import gradio as gr
import requests
import io
from PIL import Image
import json
import os
import logging
import math 
from tqdm import tqdm
import time

#logging.basicConfig(level=logging.DEBUG)

with open('loras.json', 'r') as f:
    loras = json.load(f)

def update_selection(selected_state: gr.SelectData):
    logging.debug(f"Inside update_selection, selected_state: {selected_state}")
    selected_lora_index = selected_state.index
    selected_lora = loras[selected_lora_index]
    new_placeholder = f"Type a prompt for {selected_lora['title']}"
    lora_repo = selected_lora["repo"]
    updated_text = f"### Selected: [{lora_repo}](https://huggingface.co/{lora_repo}) ✨"
    return (
        gr.update(placeholder=new_placeholder),
        updated_text,  # Retorna o texto Markdown atualizado
        selected_state
    )


def run_lora(prompt, selected_state, progress=gr.Progress(track_tqdm=True)):
    logging.debug(f"Inside run_lora, selected_state: {selected_state}")
    if not selected_state:
        logging.error("selected_state is None or empty.")
        raise gr.Error("You must select a LoRA before proceeding.")  # Popup error when no LoRA is selected

    selected_lora_index = selected_state.index  # Changed this line
    selected_lora = loras[selected_lora_index]
    api_url = f"https://api-inference.huggingface.co/models/{selected_lora['repo']}"
    trigger_word = selected_lora["trigger_word"]
    #token = os.getenv("API_TOKEN")
    payload = {
        "inputs": f"{prompt} {trigger_word}",
        "parameters":{"negative_prompt": "bad art, ugly, watermark, deformed", "num_inference_steps": 30, "scheduler":"DPMSolverMultistepScheduler"},
    }
    #headers = {"Authorization": f"Bearer {token}"}
    
    # Add a print statement to display the API request
    print(f"API Request: {api_url}")
    #print(f"API Headers: {headers}")
    print(f"API Payload: {payload}")

    error_count = 0
    pbar = tqdm(total=None, desc="Loading model")
    while(True):
        response = requests.post(api_url, json=payload)
        if response.status_code == 200:
            return Image.open(io.BytesIO(response.content))
        elif response.status_code == 503:
            #503 is triggered when the model is doing cold boot. It also gives you a time estimate from when the model is loaded but it is not super precise
            time.sleep(1)
            pbar.update(1)
        elif response.status_code == 500 and error_count < 5:
            print(response.content)
            time.sleep(1)
            error_count += 1
            continue
        else:
            logging.error(f"API Error: {response.status_code}")
            raise gr.Error("API Error: Unable to fetch the image.")  # Raise a Gradio error here



with gr.Blocks(css="custom.css") as app:
    title = gr.Markdown("# artificialguybr LoRA portfolio")
    description = gr.Markdown(
        "### This is my portfolio. Follow me on Twitter [@artificialguybr](https://twitter.com/artificialguybr). \n"
        "**Note**: The speed and generation quality are for demonstration purposes. "
        "For best quality, use Auto or Comfy or Diffusers. \n"
        "**Warning**: The API might take some time to deliver the image. \n"
        "Special thanks to Hugging Face for their free inference API."
    )
    selected_state = gr.State()
    with gr.Row():
        gallery = gr.Gallery(
            [(item["image"], item["title"]) for item in loras],
            label="LoRA Gallery",
            allow_preview=False,
            columns=3
        )
        with gr.Column():
            prompt_title = gr.Markdown("### Click on a LoRA in the gallery to select it")
            selected_info = gr.Markdown("")  # Novo componente Markdown para exibir o texto da LoRA selecionada
            with gr.Row():
                prompt = gr.Textbox(label="Prompt", show_label=False, lines=1, max_lines=1, placeholder="Type a prompt after selecting a LoRA")
                button = gr.Button("Run")
            result = gr.Image(interactive=False, label="Generated Image")

    gallery.select(
        update_selection,
        outputs=[prompt, selected_info, selected_state]  # Adicionado selected_info aqui
    )
    prompt.submit(
        fn=run_lora,
        inputs=[prompt, selected_state],
        outputs=[result]
    )
    button.click(
        fn=run_lora,
        inputs=[prompt, selected_state],
        outputs=[result]
    )

app.queue(max_size=20, concurrency_count=5)
app.launch()