import time
import subprocess
import os
import argparse
import cv2
import sys
from PIL import Image
import torch
import gradio as gr
import urllib.parse

TESTdevice = "cpu"
index = 1

def mainTest(inputpath, outpath):
    watermark = deep_nude_process(inputpath)
    watermark1 = cv2.cvtColor(watermark, cv2.COLOR_BGRA2RGBA)
    return watermark1

def deep_nude_process(inputpath):
    dress = cv2.imread(inputpath)
    h = dress.shape[0]
    w = dress.shape[1]
    dress = cv2.resize(dress, (512, 512), interpolation=cv2.INTER_CUBIC)
    watermark = process(dress)
    watermark = cv2.resize(watermark, (w, h), interpolation=cv2.INTER_CUBIC)
    return watermark

def inference(img):
    global index
    bgra = cv2.cvtColor(img, cv2.COLOR_RGBA2BGRA)
    inputpath = f"input_{index}.jpg"
    cv2.imwrite(inputpath, bgra)
    outputpath = f"out_{index}.jpg"
    index += 1
    print(time.strftime("START!!!!!!!!! %Y-%m-%d %H:%M:%S", time.localtime()))
    output = mainTest(inputpath, outputpath)
    print(time.strftime("Finish!!!!!!!!! %Y-%m-%d %H:%M:%S", time.localtime()))
    return output

def parse_url_params():
    url = os.environ.get('GRADIO_SERVER_URL', '')
    parsed_url = urllib.parse.urlparse(url)
    params = urllib.parse.parse_qs(parsed_url.query)
    
    width = params.get('width', ['auto'])[0]
    height = params.get('height', ['100%'])[0]
    bg_color = params.get('bg_color', ['rgb(17, 24, 39)'])[0]
    
    return width, height, bg_color

title = "Undress AI"
description = "⛔ Input photos of people, similar to the test picture at the bottom, and undress pictures will be produced. You may have to wait 30 seconds for a picture. 🔞 Do not upload personal photos 🔞 There is a queue system. According to the logic of first come, first served, only one picture will be made at a time. Must be able to at least see the outline of a human body ⛔"

examples = [
    ['input.png', 'Test'],
    ['input.jpg', 'Test'],
]

css = """
body {{
    background-color: {bg_color};
    color: white;
    overflow: hidden;
}}
.gradio-container {{
    background-color: {bg_color} !important;
    border: none !important;
    width: {width} !important;
    height: {height} !important;
    max-width: 100%;
    max-height: 100%;
    overflow: hidden;
}}
footer {{
    display: none !important;
}}
"""

width, height, bg_color = parse_url_params()
formatted_css = css.format(width=width, height=height, bg_color=bg_color)

with gr.Blocks(css=formatted_css) as demo:
    with gr.Column():
        image_input = gr.Image(type="numpy", label="Upload Image", height=512, width=512)
        process_button = gr.Button("Process Image")

    def update_status(img):
        return inference(img), gr.update(value="Processing complete!")

    process_button.click(update_status, inputs=image_input, outputs=[image_input])

demo.launch()