import gradio as gr
import numpy as np
import spaces
import torch
import random
from PIL import Image
from diffusers import FluxKontextPipeline
from diffusers.utils import load_image
import requests

MAX_SEED = np.iinfo(np.int32).max

pipe = FluxKontextPipeline.from_pretrained("black-forest-labs/FLUX.1-Kontext-dev", torch_dtype=torch.bfloat16).to("cuda")

@spaces.GPU
def translate_albanian_to_english(text):
    """Translate Albanian to English using sepioo-facebook-translation API."""
    if not text.strip():
        return ""
    for attempt in range(2):
        try:
            response = requests.post(
                "https://hal1993-mdftranslation1234567890abcdef1234567890-fc073a6.hf.space/v1/translate",
                json={"from_language": "sq", "to_language": "en", "input_text": text},
                headers={"accept": "application/json", "Content-Type": "application/json"},
                timeout=5
            )
            response.raise_for_status()
            translated = response.json().get("translate", "")
            return translated
        except Exception as e:
            if attempt == 1:
                raise gr.Error(f"Përkthimi dështoi: {str(e)}")
    raise gr.Error("Përkthimi dështoi. Ju lutem provoni përsëri.")

@spaces.GPU
def infer(input_image, prompt, seed=42, randomize_seed=False, guidance_scale=2.5, steps=12, progress=gr.Progress(track_tqdm=True)):
    """
    Perform image editing using the FLUX.1 Kontext pipeline.
    """
    # Translate Albanian prompt to English
    final_prompt = translate_albanian_to_english(prompt.strip()) if prompt.strip() else ""
    if not final_prompt:
        return None, seed, gr.Button(visible=False)

    if randomize_seed:
        seed = random.randint(0, MAX_SEED)
    
    if input_image:
        input_image = input_image.convert("RGB")
        image = pipe(
            image=input_image, 
            prompt=final_prompt,
            guidance_scale=guidance_scale,
            width=input_image.size[0],
            height=input_image.size[1],
            num_inference_steps=steps,
            generator=torch.Generator().manual_seed(seed),
        ).images[0]
    else:
        image = pipe(
            prompt=final_prompt,
            guidance_scale=guidance_scale,
            width=1024,
            height=1024,
            num_inference_steps=steps,
            generator=torch.Generator().manual_seed(seed),
        ).images[0]
    return image, seed, gr.Button(visible=True)

@spaces.GPU
def infer_example(input_image, prompt):
    image, seed, _ = infer(input_image, prompt)
    return image, seed

with gr.Blocks() as demo:
    gr.HTML("""
    <style>
    body::before {
        content: "";
        display: block;
        height: 320px;
        background-color: var(--body-background-fill);
    }
    button[aria-label="Fullscreen"], button[aria-label="Fullscreen"]:hover {
        display: none !important;
        visibility: hidden !important;
        opacity: 0 !important;
        pointer-events: none !important;
    }
    button[aria-label="Share"], button[aria-label="Share"]:hover {
        display: none !important;
    }
    button[aria-label="Download"] {
        transform: scale(3);
        transform-origin: top right;
        margin: 0 !important;
        padding: 6px !important;
    }
    </style>
    """)

    gr.Markdown("# Modifiko imazhet")
    gr.Markdown("Modifiko imazhet ne menyre universale ne baze te pershkrimit")

    with gr.Column():
        input_image = gr.Image(label="Ngarko Imazhin për Editim", type="pil")
        prompt = gr.Textbox(
            label="Përshkrimi",
            placeholder="Shkruani përshkrimin këtu",
            lines=3
        )
        run_button = gr.Button(value="Gjenero")
        reuse_button = gr.Button("Rivendos këtë imazh", visible=False)
        # Hidden advanced settings
        seed = gr.Slider(
            minimum=0,
            maximum=MAX_SEED,
            step=1,
            value=0,
            visible=False
        )
        randomize_seed = gr.Checkbox(value=True, visible=False)
        guidance_scale = gr.Slider(
            minimum=1,
            maximum=10,
            step=0.1,
            value=2.5,
            visible=False
        )
        steps = gr.Slider(
            minimum=1,
            maximum=30,
            value=12,
            step=1,
            visible=False
        )

    with gr.Row():
        result = gr.Image(label="Imazhi i Gjeneruar", interactive=False)
    with gr.Row():
        reuse_button = gr.Button("Përdor imazhin e gjeneruar", visible=False)         

    gr.on(
        triggers=[run_button.click, prompt.submit],
        fn=infer,
        inputs=[input_image, prompt, seed, randomize_seed, guidance_scale, steps],
        outputs=[result, seed, reuse_button]
    )
    reuse_button.click(
        fn=lambda image: image,
        inputs=[result],
        outputs=[input_image]
    )

demo.launch(mcp_server=True)