Prompt Engineering uses old version of openai lib causing an excpetion

[etoews]

Describe the bug
The Prompt Engineering: A Practical Example – Real Python tutorial uses an old version of openai lib causing an openai.lib._old_api.APIRemovedInV1 exception.

To Reproduce
Steps to reproduce the behavior:

% git clone git@github.com:realpython/materials.git

% cd materials/prompt-engineering

% python -m venv venv

% source venv/bin/activate
(venv) % python -m pip install openai
Collecting openai
  Downloading openai-1.5.0-py3-none-any.whl (223 kB)
...

% python app.py chats.txt
Traceback (most recent call last):
  File "/Users/etoews/dev/temp/prompt-engineering/app.py", line 99, in <module>
    main(parse_args())
  File "/Users/etoews/dev/temp/prompt-engineering/app.py", line 54, in main
    print(get_completion(file_content, settings))
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/etoews/dev/temp/prompt-engineering/app.py", line 59, in get_completion
    response = openai.Completion.create(
               ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/etoews/dev/temp/prompt-engineering/venv/lib/python3.11/site-packages/openai/lib/_old_api.py", line 39, in __call__
    raise APIRemovedInV1(symbol=self._symbol)
openai.lib._old_api.APIRemovedInV1:

You tried to access openai.Completion, but this is no longer supported in openai>=1.0.0 - see the README at https://github.com/openai/openai-python for the API.

You can run `openai migrate` to automatically upgrade your codebase to use the 1.0.0 interface.

Alternatively, you can pin your installation to the old version, e.g. `pip install openai==0.28`

A detailed migration guide is available here: https://github.com/openai/openai-python/discussions/742

FWIW, running openai migrate does work but I don't know enough about the lib to put this into a PR yet.

% venv/bin/openai migrate
Retrieving Grit CLI metadata from https://api.keygen.sh/v1/accounts/custodian-dev/artifacts/marzano-macos-arm64
./app.py
     import tomllib
     from pathlib import Path

    -import openai
    +from openai import OpenAI
    +
    +client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))

     # Authenticate
    -openai.api_key = os.getenv("OPENAI_API_KEY")


     class Settings(dict):

     def get_completion(content: str, settings: Settings) -> str:
         """Send a request to the /completions endpoint."""
    -    response = openai.Completion.create(
    -        model=settings.model,
    -        prompt=assemble_prompt(content, settings),
    -        max_tokens=settings.max_tokens,
    -        temperature=settings.temperature,
    -    )
    +    response = client.completions.create(model=settings.model,
    +    prompt=assemble_prompt(content, settings),
    +    max_tokens=settings.max_tokens,
    +    temperature=settings.temperature)
         return response["choices"][0]["text"]


     def get_chat_completion(content: str, settings: Settings) -> str:
         """Send a request to the /chat/completions endpoint."""
    -    response = openai.ChatCompletion.create(
    -        model=settings.model,
    -        messages=assemble_chat_messages(content, settings),
    -        temperature=settings.temperature,
    -    )
    +    response = client.chat.completions.create(model=settings.model,
    +    messages=assemble_chat_messages(content, settings),
    +    temperature=settings.temperature)
         return response["choices"][0]["message"]["content"]




Log in ./venv/lib/python3.11/site-packages/openai/cli/_cli.py: TODO: The 'openai.http_client' option isn't read in the client API. You will need to pass it when you instantiate the client, e.g. 'OpenAI(http_client=http_client)'

  - Range: 183:5 - 183:37
  - Source:
  openai.http_client = http_client
  - Syntax tree:
  (assignment left: (attribute object: (identifier) attribute: (identifier)) right: (identifier))
./venv/lib/python3.11/site-packages/openai/cli/_cli.py
             proxies=proxies or None,
             http2=can_use_http2(),
         )
    -    openai.http_client = http_client
    +    # TODO: The 'openai.http_client' option isn't read in the client API. You will need to pass it when you instantiate the client, e.g. 'OpenAI(http_client=http_client)'
    +    # openai.http_client = http_client

         if args.organization:
    -        openai.organization = args.organization
    +        # TODO: The 'openai.organization' option isn't read in the client API. You will need to pass it when you instantiate the client, e.g. 'OpenAI(organization=args.organization)'
    +        # openai.organization = args.organization

         if args.api_key:
    -        openai.api_key = args.api_key

         if args.api_base:
    -        openai.base_url = args.api_base
    +        # TODO: The 'openai.base_url' option isn't read in the client API. You will need to pass it when you instantiate the client, e.g. 'OpenAI(base_url=args.api_base)'
    +        # openai.base_url = args.api_base

         # azure
         if args.api_type is not None:
    -        openai.api_type = args.api_type

         if args.azure_endpoint is not None:
    -        openai.azure_endpoint = args.azure_endpoint
    +        # TODO: The 'openai.azure_endpoint' option isn't read in the client API. You will need to pass it when you instantiate the client, e.g. 'OpenAI(azure_endpoint=args.azure_endpoint)'
    +        # openai.azure_endpoint = args.azure_endpoint

         if args.api_version is not None:
    -        openai.api_version = args.api_version

         if args.azure_ad_token is not None:
    -        openai.azure_ad_token = args.azure_ad_token
    +        # TODO: The 'openai.azure_ad_token' option isn't read in the client API. You will need to pass it when you instantiate the client, e.g. 'OpenAI(azure_ad_token=args.azure_ad_token)'
    +        # openai.azure_ad_token = args.azure_ad_token

         try:
             if args.args_model:


Processed 1167 files and found 2 matches

Expected behavior
The Prompt Engineering: A Practical Example – Real Python tutorial should work with openai>=1.0.0.

Additional context

% python --version
Python 3.11.3

[etoews]

Or change this part of the tutorial

To this

% python -m pip install -r requirements.txt
Collecting openai==0.27.8
  Downloading openai-0.27.8-py3-none-any.whl (73 kB)
...

[dbader]

Thanks -- we're looking into it! :)

[dbader]

Quick update, we've added instructions for pinning the openai module to the article. We've also slated it for a bigger refresh in the new year to take into account the new models and updated OpenAI API endpoints. I'm closing this issue for now. Thanks again for the report.

[etoews]

Thanks!