for t in asyncio.all_tasks() if t is not
-                 asyncio.current_task()]
-        [task.cancel() for task in tasks]
-
-        cls._logger.debug("complete remaining tasks...")
-        await asyncio.gather(*tasks, return_exceptions=True)
-        loop = asyncio.get_event_loop()
-        loop.stop()

pytchat/core_multithread/buffer.py

@@ -4,13 +4,13 @@ import queue
 
 class Buffer(queue.Queue):
     '''
-    チャットデータを格納するバッファの役割を持つFIFOキュー
+    Buffer for storing chat data.
 
     Parameter
     ---------
-    max_size : int
+    maxsize : int
-        格納するチャットブロックの最大個数。0の場合は無限。
+        Maximum number of chat blocks to be stored.
-        最大値を超える場合は古いチャットブロックから破棄される。
+        If it exceeds the maximum, the oldest chat block will be discarded.
     '''
 
     def __init__(self, maxsize=0):

pytchat/core_multithread/livechat.py

@@ -21,54 +21,53 @@ MAX_RETRY = 10
 
 
 class LiveChat:
-    ''' スレッドプールを利用してYouTubeのライブ配信のチャットデータを取得する
+    '''
+    LiveChat object fetches chat data and stores them
+    in a buffer with ThreadpoolExecutor.
 
     Parameter
     ---------
     video_id : str
-        動画ID
 
     seektime : int
-        （ライブチャット取得時は無視）
+        start position of fetching chat (seconds).
-        取得開始するアーカイブ済みチャットの経過時間(秒）
+        This option is valid for archived chat only.
-        マイナス値を指定した場合は、配信開始前のチャットも取得する。
+        If negative value, chat data posted before the start of the broadcast
+        will be retrieved as well.
 
     processor : ChatProcessor
-        チャットデータを加工するオブジェクト
 
-    buffer : Buffer(maxsize:20[default])
+    buffer : Buffer
-        チャットデータchat_componentを格納するバッファ。
+        buffer of chat data fetched background.
-        maxsize : 格納できるchat_componentの個数
-        default値20個。1個で約5~10秒分。
 
     interruptable : bool
-        Ctrl+Cによる処理中断を行うかどうか。
+        Allows keyboard interrupts.
+        Set this parameter to False if your own threading program causes
+        the problem.
 
     callback : func
-        _listen()関数から一定間隔で自動的に呼びだす関数。
+        function called periodically from _listen().
 
     done_callback : func
-        listener終了時に呼び出すコールバック。
+        function called when listener ends.
 
     direct_mode : bool
-        Trueの場合、bufferを使わずにcallbackを呼ぶ。
+        If True, invoke specified callback function without using buffer.
-        Trueの場合、callbackの設定が必須
+        callback is required. If not, IllegalFunctionCall will be raised.
-        (設定していない場合IllegalFunctionCall例外を発生させる）
 
     force_replay : bool
-        Trueの場合、ライブチャットが取得できる場合であっても
+        force to fetch archived chat data, even if specified video is live.
-        強制的にアーカイブ済みチャットを取得する。
 
     topchat_only : bool
-        Trueの場合、上位チャットのみ取得する。
+        If True, get only top chat.
 
     Attributes
     ---------
     _executor : ThreadPoolExecutor
-        チャットデータ取得ループ（_listen）用のスレッド
+        This is used for _listen() loop.
 
     _is_alive : bool
-        チャット取得を停止するためのフラグ
+        Flag to stop getting chat.
     '''
 
     _setup_finished = False
@@ -112,24 +111,24 @@ class LiveChat:
         self._setup()
 
     def _setup(self):
-        # direct modeがTrueでcallback未設定の場合例外発生。
+        # An exception is raised when direct mode is true and no callback is set.
         if self._direct_mode:
             if self._callback is None:
                 raise exceptions.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
+            # Create a default buffer if `direct_mode` is False and buffer is not set.
             if self._buffer is None:
                 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
+                # Create a loop task to call callback if the `callback` param is specified.
             if self._callback is None:
                 pass
             else:
-                # callbackを呼ぶループタスクの開始
+                # Start a loop task calling callback function.
                 self._executor.submit(self._callback_loop, self._callback)
-        # _listenループタスクの開始
+        # Start a loop task for _listen()
         self.listen_task = self._executor.submit(self._startlisten)
-        # add_done_callbackの登録
+        # Register add_done_callback
         if self._done_callback is None:
             self.listen_task.add_done_callback(self._finish)
         else:
@@ -236,30 +235,30 @@ 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"
         for _ in range(MAX_RETRY + 1):
             with client:
                 try:
                     livechat_json = client.get(url, headers=headers).json()
                     break
-                except json.JSONDecodeError:
+                except (json.JSONDecodeError, httpx.HTTPError):
-                    time.sleep(1)
+                    time.sleep(2)
                     continue
         else:
             self._logger.error(f"[{self._video_id}]"
-                               f"Exceeded retry count. status_code={status_code}")
+                               f"Exceeded retry count.")
             raise exceptions.RetryExceedMaxCount()
         return livechat_json
 
     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
+        """ If a callback is specified in the constructor,
-        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。
+        it throws chat data at regular intervals to the
+        function specified in the callback in the backgroun
 
         Parameter
         ---------
         callback : func
-            加工済みのチャットデータを渡す先の関数。
+            function to which the processed chat data is passed.
         """
         while self.is_alive():
             items = self._buffer.get()
@@ -270,11 +269,13 @@ class LiveChat:
                 self._callback(processed_chat)
 
     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。
+        Retrieves data from the buffer,
+        throws it to the processor,
+        and returns the processed chat data.
 
