🐛 Bug Report: Python Prompt-chatgpt template

[sonicviz]

👟 Reproduction steps

https://github.com/appwrite/templates/tree/main/python/prompt_chatgpt
Fails to run from the included static HTML test interface.
Always returns "Failed to query model."

👍 Expected behavior

It should work?

👎 Actual Behavior

Always returns "Failed to query model."
No errors as there was no logging.

So I debugged it according to the latest OpenAI library https://github.com/openai/openai-python/tree/main
It appears the OpenAI calls have changed.

Debugged and working version is below, with logging. Suggest including sample logging in any templates to make it easier to debug and also helps users not so familiar with python upskill faster.
File: src\main.py

import os
from openai import OpenAI
from .utils import get_static_file, throw_if_missing


def main(context):
    throw_if_missing(os.environ, ["OPENAI_API_KEY"])

    if context.req.method == "GET":
        return context.res.send(
            get_static_file("index.html"),
            200,
            {
                "content-type": "text/html; charset=utf-8"
            },
        )

    context.log(context.req.body)

    try:
        throw_if_missing(context.req.body, ["prompt"])
    except ValueError as err:
        return context.res.json({"ok": False, "error": err.message}, 400)

    # openai.api_key = os.environ["OPENAI_API_KEY"]

    client = OpenAI(
    # This is the default and can be omitted
    api_key=os.environ.get("OPENAI_API_KEY"),
    )       

    context.log("OPENAI_API_KEY" + os.environ["OPENAI_API_KEY"])
    context.log("OPENAI_MAX_TOKENS" + os.environ["OPENAI_MAX_TOKENS"])

    try:
        # response = openai.ChatCompletion.create(
        #     model="gpt-3.5-turbo",
        #     max_tokens=int(os.environ.get("OPENAI_MAX_TOKENS", "512")),
        #     messages=[{"role": "user", "content": context.req.body["prompt"]}],
        # )
        response = client.chat.completions.create(
            model="gpt-4o",
            max_tokens=int(os.environ.get("OPENAI_MAX_TOKENS", "512")),
            messages=[{"role": "user", "content": context.req.body["prompt"]}],
        )
        context.log(response)
        completion = response.choices[0].message.content
        return context.res.json({"ok": True, "completion": completion}, 200)

    except Exception:
        return context.res.json({"ok": False, "error": "Failed to query model."}, 500)

🎲 Appwrite version

Appwrite Cloud

💻 Operating system

Windows

🧱 Your Environment

No response

👀 Have you spent some time to check if this issue has been raised before?

• I checked and didn't find similar issue

🏢 Have you read the Code of Conduct?

• I have read the Code of Conduct

[loks0n]

Thanks for the report! 🙏

Feel free to pick this up and open a PR if you are interested.

[ChiragAgg5k]

since @sonicviz isnt working on this,
i can work on this issue, can it be assigned to me?

[sonicviz]

Yeah, sorry, too busy on other projects atm.
All the details are there though. Suerte!