Upgrade to OpenAI Python SDK - Global Client

Convert OpenAI from openai version to the v1 version, while continuing to use the global client. This is a variant of the client-based version.

Apply with the Grit CLI

grit apply openai_global

Rewrite completions

BEFORE

import openai

completion = await openai.Completion.acreate(model="davinci-002", prompt="Hello world")
chat_completion = await openai.ChatCompletion.acreate(model="gpt-3.5-turbo", messages=[{"role": "user", "content": "Hello world"}])

AFTER

import openai

completion = await openai.completions.create(model="davinci-002", prompt="Hello world")
chat_completion = await openai.chat.completions.create(model="gpt-3.5-turbo", messages=[{"role": "user", "content": "Hello world"}])

Global settings

If you use the global client, options can be set on itself.

BEFORE

import openai

if openai_proxy:
    openai.proxy = openai_proxy
    openai.api_base = self.openai_api_base

AFTER

import openai

if openai_proxy:
    openai.proxy = openai_proxy
    openai.api_base = self.openai_api_base

Remap errors

BEFORE

import openai

try:
    completion = openai.Completion.create(model="davinci-002", prompt="Hello world")
    chat_completion = openai.ChatCompletion.create(model="gpt-3.5-turbo", messages=[{"role": "user", "content": "Hello world"}])
except openai.error.RateLimitError as err:
    pass

AFTER

import openai

try:
    completion = openai.completions.create(model="davinci-002", prompt="Hello world")
    chat_completion = openai.chat.completions.create(model="gpt-3.5-turbo", messages=[{"role": "user", "content": "Hello world"}])
except openai.RateLimitError as err:
    pass