         Returns
-             : Processorによって加工されたチャットデータ
+             : Chat data processed by the Processor
         """
         if self._callback is None:
             if self.is_alive():
@@ -283,7 +284,7 @@ class LiveChat:
             else:
                 return []
         raise exceptions.IllegalFunctionCall(
-            "既にcallbackを登録済みのため、get()は実行できません。")
+            "Callback parameter is already set, so get() cannot be performed.")
 
     def is_replay(self):
         return self._is_replay
@@ -304,13 +305,16 @@ class LiveChat:
         return self._is_alive
 
     def _finish(self, sender):
-        '''Listener終了時のコールバック'''
+        '''Called when the _listen() task finished.'''
         try:
             self._task_finished()
         except CancelledError:
             self._logger.debug(f'[{self._video_id}] cancelled:{sender}')
 
     def terminate(self):
+        '''
+        Terminate fetching chats.
+        '''
         if self._pauser.empty():
             self._pauser.put_nowait(None)
         self._is_alive = False
@@ -319,9 +323,6 @@ class LiveChat:
         self.processor.finalize()
 
     def _task_finished(self):
-        '''
-        Listenerを終了する。
-        '''
         if self.is_alive():
             self.terminate()
         try:
@@ -330,7 +331,7 @@ class LiveChat:
             self.exception = e
             if not isinstance(e, exceptions.ChatParseException):
                 self._logger.error(f'Internal exception - {type(e)}{str(e)}')
-        self._logger.info(f'[{self._video_id}]終了しました')
+        self._logger.info(f'[{self._video_id}] finished.')
 
     def raise_for_status(self):
         if self.exception is not None:

pytchat/exceptions.py

@@ -76,6 +76,6 @@ class PatternUnmatchError(VideoInfoParseError):
     '''
     Thrown when failed to parse video info with unmatched pattern.
     '''
-    def __init__(self, doc):
+    def __init__(self, doc=''):
         self.msg = "PatternUnmatchError"
         self.doc = doc

pytchat/paramgen/arcparam_mining.py

@@ -1,133 +0,0 @@
-from base64 import urlsafe_b64encode as b64enc
-from functools import reduce
-import urllib.parse
-
-'''
-Generate continuation parameter of youtube replay chat.
-
-Author: taizan-hokuto (2019) @taizan205
-
-ver 0.0.1 2019.10.05
-'''
-
-
-def _gen_vid_long(video_id):
-    """generate video_id parameter.
-    Parameter
-    ---------
-        video_id : str
-
-    Return
-    ---------
-    byte[] : base64 encoded video_id parameter.
-    """
-    header_magic = b'\x0A\x0F\x1A\x0D\x0A'
-    header_id = video_id.encode()
-    header_sep_1 = b'\x1A\x13\xEA\xA8\xDD\xB9\x01\x0D\x0A\x0B'
-    header_terminator = b'\x20\x01'
-
-    item = [
-        header_magic,
-        _nval(len(header_id)),
-        header_id,
-        header_sep_1,
-        header_id,
-        header_terminator
-    ]
-
-    return urllib.parse.quote(
-        b64enc(reduce(lambda x, y: x + y, item)).decode()
-    ).encode()
-
-
-def _gen_vid(video_id):
-    """generate video_id parameter.
-    Parameter
-    ---------
-        video_id : str
-
-    Return
-    ---------
-        bytes : base64 encoded video_id parameter.
-    """
-    header_magic = b'\x0A\x0F\x1A\x0D\x0A'
-    header_id = video_id.encode()
-    header_terminator = b'\x20\x01'
-
-    item = [
-        header_magic,
-        _nval(len(header_id)),
-        header_id,
-        header_terminator
-    ]
-
-    return urllib.parse.quote(
-        b64enc(reduce(lambda x, y: x + y, item)).decode()
-    ).encode()
-
-
-def _nval(val):
-    """convert value to byte array"""
-    if val < 0:
-        raise ValueError
-    buf = b''
-    while val >> 7:
-        m = val & 0xFF | 0x80
-        buf += m.to_bytes(1, 'big')
-        val >>= 7
-    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:
-        raise ValueError("seektime must be greater than or equal to zero.")
-    if seektime == 0:
-        times = b''
-    else:
-        times = _nval(int(seektime * 1000))
-    if seektime > 0:
-        _len_time = b'\x5A' + (len(times) + 1).to_bytes(1, 'big') + b'\x10'
-    else:
-        _len_time = b''
-
-    header_magic = b'\xA2\x9D\xB0\xD3\x04'
-    sep_0 = b'\x1A'
-    vid = _gen_vid(video_id)
-    _tag = b'\x40\x01'
-    timestamp1 = times
-    sep_1 = b'\x60\x04\x72\x02\x08'
-    terminator = b'\x78\x01'
-
-    body = [
-        sep_0,
-        _nval(len(vid)),
-        vid,
-        _tag,
-        _len_time,
-        timestamp1,
-        sep_1,
-        switch_01,
-        terminator
-    ]
-
-    body = reduce(lambda x, y: x + y, body)
-
-    return urllib.parse.quote(
-        b64enc(header_magic + _nval(len(body)) + body
-               ).decode()
-    )
-
-
-def getparam(video_id, seektime=0.0, topchat_only=False):
-    '''
-    Parameter
-    ---------
-    seektime : int
-        unit:seconds
-        start position of fetching chat data.
-    topchat_only : bool
-        if True, fetch only 'top chat'
-    '''
-    return _build(video_id, seektime, topchat_only)

pytchat/parser/live.py

@@ -8,15 +8,26 @@ from .. import exceptions
 
 
 class Parser:
+    '''
+    Parser of chat json.
     
-    __slots__ = ['is_replay']
+    Parameter
+    ----------
+    is_replay : bool
 
-    def __init__(self, is_replay):
+    exception_holder : Object [default:Npne]
+        The object holding exceptions.
+        This is passed from the parent livechat object.
+    '''
+    __slots__ = ['is_replay', 'exception_holder']
+
+    def __init__(self, is_replay, exception_holder=None):
         self.is_replay = is_replay
+        self.exception_holder = exception_holder
 
     def get_contents(self, jsn):
         if jsn is None:
-            raise exceptions.IllegalFunctionCall('Called with none JSON object.')
+            self.raise_exception(exceptions.IllegalFunctionCall('Called with none JSON object.'))
         if jsn['response']['responseContext'].get('errors'):
             raise exceptions.ResponseContextError(
                 'The video_id would be wrong, or video is deleted or private.')
@@ -42,11 +53,11 @@ class Parser:
 
         if contents is None:
             '''Broadcasting end or cannot fetch chat stream'''
-            raise exceptions.NoContents('Chat data stream is empty.')
+            self.raise_exception(exceptions.NoContents('Chat data stream is empty.'))
 
