llm-workflow-engine: New chat title generation POST request hanging

I encountered a problem when calling the chatgpt.ask() function. My question is successfully passed to the OpenAI website, and the dialogue content (i.e., the response from chatgpt to my question) is also generated correctly on the OpenAI platform. However, my local program is unable to retrieve the answer. During the ‘await response’ process, the following error occurs: playwright._impl._api_types.Error: Request context disposed. Can you help me understand and resolve this issue?

My test code:

from chatgpt_wrapper import ChatGPT
from chatgpt_wrapper.config import Config

config = Config()
config.set('browser.debug', True)
config.set('chat.model', 'legacy-paid')

bot = ChatGPT(config)
success, response, message = bot.ask("Can you tell me a joke?")
if success:
    print(response)
else:
    raise RuntimeError(message)
Traceback (most recent call last):
  File "c:\Users\Administrator\Desktop\chatgpt-test\test.py", line 9, in <module>
    success, response, message = bot.ask("Can you tell me a joke?")
  File "C:\ProgramData\Anaconda3\lib\site-packages\chatgpt_wrapper\chatgpt.py", line 518, in ask
    return self.async_run(self.agpt.ask(message, title=title))
  File "C:\ProgramData\Anaconda3\lib\site-packages\chatgpt_wrapper\chatgpt.py", line 494, in async_run        
    return asyncio.get_event_loop().run_until_complete(awaitable)
  File "C:\ProgramData\Anaconda3\lib\asyncio\base_events.py", line 647, in run_until_complete
    return future.result()
  File "C:\ProgramData\Anaconda3\lib\site-packages\chatgpt_wrapper\chatgpt.py", line 468, in ask
    response = list([i async for i in self.ask_stream(message, title=title)])
  File "C:\ProgramData\Anaconda3\lib\site-packages\chatgpt_wrapper\chatgpt.py", line 468, in <listcomp>       
    response = list([i async for i in self.ask_stream(message, title=title)])
  File "C:\ProgramData\Anaconda3\lib\site-packages\chatgpt_wrapper\chatgpt.py", line 444, in ask_stream       
    await self._gen_title()
  File "C:\ProgramData\Anaconda3\lib\site-packages\chatgpt_wrapper\chatgpt.py", line 204, in _gen_title       
    ok, json, response = await self._api_post_request(url, data)
  File "C:\ProgramData\Anaconda3\lib\site-packages\chatgpt_wrapper\chatgpt.py", line 188, in _api_post_request
    response = await self.page.request.post(url, headers=headers, data=data)
  File "C:\ProgramData\Anaconda3\lib\site-packages\playwright\async_api\_generated.py", line 18313, in post
    await self._impl_obj.post(
  File "C:\ProgramData\Anaconda3\lib\site-packages\playwright\_impl\_fetch.py", line 248, in post
    return await self.fetch(
  File "C:\ProgramData\Anaconda3\lib\site-packages\playwright\_impl\_fetch.py", line 285, in fetch
    return await self._inner_fetch(
  File "C:\ProgramData\Anaconda3\lib\site-packages\playwright\_impl\_fetch.py", line 372, in _inner_fetch
    response = await self._channel.send(
  File "C:\ProgramData\Anaconda3\lib\site-packages\playwright\_impl\_connection.py", line 44, in send
    return await self._connection.wrap_api_call(
  File "C:\ProgramData\Anaconda3\lib\site-packages\playwright\_impl\_connection.py", line 419, in wrap_api_call
    return await cb()
  File "C:\ProgramData\Anaconda3\lib\site-packages\playwright\_impl\_connection.py", line 79, in inner_send
    result = next(iter(done)).result()
playwright._impl._api_types.Error: Request context disposed.

The dialogue content (i.e., the response from chatgpt to my question) is also generated correctly on the OpenAI platform: image

About this issue

  • Original URL
  • State: closed
  • Created a year ago
  • Comments: 18

Most upvoted comments

@He2hiwei @cloudyskyy can you please try the latest release and see if that fixes the issue for you?

I’ve added a sensible timeout for the auto title generation, and put in better error handling – I think it will work fine in your cases.

后来解决了吗

The issue has not been resolved yet. If you have any solutions or other ideas, please feel free to share and contribute

Yes, kind of ‘solved’ … I found a strange phenomenon: When I rolled back the version to chatgpt 0.3.16, it suddenly stopped reporting this error, and the python API could return normal response of chatgpt. Once upgraded to chatgpt 0.3.17 or higher, the problem reappeared.