app.py

import gradio as gr
import requests
import base64
from pathlib import Path
import jwt
import time
import logging

# Set up logging for debugging
logging.basicConfig(level=logging.DEBUG)
logger = logging.getLogger(__name__)

# Kling AI API configuration
ACCESS_KEY_ID = "AFyHfnQATghFdCMyAG3gRPbNY4TNKFGB"
ACCESS_KEY_SECRET = "TTepeLyBterLNM3brYPGmdndBnnyKJBA"
API_URL = "https://api-singapore.klingai.com/v1/images/generations"

def generate_jwt_token():
    """Generate JWT token for Kling AI API authentication."""
    payload = {
        "iss": ACCESS_KEY_ID,
        "exp": int(time.time()) + 1800,
        "nbf": int(time.time()) - 5
    }
    token = jwt.encode(payload, ACCESS_KEY_SECRET, algorithm="HS256")
    return token if isinstance(token, str) else token.decode("utf-8")

def generate_image(image, prompt=""):
    """Call Kling AI API for image-to-image generation."""
    if not image:
        return None, "Error: Please upload a valid image."

    try:
        with open(image, "rb") as img_file:
            image_base64 = base64.b64encode(img_file.read()).decode("utf-8")
    except Exception as e:
        return None, f"Error: Failed to process image. Details: {str(e)}"

    headers = {
        "Authorization": f"Bearer {generate_jwt_token()}",
        "Content-Type": "application/json"
    }

    payload = {
        "model": "kolors-v2.1",
        "prompt": prompt or "Transform the face into a cartoon style while preserving identity",
        "image_base64_list": [image_base64],
        "strength": 0.97,
        "output_format": "png",
        "num_outputs": 1,
        "aspect_ratio": "1:1"
    }

    try:
        logger.debug(f"Sending POST to {API_URL} with payload")
        response = requests.post(API_URL, json=payload, headers=headers, timeout=30)
        response.raise_for_status()
        data = response.json()
        logger.debug(f"API response: {data}")

        task_id = data.get("data", {}).get("task_id")
        if not task_id:
            return None, "Error: No task ID returned."

        # Poll task until completion
        status_url = f"https://api-singapore.klingai.com/v1/images/task/{task_id}"
        for _ in range(60):  # up to 5 minutes
            status_response = requests.get(status_url, headers=headers, timeout=30)
            status_response.raise_for_status()
            status_data = status_response.json()
            logger.debug(f"Status: {status_data}")
            task_status = status_data.get("data", {}).get("task_status")
            if task_status == "succeed":
                image_url = status_data["data"]["images"][0]["url"]
                img_resp = requests.get(image_url, timeout=30)
                output_path = Path("/tmp/output_image.png")
                with open(output_path, "wb") as f:
                    f.write(img_resp.content)
                return str(output_path), None
            elif task_status in ("failed", "canceled"):
                return None, f"Error: Task failed. Details: {status_data}"
            time.sleep(5)
        return None, "Error: Task polling timed out."

    except requests.exceptions.RequestException as e:
        return None, f"Error: API request failed. Details: {str(e)}"

def chatbot_interface(image, prompt):
    output_path, error = generate_image(image, prompt)
    if error:
        return None, None, error
    return output_path, output_path, None

# Define Gradio interface
with gr.Blocks() as iface:
    gr.Markdown("# Kling AI Image-to-Image Generator")
    with gr.Row():
        with gr.Column():
            image_input = gr.Image(type="filepath", label="Upload Image")
            prompt_input = gr.Textbox(lines=2, placeholder="Enter prompt", label="Prompt")
            generate_button = gr.Button("Generate")
        with gr.Column():
            output_image = gr.Image(label="Generated Image")
            output_file = gr.File(label="Download Generated Image")
            error_message = gr.Textbox(label="Status/Error Message", interactive=False)

    generate_button.click(
        fn=chatbot_interface,
        inputs=[image_input, prompt_input],
        outputs=[output_image, output_file, error_message]
    )

if __name__ == "__main__":
    iface.launch(server_name="0.0.0.0", server_port=7860)