         cont = contents['liveChatContinuation']['continuations'][0]
         if cont is None:
-            raise exceptions.NoContinuation('No Continuation')
+            self.raise_exception(exceptions.NoContinuation('No Continuation'))
         metadata = (cont.get('invalidationContinuationData')
                     or cont.get('timedContinuationData')
                     or cont.get('reloadContinuationData')
@@ -54,13 +65,13 @@ class Parser:
                     )
         if metadata is None:
             if cont.get("playerSeekContinuationData"):
-                raise exceptions.ChatDataFinished('Finished chat data')
+                self.raise_exception(exceptions.ChatDataFinished('Finished chat data'))
             unknown = list(cont.keys())[0]
             if unknown:
-                raise exceptions.ReceivedUnknownContinuation(
+                self.raise_exception(exceptions.ReceivedUnknownContinuation(
-                    f"Received unknown continuation type:{unknown}")
+                    f"Received unknown continuation type:{unknown}"))
             else:
-                raise exceptions.FailedExtractContinuation('Cannot extract continuation data')
+                self.raise_exception(exceptions.FailedExtractContinuation('Cannot extract continuation data'))
         return self._create_data(metadata, contents)
 
     def reload_continuation(self, contents):
@@ -72,7 +83,7 @@ class Parser:
         """
         if contents is None:
             '''Broadcasting end or cannot fetch chat stream'''
-            raise exceptions.NoContents('Chat data stream is empty.')
+            self.raise_exception(exceptions.NoContents('Chat data stream is empty.'))
         cont = contents['liveChatContinuation']['continuations'][0]
         if cont.get("liveChatReplayContinuationData"):
             # chat data exist.
@@ -81,7 +92,7 @@ class Parser:
         init_cont = cont.get("playerSeekContinuationData")
         if init_cont:
             return init_cont.get("continuation")
-        raise exceptions.ChatDataFinished('Finished chat data')
+        self.raise_exception(exceptions.ChatDataFinished('Finished chat data'))
 
     def _create_data(self, metadata, contents):
         actions = contents['liveChatContinuation'].get('actions')
@@ -103,3 +114,8 @@ class Parser:
         start = int(actions[0]["replayChatItemAction"]["videoOffsetTimeMsec"])
         last = int(actions[-1]["replayChatItemAction"]["videoOffsetTimeMsec"])
         return (last - start)
+
+    def raise_exception(self, exception):
+        if self.exception_holder is None:
+            raise exception
+        self.exception_holder = exception

pytchat/processors/combinator.py

@@ -36,3 +36,7 @@ class Combinator(ChatProcessor):
         '''
         return tuple(processor.process(chat_components)
                      for processor in self.processors)
+
+    def finalize(self, *args, **kwargs):
+        [processor.finalize(*args, **kwargs)
+                     for processor in self.processors]

pytchat/processors/default/custom_encoder.py

@@ -0,0 +1,11 @@
+import json
+from .renderer.base import Author
+from .renderer.paidmessage import Colors
+from .renderer.paidsticker import Colors2
+
+
+class CustomEncoder(json.JSONEncoder):
+    def default(self, obj):
+        if isinstance(obj, Author) or isinstance(obj, Colors) or isinstance(obj, Colors2):
+            return vars(obj)
+        return json.JSONEncoder.default(self, obj)

pytchat/processors/default/processor.py

@@ -1,5 +1,7 @@
 import asyncio
+import json
 import time
+from .custom_encoder import CustomEncoder
 from .renderer.textmessage import LiveChatTextMessageRenderer
 from .renderer.paidmessage import LiveChatPaidMessageRenderer
 from .renderer.paidsticker import LiveChatPaidStickerRenderer
@@ -11,25 +13,120 @@ from ... import config
 logger = config.logger(__name__)
 
 
+class Chat:
+    def json(self) -> str:
+        return json.dumps(vars(self), ensure_ascii=False, cls=CustomEncoder)
+
+
 class Chatdata:
-    def __init__(self, chatlist: list, timeout: float):
+
+    def __init__(self, chatlist: list, timeout: float, abs_diff):
         self.items = chatlist
         self.interval = timeout
+        self.abs_diff = abs_diff
+        self.itemcount = 0
 
     def tick(self):
-        if self.interval == 0:
+        '''DEPRECATE
+            Use sync_items()
+        '''
+        if len(self.items) < 1:
             time.sleep(1)
             return
-        time.sleep(self.interval / len(self.items))
+        if self.itemcount == 0:
+            self.starttime = time.time()
+        if len(self.items) == 1:
+            total_itemcount = 1
+        else:
+            total_itemcount = len(self.items) - 1
+        next_chattime = (self.items[0].timestamp + (self.items[-1].timestamp - self.items[0].timestamp) / total_itemcount * self.itemcount) / 1000
+        tobe_disptime = self.abs_diff + next_chattime
+        wait_sec = tobe_disptime - time.time()
+        self.itemcount += 1
+        
+        if wait_sec < 0:
+            wait_sec = 0
+       
+        time.sleep(wait_sec)
 
     async def tick_async(self):
-        if self.interval == 0:
+        '''DEPRECATE
+            Use async_items()
+        '''
+        if len(self.items) < 1:
             await asyncio.sleep(1)
             return
-        await asyncio.sleep(self.interval / len(self.items))
+        if self.itemcount == 0:
+            self.starttime = time.time()
+        if len(self.items) == 1:
+            total_itemcount = 1
+        else:
+            total_itemcount = len(self.items) - 1
+        next_chattime = (self.items[0].timestamp + (self.items[-1].timestamp - self.items[0].timestamp) / total_itemcount * self.itemcount) / 1000
+        tobe_disptime = self.abs_diff + next_chattime
+        wait_sec = tobe_disptime - time.time()
+        self.itemcount += 1
+        
+        if wait_sec < 0:
+            wait_sec = 0
+       
+        await asyncio.sleep(wait_sec)
+
+    def sync_items(self):
+        starttime = time.time()
+        if len(self.items) > 0:
+            last_chattime = self.items[-1].timestamp / 1000
+            tobe_disptime = self.abs_diff + last_chattime
+            wait_total_sec = max(tobe_disptime - time.time(), 0)
+            if len(self.items) > 1:
+                wait_sec = wait_total_sec / len(self.items)
+            elif len(self.items) == 1:
+                wait_sec = 0
+            for c in self.items:
+                if wait_sec < 0:
+                    wait_sec = 0
+                time.sleep(wait_sec)
+                yield c
+        stop_interval = time.time() - starttime
+        if stop_interval < 1:
+            time.sleep(1 - stop_interval)
+
+    async def async_items(self):
+        starttime = time.time()
+        if len(self.items) > 0:
+            last_chattime = self.items[-1].timestamp / 1000
+            tobe_disptime = self.abs_diff + last_chattime
+            wait_total_sec = max(tobe_disptime - time.time(), 0)
+            if len(self.items) > 1:
+                wait_sec = wait_total_sec / len(self.items)
+            elif len(self.items) == 1:
+                wait_sec = 0
+            for c in self.items:
+                if wait_sec < 0:
+                    wait_sec = 0
+                await asyncio.sleep(wait_sec)
+                yield c
+                
+        stop_interval = time.time() - starttime
+        if stop_interval < 1:
+            await asyncio.sleep(1 - stop_interval)
+
+    def json(self) -> str:
+        return json.dumps([vars(a) for a in self.items], ensure_ascii=False, cls=CustomEncoder)
 
