Closed 5ocworkshop closed 7 months ago
I don't know if this is new or not but I'm now seeing with V18 that I sometimes get this output in red in the console (I'm using VS Code native terminal which is using reduced colours): summarizer unexpectedly failed for gpt-3.5-turbo
I'm using the gpt-4 turbo 1106 api so not sure where the 3.5 reference comes from or is that an internal call?
We upgraded aider to the new openai client, so it's unlikely that this particular bug still happens.
I'm going to close this issue for now, but feel free to add a comment here and I will re-open or file a new issue any time.
Upgraded earlier today. Amazing project. Thank you so much.
Using the gpt-4 1106 preview. It was in the process of providing a search/replace snippet and:
Ensure that the column names are formatted and escaped properly
File "/home/jac/.local/bin/aider", line 8, in
sys.exit(main())
^^^^^^
File "/home/jac/.local/pipx/venvs/aider-chat/lib/python3.11/site-packages/aider/main.py", line 566, in main
coder.run()
File "/home/jac/.local/pipx/venvs/aider-chat/lib/python3.11/site-packages/aider/coders/base_coder.py", line 329, in run
new_user_message = self.run_loop()
^^^^^^^^^^^^^^^
File "/home/jac/.local/pipx/venvs/aider-chat/lib/python3.11/site-packages/aider/coders/base_coder.py", line 412, in run_loop
return self.send_new_user_message(inp)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/jac/.local/pipx/venvs/aider-chat/lib/python3.11/site-packages/aider/coders/base_coder.py", line 466, in send_new_user_message
interrupted = self.send(messages, functions=self.functions)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/jac/.local/pipx/venvs/aider-chat/lib/python3.11/site-packages/aider/coders/base_coder.py", line 590, in send
self.show_send_output_stream(completion)
File "/home/jac/.local/pipx/venvs/aider-chat/lib/python3.11/site-packages/aider/coders/base_coder.py", line 669, in show_send_output_stream
for chunk in completion:
File "/home/jac/.local/pipx/venvs/aider-chat/lib/python3.11/site-packages/openai/api_resources/abstract/engine_api_resource.py", line 168, in
return (
^
File "/home/jac/.local/pipx/venvs/aider-chat/lib/python3.11/site-packages/openai/api_requestor.py", line 703, in
self._interpret_response_line(
File "/home/jac/.local/pipx/venvs/aider-chat/lib/python3.11/site-packages/openai/api_requestor.py", line 775, in _interpret_response_line
raise self.handle_error_response(
openai.error.APIError: Request failed due to server shutdown (Error occurred while streaming.)
Seemed like something that could be handled gracefully and reported back to the UI?