#!/usr/bin/env python

from __future__ import annotations
import argparse
import functools
import os
import pathlib
import sys
from typing import Callable


import gradio as gr
import huggingface_hub
import numpy as np
import PIL.Image

from io import BytesIO
from fastai.vision import *
from fastai.vision import load_learner

from core import FeatureLoss
import torchvision.transforms as T

ORIGINAL_REPO_URL = 'https://github.com/vijishmadhavan/ArtLine'
TITLE = 'vijishmadhavan/ArtLine'
DESCRIPTION = f"""This is a demo for {ORIGINAL_REPO_URL}.

"""
ARTICLE = """

"""


MODEL_REPO = 'hylee/artline_model'

def parse_args() -> argparse.Namespace:
    parser = argparse.ArgumentParser()
    parser.add_argument('--device', type=str, default='cpu')
    parser.add_argument('--theme', type=str)
    parser.add_argument('--live', action='store_true')
    parser.add_argument('--share', action='store_true')
    parser.add_argument('--port', type=int)
    parser.add_argument('--disable-queue',
                        dest='enable_queue',
                        action='store_false')
    parser.add_argument('--allow-flagging', type=str, default='never')
    parser.add_argument('--allow-screenshot', action='store_true')
    return parser.parse_args()

def load_model():
    dir = 'model'
    name = 'ArtLine_650.pkl'
    model_path = huggingface_hub.hf_hub_download(MODEL_REPO,
                                                      name,
                                                      cache_dir=dir,
                                                      force_filename=name)
    return model_path



def run(
    image,
    learn,
) -> tuple[PIL.Image.Image]:

    img = PIL.Image.open(image.name)
    img_t = T.ToTensor()(img)
    img_fast = PIL.Image(img_t)

    p, img_hr, b = learn.predict(img_fast)
    r = PIL.Image(img_hr)

    return r

learn = None
def main():
    gr.close_all()

    args = parse_args()

    model_path = load_model()

    # singleton start
    def load_pkl(self) -> Any:
        global learn
        path = Path("model")
        learn = load_learner(path, 'ArtLine_650.pkl')

    PklLoader = type('PklLoader', (), {"load_pkl": load_pkl})
    pl = PklLoader()
    pl.load_pkl()


    func = functools.partial(run, learn=learn)
    func = functools.update_wrapper(func, run)

    
    gr.Interface(
        func,
        [
            gr.inputs.Image(type='file', label='Input Image'),
        ],
        [
            gr.outputs.Image(
                type='pil',
                label='Result'),
        ],
        #examples=examples,
        theme=args.theme,
        title=TITLE,
        description=DESCRIPTION,
        article=ARTICLE,
        allow_screenshot=args.allow_screenshot,
        allow_flagging=args.allow_flagging,
        live=args.live,
    ).launch(
        enable_queue=args.enable_queue,
        server_port=args.port,
        share=args.share,
    )


if __name__ == '__main__':
    main()