 
 class DefaultProcessor(ChatProcessor):
+    def __init__(self):
+        self.first = True
+        self.abs_diff = 0
+        self.renderers = {
+            "liveChatTextMessageRenderer": LiveChatTextMessageRenderer(),
+            "liveChatPaidMessageRenderer": LiveChatPaidMessageRenderer(),
+            "liveChatPaidStickerRenderer": LiveChatPaidStickerRenderer(),
+            "liveChatLegacyPaidMessageRenderer": LiveChatLegacyPaidMessageRenderer(),
+            "liveChatMembershipItemRenderer": LiveChatMembershipItemRenderer()
+        }
+
     def process(self, chat_components: list):
 
         chatlist = []
@@ -37,6 +134,8 @@ class DefaultProcessor(ChatProcessor):
 
         if chat_components:
             for component in chat_components:
+                if component is None:
+                    continue
                 timeout += component.get('timeout', 0)
                 chatdata = component.get('chatdata')
                 if chatdata is None:
@@ -46,43 +145,35 @@ class DefaultProcessor(ChatProcessor):
                         continue
                     if action.get('addChatItemAction') is None:
                         continue
-                    if action['addChatItemAction'].get('item') is None:
+                    item = action['addChatItemAction'].get('item')
+                    if item is None:
                         continue
-
+                    chat = self._parse(item)
-                    chat = self._parse(action)
                     if chat:
                         chatlist.append(chat)
-        return Chatdata(chatlist, float(timeout))
         
-    def _parse(self, sitem):
+        if self.first and chatlist:
-        action = sitem.get("addChatItemAction")
+            self.abs_diff = time.time() - chatlist[0].timestamp / 1000 + 2
-        if action:
+            self.first = False
-            item = action.get("item")
+
-        if item is None:
+        chatdata = Chatdata(chatlist, float(timeout), self.abs_diff)
-            return None
+
+        return chatdata
+
+    def _parse(self, item):
         try:
-            renderer = self._get_renderer(item)
+            key = list(item.keys())[0]
+            renderer = self.renderers.get(key)
             if renderer is None:
                 return None
-
+            renderer.setitem(item.get(key), Chat())
+            renderer.settype()
             renderer.get_snippet()
             renderer.get_authordetails()
+            rendered_chatobj = renderer.get_chatobj()
+            renderer.clear()
         except (KeyError, TypeError) as e:
-            logger.error(f"{str(type(e))}-{str(e)} sitem:{str(sitem)}")
+            logger.error(f"{str(type(e))}-{str(e)} item:{str(item)}")
             return None
-        return renderer
         
-    def _get_renderer(self, item):
+        return rendered_chatobj
-        if item.get("liveChatTextMessageRenderer"):
-            renderer = LiveChatTextMessageRenderer(item)
-        elif item.get("liveChatPaidMessageRenderer"):
-            renderer = LiveChatPaidMessageRenderer(item)
-        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/default/renderer/base.py

@@ -6,50 +6,51 @@ class Author:
 
 
 class BaseRenderer:
-    def __init__(self, item, chattype):
+    def setitem(self, item, chat):
-        self.renderer = list(item.values())[0]
+        self.item = item
-        self.chattype = chattype
+        self.chat = chat
-        self.author = Author()
+        self.chat.author = Author()
+
+    def settype(self):
+        pass
 
     def get_snippet(self):
-        self.type = self.chattype
+        self.chat.id = self.item.get('id')
-        self.id = self.renderer.get('id')
+        timestampUsec = int(self.item.get("timestampUsec", 0))
-        timestampUsec = int(self.renderer.get("timestampUsec", 0))
+        self.chat.timestamp = int(timestampUsec / 1000)
-        self.timestamp = int(timestampUsec / 1000)
+        tst = self.item.get("timestampText")
-        tst = self.renderer.get("timestampText")
         if tst:
-            self.elapsedTime = tst.get("simpleText")
+            self.chat.elapsedTime = tst.get("simpleText")
         else:
-            self.elapsedTime = ""
+            self.chat.elapsedTime = ""
-        self.datetime = self.get_datetime(timestampUsec)
+        self.chat.datetime = self.get_datetime(timestampUsec)
-        self.message, self.messageEx = self.get_message(self.renderer)
+        self.chat.message, self.chat.messageEx = self.get_message(self.item)
-        self.id = self.renderer.get('id')
+        self.chat.id = self.item.get('id')
-        self.amountValue = 0.0
+        self.chat.amountValue = 0.0
-        self.amountString = ""
+        self.chat.amountString = ""
-        self.currency = ""
+        self.chat.currency = ""
-        self.bgColor = 0
+        self.chat.bgColor = 0
 
     def get_authordetails(self):
-        self.author.badgeUrl = ""
+        self.chat.author.badgeUrl = ""
-        (self.author.isVerified,
+        (self.chat.author.isVerified,
-         self.author.isChatOwner,
+         self.chat.author.isChatOwner,
-         self.author.isChatSponsor,
+         self.chat.author.isChatSponsor,
-         self.author.isChatModerator) = (
+         self.chat.author.isChatModerator) = (
-            self.get_badges(self.renderer)
+            self.get_badges(self.item)
         )
-        self.author.channelId = self.renderer.get("authorExternalChannelId")
+        self.chat.author.channelId = self.item.get("authorExternalChannelId")
-        self.author.channelUrl = "http://www.youtube.com/channel/" + self.author.channelId
+        self.chat.author.channelUrl = "http://www.youtube.com/channel/" + self.chat.author.channelId
-        self.author.name = self.renderer["authorName"]["simpleText"]
+        self.chat.author.name = self.item["authorName"]["simpleText"]
-        self.author.imageUrl = self.renderer["authorPhoto"]["thumbnails"][1]["url"]
+        self.chat.author.imageUrl = self.item["authorPhoto"]["thumbnails"][1]["url"]
 
