import os
import gradio as gr
import base64
import prompts
import json
from openai import OpenAI
from dotenv import load_dotenv

load_dotenv()
client = OpenAI()
PROMPT = prompts.SINGLE_QCM_PROMPT


# Function to encode the image
def encode_image(image_path):
    with open(image_path, "rb") as image_file:
        return base64.b64encode(image_file.read()).decode("utf-8")


def load_qcm(file_path):
    try:
        with open(file_path, "r", encoding="utf-8") as file:
            return json.load(file)
    except json.JSONDecodeError as e:
        print(f"Error decoding JSON: {e}")
        return {}


def get_answers(qcm):  # qcm is in json format
    answers = [answer["value"] for answer in qcm["Answers"]]
    correct_answers = [
        answer["value"] for answer in qcm["Answers"] if answer["correct"]
    ]
    md_answers = "\n".join([f"* {answer}" for answer in answers])
    md_correct_answers = "\n".join([f"* {answer}" for answer in correct_answers])
    return {"md_answers": md_answers, "md_correct_answers": md_correct_answers}


def process(image_path):
    try:
        response = client.chat.completions.create(
            model="gpt-4o",
            # response_format={ "type": "json_object" }, # si nécessaire
            messages=[
                # {"role": "system", "content": "You are a helpful assistant designed to output JSON."},
                {
                    "role": "user",
                    "content": [
                        {"type": "text", "text": PROMPT},
                        {
                            "type": "image_url",
                            "image_url": {
                                "url": f"data:image/jpeg;base64,{encode_image(image_path)}"
                            },
                        },
                    ],
                }
            ],
            temperature=0.2,
            #    max_tokens=256,
            #    top_p=1,
            #    frequency_penalty=0,
            #    presence_penalty=0
        )
        #        print(response["usage"]["total_tokens"])
        json_output = response.choices[0].message.content
        return json.loads(json_output)

    except Exception as e:
        print(f"an error occurred : {e}")
        return {"error": str(e)}, str(e)


with gr.Blocks() as demo:

    with gr.Row():
        image = gr.Image(label="Image", type="filepath")
        with gr.Column():
            submit_btn = gr.Button("Soumettre")
            progress = gr.Textbox(label="Traitement")
            with gr.Accordion(
                open=False,
            ):
                gr_json_output = gr.JSON(label="json output")

    with gr.Tab(label="QCM", visible=False) as gr_qcm_column:

        gr_question = gr.Textbox(label="Question")
        with gr.Accordion(label="Réponses possibles"):
            gr_answers = gr.Markdown()
        gr_hint = gr.Textbox(label="Aide à la réponse")
        with gr.Accordion(label="Bonnes réponses"):
            gr_correct_answers = gr.Markdown()
        gr_explanation = gr.Textbox(label="Explication")

    def submit(image_path):

        qcm = process(image_path)
        # qcm = load_qcm("questions.json")
        ga = get_answers(qcm)

        return {
            progress: "Terminé !",
            gr_qcm_column: gr.Tab(visible=True),
            gr_json_output: qcm,
            gr_question: qcm["Question"],
            gr_answers: ga["md_answers"],
            gr_hint: qcm["hint"],
            gr_correct_answers: ga["md_correct_answers"],
            gr_explanation: qcm["explanation"],
        }

    submit_btn.click(
        fn=submit,
        inputs=image,
        outputs=[
            progress,
            gr_qcm_column,
            gr_json_output,
            gr_question,
            gr_hint,
            gr_answers,
            gr_correct_answers,
            gr_explanation,
        ],
        api_name="submit",
    )

if __name__ == "__main__":
    authorized_users = [("test", os.environ["TEST_PASSWORD"])]
    demo.launch(auth=authorized_users)