-    def get_message(self, renderer):
+    def get_message(self, item):
         message = ''
         message_ex = []
-        if renderer.get("message"):
+        runs = item.get("message", {}).get("runs", {})
-            runs = renderer["message"].get("runs")
-            if runs:
         for r in runs:
-                    if r:
+            if not hasattr(r, "get"):
+                continue
             if r.get('emoji'):
                 message += r['emoji'].get('shortcuts', [''])[0]
                 message_ex.append({
@@ -63,17 +64,16 @@ class BaseRenderer:
         return message, message_ex
 
     def get_badges(self, renderer):
-        self.author.type = ''
+        self.chat.author.type = ''
         isVerified = False
         isChatOwner = False
         isChatSponsor = False
         isChatModerator = False
-        badges = renderer.get("authorBadges")
+        badges = renderer.get("authorBadges", {})
-        if badges:
         for badge in badges:
             if badge["liveChatAuthorBadgeRenderer"].get("icon"):
                 author_type = badge["liveChatAuthorBadgeRenderer"]["icon"]["iconType"]
-                    self.author.type = author_type
+                self.chat.author.type = author_type
                 if author_type == 'VERIFIED':
                     isVerified = True
                 if author_type == 'OWNER':
@@ -82,13 +82,20 @@ class BaseRenderer:
                     isChatModerator = True
             if badge["liveChatAuthorBadgeRenderer"].get("customThumbnail"):
                 isChatSponsor = True
-                    self.author.type = 'MEMBER'
+                self.chat.author.type = 'MEMBER'
                 self.get_badgeurl(badge)
         return isVerified, isChatOwner, isChatSponsor, isChatModerator
 
     def get_badgeurl(self, badge):
-        self.author.badgeUrl = badge["liveChatAuthorBadgeRenderer"]["customThumbnail"]["thumbnails"][0]["url"]
+        self.chat.author.badgeUrl = badge["liveChatAuthorBadgeRenderer"]["customThumbnail"]["thumbnails"][0]["url"]
 
     def get_datetime(self, timestamp):
         dt = datetime.fromtimestamp(timestamp / 1000000)
         return dt.strftime('%Y-%m-%d %H:%M:%S')
+
+    def get_chatobj(self):
+        return self.chat
+
+    def clear(self):
+        self.item = None
+        self.chat = None

pytchat/processors/default/renderer/legacypaid.py

@@ -2,14 +2,14 @@ from .base import BaseRenderer
 
 
 class LiveChatLegacyPaidMessageRenderer(BaseRenderer):
-    def __init__(self, item):
+    def settype(self):
-        super().__init__(item, "newSponsor")
+        self.chat.type = "newSponsor"
 
     def get_authordetails(self):
         super().get_authordetails()
-        self.author.isChatSponsor = True
+        self.chat.author.isChatSponsor = True
 
-    def get_message(self, renderer):
+    def get_message(self, item):
-        message = (renderer["eventText"]["runs"][0]["text"]
+        message = (item["eventText"]["runs"][0]["text"]
-                   ) + ' / ' + (renderer["detailText"]["simpleText"])
+                   ) + ' / ' + (item["detailText"]["simpleText"])
         return message, [message]

pytchat/processors/default/renderer/membership.py

@@ -2,14 +2,17 @@ from .base import BaseRenderer
 
 
 class LiveChatMembershipItemRenderer(BaseRenderer):
-    def __init__(self, item):
+    def settype(self):
-        super().__init__(item, "newSponsor")
+        self.chat.type = "newSponsor"
 
     def get_authordetails(self):
         super().get_authordetails()
-        self.author.isChatSponsor = True
+        self.chat.author.isChatSponsor = True
 
-    def get_message(self, renderer):
+    def get_message(self, item):
+        try:
             message = ''.join([mes.get("text", "")
-                           for mes in renderer["headerSubtext"]["runs"]])
+                           for mes in item["headerSubtext"]["runs"]])
+        except KeyError:
+            return "Welcome New Member!", ["Welcome New Member!"]
         return message, [message]

pytchat/processors/default/renderer/paidmessage.py

@@ -9,23 +9,23 @@ class Colors:
 
 
 class LiveChatPaidMessageRenderer(BaseRenderer):
-    def __init__(self, item):
+    def settype(self):
-        super().__init__(item, "superChat")
+        self.chat.type = "superChat"
 
     def get_snippet(self):
         super().get_snippet()
         amountDisplayString, symbol, amount = (
-            self.get_amountdata(self.renderer)
+            self.get_amountdata(self.item)
         )
-        self.amountValue = amount
+        self.chat.amountValue = amount
-        self.amountString = amountDisplayString
+        self.chat.amountString = amountDisplayString
-        self.currency = currency.symbols[symbol]["fxtext"] if currency.symbols.get(
+        self.chat.currency = currency.symbols[symbol]["fxtext"] if currency.symbols.get(
             symbol) else symbol
-        self.bgColor = self.renderer.get("bodyBackgroundColor", 0)
+        self.chat.bgColor = self.item.get("bodyBackgroundColor", 0)
-        self.colors = self.get_colors()
+        self.chat.colors = self.get_colors()
 
-    def get_amountdata(self, renderer):
+    def get_amountdata(self, item):
-        amountDisplayString = renderer["purchaseAmountText"]["simpleText"]
+        amountDisplayString = item["purchaseAmountText"]["simpleText"]
         m = superchat_regex.search(amountDisplayString)
         if m:
             symbol = m.group(1)
@@ -36,11 +36,12 @@ class LiveChatPaidMessageRenderer(BaseRenderer):
         return amountDisplayString, symbol, amount
 
     def get_colors(self):
+        item = self.item
         colors = Colors()
-        colors.headerBackgroundColor = self.renderer.get("headerBackgroundColor", 0)
+        colors.headerBackgroundColor = item.get("headerBackgroundColor", 0)
-        colors.headerTextColor = self.renderer.get("headerTextColor", 0)
+        colors.headerTextColor = item.get("headerTextColor", 0)
-        colors.bodyBackgroundColor = self.renderer.get("bodyBackgroundColor", 0)
+        colors.bodyBackgroundColor = item.get("bodyBackgroundColor", 0)
-        colors.bodyTextColor = self.renderer.get("bodyTextColor", 0)
+        colors.bodyTextColor = item.get("bodyTextColor", 0)
-        colors.timestampColor = self.renderer.get("timestampColor", 0)
+        colors.timestampColor = item.get("timestampColor", 0)
-        colors.authorNameTextColor = self.renderer.get("authorNameTextColor", 0)
+        colors.authorNameTextColor = item.get("authorNameTextColor", 0)
         return colors

pytchat/processors/default/renderer/paidsticker.py

@@ -4,30 +4,30 @@ from .base import BaseRenderer
 superchat_regex = re.compile(r"^(\D*)(\d{1,3}(,\d{3})*(\.\d*)*\b)$")
 
 
-class Colors:
+class Colors2:
     pass
 
 
 class LiveChatPaidStickerRenderer(BaseRenderer):
-    def __init__(self, item):
+    def settype(self):
-        super().__init__(item, "superSticker")
+        self.chat.type = "superSticker"
 
     def get_snippet(self):
         super().get_snippet()
         amountDisplayString, symbol, amount = (
-            self.get_amountdata(self.renderer)
+            self.get_amountdata(self.item)
         )
-        self.amountValue = amount
+        self.chat.amountValue = amount
-        self.amountString = amountDisplayString
+        self.chat.amountString = amountDisplayString
-        self.currency = currency.symbols[symbol]["fxtext"] if currency.symbols.get(
+        self.chat.currency = currency.symbols[symbol]["fxtext"] if currency.symbols.get(
             symbol) else symbol
-        self.bgColor = self.renderer.get("backgroundColor", 0)
+        self.chat.bgColor = self.item.get("backgroundColor", 0)
-        self.sticker = "".join(("https:",
+        self.chat.sticker = "".join(("https:",
-            self.renderer["sticker"]["thumbnails"][0]["url"]))
+            self.item["sticker"]["thumbnails"][0]["url"]))
-        self.colors = self.get_colors()
+        self.chat.colors = self.get_colors()
 
-    def get_amountdata(self, renderer):
+    def get_amountdata(self, item):
-        amountDisplayString = renderer["purchaseAmountText"]["simpleText"]
+        amountDisplayString = item["purchaseAmountText"]["simpleText"]
         m = superchat_regex.search(amountDisplayString)
         if m:
             symbol = m.group(1)
@@ -38,9 +38,10 @@ class LiveChatPaidStickerRenderer(BaseRenderer):
         return amountDisplayString, symbol, amount
 
     def get_colors(self):
-        colors = Colors()
+        item = self.item
-        colors.moneyChipBackgroundColor = self.renderer.get("moneyChipBackgroundColor", 0)
+        colors = Colors2()
-        colors.moneyChipTextColor = self.renderer.get("moneyChipTextColor", 0)
+        colors.moneyChipBackgroundColor = item.get("moneyChipBackgroundColor", 0)
-        colors.backgroundColor = self.renderer.get("backgroundColor", 0)
+        colors.moneyChipTextColor = item.get("moneyChipTextColor", 0)
-        colors.authorNameTextColor = self.renderer.get("authorNameTextColor", 0)
+        colors.backgroundColor = item.get("backgroundColor", 0)
+        colors.authorNameTextColor = item.get("authorNameTextColor", 0)
         return colors

pytchat/processors/default/renderer/textmessage.py

@@ -2,5 +2,5 @@ from .base import BaseRenderer
 
 
 class LiveChatTextMessageRenderer(BaseRenderer):
-    def __init__(self, item):
+    def settype(self):
-        super().__init__(item, "textMessage")
+        self.chat.type = "textMessage"

pytchat/processors/html_archiver.py

@@ -3,7 +3,7 @@ import os
 import re
 import time
 from base64 import standard_b64encode
-from httpx import NetworkError, ReadTimeout
+from concurrent.futures import ThreadPoolExecutor
 from .chat_processor import ChatProcessor
 from .default.processor import DefaultProcessor
 from ..exceptions import UnknownConnectionError
@@ -48,12 +48,14 @@ class HTMLArchiver(ChatProcessor):
     '''
     def __init__(self, save_path, callback=None):
         super().__init__()
+        self.client = httpx.Client(http2=True)
         self.save_path = self._checkpath(save_path)
         self.processor = DefaultProcessor()
-        self.emoji_table = {}  # tuble for custom emojis. key: emoji_id, value: base64 encoded image binary.
+        self.emoji_table = {}  # dict for custom emojis. key: emoji_id, value: base64 encoded image binary.
         self.header = [HEADER_HTML]
         self.body = ['<body>\n', '<table class="css">\n', self._parse_table_header(fmt_headers)]
         self.callback = callback
+        self.executor = ThreadPoolExecutor(max_workers=10)
 
     def _checkpath(self, filepath):
         splitter = os.path.splitext(os.path.basename(filepath))
@@ -80,7 +82,7 @@ class HTMLArchiver(ChatProcessor):
             save_path : str :
                 Actual save path of file.
             total_lines : int :
-                count of total lines written to the file.
+                Count of total lines written to the file.
         """
         if chat_components is None or len(chat_components) == 0:
             return
@@ -118,10 +120,9 @@ class HTMLArchiver(ChatProcessor):
         err = None
         for _ in range(5):
             try:
-                resp = httpx.get(url, timeout=30)
+                resp = self.client.get(url, timeout=30)
                 break
-            except (NetworkError, ReadTimeout) as e:
+            except httpx.HTTPError as e:
-                print("Network Error. retrying...")
                 err = e
                 time.sleep(3)
         else:
@@ -130,9 +131,9 @@ class HTMLArchiver(ChatProcessor):
         return standard_b64encode(resp.content).decode()
 
     def _set_emoji_table(self, item: dict):
-        emoji_id = item['id']
+        emoji_id = ''.join(('Z', item['id'])) if 48 <= ord(item['id'][0]) <= 57 else item['id']
         if emoji_id not in self.emoji_table:
-            self.emoji_table.setdefault(emoji_id, self._encode_img(item['url']))
+            self.emoji_table.setdefault(emoji_id, self.executor.submit(self._encode_img, item['url']))
         return emoji_id
 
     def _stylecode(self, name, code, width, height):
@@ -143,11 +144,12 @@ class HTMLArchiver(ChatProcessor):
     def _create_styles(self):
         return '\n'.join(('<style type="text/css">',
                           TABLE_CSS,
-                          '\n'.join(self._stylecode(key, self.emoji_table[key], 24, 24)
+                          '\n'.join(self._stylecode(key, self.emoji_table[key].result(), 24, 24)
                                 for key in self.emoji_table.keys()),
                           '</style>\n'))
     
     def finalize(self):
+        self.executor.shutdown()
         self.header.extend([self._create_styles(), '</head>\n'])
         self.body.extend(['</table>\n</body>\n</html>'])
         with open(self.save_path, mode='a', encoding='utf-8') as f:

pytchat/tool/extract/asyncdl.py

@@ -1,5 +1,6 @@
-import httpx
 import asyncio
+import httpx
+import socket
 from . import parser
 from . block import Block
 from . worker import ExtractWorker
@@ -8,7 +9,6 @@ from ... import config
 from ... paramgen import arcparam
 from ... exceptions import UnknownConnectionError
 from concurrent.futures import CancelledError
-from httpx import NetworkError, ReadTimeout
 from json import JSONDecodeError
 from urllib.parse import quote
 
@@ -75,12 +75,12 @@ def ready_blocks(video_id, duration, div, callback):
                     next_continuation, actions = None, []
                     break
                 param_set.add(continuation)
-                resp = await session.get(url, headers=headers)
+                resp = await session.get(url, headers=headers, timeout=10)
                 next_continuation, actions = parser.parse(resp.json())
                 break
             except JSONDecodeError:
                 await asyncio.sleep(3)
-            except (NetworkError, ReadTimeout) as e:
+            except httpx.HTTPError as e:
                 err = e
                 await asyncio.sleep(3)
         else:
@@ -136,9 +136,12 @@ def fetch_patch(callback, blocks, video_id):
                 break
             except JSONDecodeError:
                 await asyncio.sleep(3)
-            except (NetworkError, ReadTimeout) as e:
+            except httpx.HTTPError as e:
                 err = e
                 await asyncio.sleep(3)
+            except socket.error as error:
+                print("socket error", error.errno)
+                await asyncio.sleep(3)
         else:
             cancel()
             raise UnknownConnectionError("Abort:" + str(err))
@@ -162,15 +165,10 @@ def fetch_patch(callback, blocks, video_id):
 
 
 async def _shutdown():
-    print("\nshutdown...")
     tasks = [t for t in asyncio.all_tasks()
              if t is not asyncio.current_task()]
     for task in tasks:
         task.cancel()
-        try:
-            await task
-        except asyncio.CancelledError:
-            pass
 
 
 def cancel():

pytchat/tool/extract/extractor.py

@@ -93,5 +93,4 @@ class Extractor:
         return ret
 
     def cancel(self):
-        print("cancel")
         asyncdl.cancel()

pytchat/tool/videoinfo.py

@@ -2,15 +2,14 @@ import httpx
 import json
 import re
 import time
-from httpx import ConnectError, NetworkError
 from .. import config
 from ..exceptions import InvalidVideoIdException, PatternUnmatchError, UnknownConnectionError
 from ..util.extract_video_id import extract_video_id
 
 
 headers = config.headers
-                         
 pattern = re.compile(r"['\"]PLAYER_CONFIG['\"]:\s*({.*})")
+pattern2 = re.compile(r"yt\.setConfig\((\{[\s\S]*?\})\);")
 
 item_channel_id = [
     "videoDetails",
@@ -32,6 +31,10 @@ item_response = [
     "embedded_player_response"
 ]
 
+item_response2 = [
+    "PLAYER_VARS",
+    "embedded_player_response"
+]
 item_author_image = [
     "videoDetails",
     "embeddedPlayerOverlayVideoDetailsRenderer",
@@ -83,26 +86,32 @@ class VideoInfo:
 
     def __init__(self, video_id):
         self.video_id = extract_video_id(video_id)
+        self.client = httpx.Client(http2=True)
+        self.new_pattern_text = False
+        err = None
         for _ in range(3):
             try:
                 text = self._get_page_text(self.video_id)
                 self._parse(text)
                 break
-            except PatternUnmatchError:
+            except (InvalidVideoIdException, UnknownConnectionError) as e:
+                raise e
+            except Exception as e:
+                err = e
                 time.sleep(2)
                 pass
         else:
-            raise PatternUnmatchError("Pattern Unmatch")
+            raise err
 
     def _get_page_text(self, video_id):
         url = f"https://www.youtube.com/embed/{video_id}"
         err = None
         for _ in range(3):
             try:
-                resp = httpx.get(url, headers=headers)
+                resp = self.client.get(url, headers=headers)
                 resp.raise_for_status()
                 break
-            except (ConnectError, NetworkError) as e:
+            except httpx.HTTPError as e:
                 err = e
                 time.sleep(3)
         else:
@@ -113,11 +122,24 @@ class VideoInfo:
     def _parse(self, text):
         result = re.search(pattern, text)
         if result is None:
-            raise PatternUnmatchError()
+            result = re.search(pattern2, text)
+            if result is None:
+                raise PatternUnmatchError(doc=text)
+            else:
+                self.new_pattern_text = True
         decoder = json.JSONDecoder()
+        if self.new_pattern_text:
+            res = decoder.raw_decode(result.group(1))[0]
+        else:
             res = decoder.raw_decode(result.group(1)[:-1])[0]
+        if self.new_pattern_text:
+            response = self._get_item(res, item_response2)
+        else:
             response = self._get_item(res, item_response)
         if response is None:
+            if self.new_pattern_text:
+                self._check_video_is_private(res.get("PLAYER_VARS"))
+            else:
                 self._check_video_is_private(res.get("args"))
         self._renderer = self._get_item(json.loads(response), item_renderer)
         if self._renderer is None:

pytchat/util/extract_video_id.py

@@ -8,18 +8,21 @@ YT_VIDEO_ID_LENGTH = 11
 
 def extract_video_id(url_or_id: str) -> str:
     ret = ''
+    if '[' in url_or_id:
+        url_or_id = url_or_id.replace('[', '').replace(']', '')
+
     if type(url_or_id) != str:
         raise TypeError(f"{url_or_id}: URL or VideoID must be str, but {type(url_or_id)} is passed.")
     if len(url_or_id) == YT_VIDEO_ID_LENGTH:
         return url_or_id
     match = re.search(PATTERN, url_or_id)
     if match is None:
-        raise InvalidVideoIdException(url_or_id)
+        raise InvalidVideoIdException(f"Invalid video id: {url_or_id}")
     try:
         ret = match.group(4)
     except IndexError:
-        raise InvalidVideoIdException(url_or_id)
+        raise InvalidVideoIdException(f"Invalid video id: {url_or_id}")
 
     if ret is None or len(ret) != YT_VIDEO_ID_LENGTH:
-        raise InvalidVideoIdException(url_or_id)
+        raise InvalidVideoIdException(f"Invalid video id: {url_or_id}")
     return ret

requirements.txt

@@ -1,4 +1,4 @@
-httpx[http2]==0.14.1
+httpx[http2]==0.16.1
-protobuf==3.13.0
+protobuf==3.14.0
 pytz
 urllib3

requirements_test.txt

@@ -1,4 +1,2 @@
-mock
+pytest-mock==3.3.1
-mocker
+pytest-httpx==0.10.0
-pytest
-pytest_httpx

tests/test_default_processor.py

@@ -1,8 +1,17 @@
 import json
+from datetime import datetime
 from pytchat.parser.live import Parser
 from pytchat.processors.default.processor import DefaultProcessor
 
 
+TEST_TIMETSTAMP = 1570678496000000
+
+
+def get_local_datetime(timestamp):
+    dt = datetime.fromtimestamp(timestamp / 1000000)
+    return dt.strftime('%Y-%m-%d %H:%M:%S')
+
+
 def test_textmessage(mocker):
     '''text message'''
     processor = DefaultProcessor()
@@ -17,11 +26,10 @@ def test_textmessage(mocker):
     }
 
     ret = processor.process([data]).items[0]
-    assert ret.chattype == "textMessage"
     assert ret.id == "dummy_id"
     assert ret.message == "dummy_message"
     assert ret.timestamp == 1570678496000
-    assert ret.datetime == "2019-10-10 12:34:56"
+    assert ret.datetime == get_local_datetime(TEST_TIMETSTAMP)
     assert ret.author.name == "author_name"
     assert ret.author.channelId == "author_channel_id"
     assert ret.author.channelUrl == "http://www.youtube.com/channel/author_channel_id"
@@ -47,13 +55,12 @@ def test_textmessage_replay_member(mocker):
     }
     
     ret = processor.process([data]).items[0]
-    assert ret.chattype == "textMessage"
     assert ret.type == "textMessage"
     assert ret.id == "dummy_id"
     assert ret.message == "dummy_message"
     assert ret.messageEx == ["dummy_message"]
     assert ret.timestamp == 1570678496000
-    assert ret.datetime == "2019-10-10 12:34:56"
+    assert ret.datetime == get_local_datetime(TEST_TIMETSTAMP)
     assert ret.elapsedTime == "1:23:45"
     assert ret.author.name == "author_name"
     assert ret.author.channelId == "author_channel_id"
@@ -80,14 +87,12 @@ def test_superchat(mocker):
     }
     
     ret = processor.process([data]).items[0]
-    print(json.dumps(chatdata, ensure_ascii=False))
-    assert ret.chattype == "superChat"
     assert ret.type == "superChat"
     assert ret.id == "dummy_id"
     assert ret.message == "dummy_message"
     assert ret.messageEx == ["dummy_message"]
     assert ret.timestamp == 1570678496000
-    assert ret.datetime == "2019-10-10 12:34:56"
+    assert ret.datetime == get_local_datetime(TEST_TIMETSTAMP)
     assert ret.elapsedTime == ""
     assert ret.amountValue == 800
     assert ret.amountString == "￥800"
@@ -124,14 +129,12 @@ def test_supersticker(mocker):
     }
    
     ret = processor.process([data]).items[0]
-    print(json.dumps(chatdata, ensure_ascii=False))
-    assert ret.chattype == "superSticker"
     assert ret.type == "superSticker"
     assert ret.id == "dummy_id"
     assert ret.message == ""
     assert ret.messageEx == []
     assert ret.timestamp == 1570678496000
-    assert ret.datetime == "2019-10-10 12:34:56"
+    assert ret.datetime == get_local_datetime(TEST_TIMETSTAMP)
     assert ret.elapsedTime == ""
     assert ret.amountValue == 200
     assert ret.amountString == "￥200"
@@ -167,14 +170,12 @@ def test_sponsor(mocker):
     }
     
     ret = processor.process([data]).items[0]
-    print(json.dumps(chatdata, ensure_ascii=False))
-    assert ret.chattype == "newSponsor"
     assert ret.type == "newSponsor"
     assert ret.id == "dummy_id"
     assert ret.message == "新規メンバー"
     assert ret.messageEx == ["新規メンバー"]
     assert ret.timestamp == 1570678496000
-    assert ret.datetime == "2019-10-10 12:34:56"
+    assert ret.datetime == get_local_datetime(TEST_TIMETSTAMP)
     assert ret.elapsedTime == ""
     assert ret.bgColor == 0
     assert ret.author.name == "author_name"
@@ -202,14 +203,12 @@ def test_sponsor_legacy(mocker):
     }
     
     ret = processor.process([data]).items[0]
-    print(json.dumps(chatdata, ensure_ascii=False))
-    assert ret.chattype == "newSponsor"
     assert ret.type == "newSponsor"
     assert ret.id == "dummy_id"
     assert ret.message == "新規メンバー / ようこそ、author_name！"
     assert ret.messageEx == ["新規メンバー / ようこそ、author_name！"]
     assert ret.timestamp == 1570678496000
-    assert ret.datetime == "2019-10-10 12:34:56"
+    assert ret.datetime == get_local_datetime(TEST_TIMETSTAMP)
     assert ret.elapsedTime == ""
     assert ret.bgColor == 0
     assert ret.author.name == "author_name"

tests/test_videoinfo.py

@@ -1,6 +1,6 @@
 from json.decoder import JSONDecodeError
 from pytchat.tool.videoinfo import VideoInfo
-from pytchat.exceptions import InvalidVideoIdException, PatternUnmatchError
+from pytchat.exceptions import InvalidVideoIdException
 
 
 def _open_file(path):
@@ -13,7 +13,7 @@ def _set_test_data(filepath, mocker):
     response_mock = mocker.Mock()
     response_mock.status_code = 200
     response_mock.text = _text
-    mocker.patch('httpx.get').return_value = response_mock
+    mocker.patch('httpx.Client.get').return_value = response_mock
 
 
 def test_archived_page(mocker):
@@ -85,7 +85,7 @@ def test_pattern_unmatch(mocker):
     try:
         _ = VideoInfo('__test_id__')
         assert False
-    except PatternUnmatchError:
+    except JSONDecodeError:
         assert True