import json
import requests
import streamlit as st
import pandas as pd
import numpy as np
from urllib.request import urlretrieve ,urlopen
import os
import torch
from distutils.dir_util import copy_tree
import tempfile
import zipfile
import subprocess
import mmcv
from mmcv import Config
import sys
from shapely.geometry import Polygon 


import shutil
import ssl

import urllib.request
from pathlib import Path

import cv2
from PIL import Image
##########
@st.cache(allow_output_mutation=True)
def loading_resources():
  dl_url = os.environ['dl_url']
  if 'utils2.py' not in os.listdir('./'):
    urlretrieve(dl_url, 'appdata.zip')
    with zipfile.ZipFile("appdata.zip","r") as zip_ref:
      zip_ref.extractall("./")
loading_resources()

import utilss
from utils2 import *
if 'model_inference' not in sys.modules:
  from mmocr.apis import init_detector ,model_inference
else:
  model_inference = sys.modules['model_inference']
  init_detector = sys.modules['init_detector']

########## utils
def draw_conts(im,conts,fcolor,stroke):
  opacity = .3
  im =im.copy()
  # fcolor =fcolor if fcolor else get_random_color()
  frontmm = np.full_like(im,fcolor,np.uint8)
  mask = np.zeros_like(im)
  for cont in conts:
    cv2.drawContours(mask, [cont], -1, color=[int(opacity*255)]*3, thickness=cv2.FILLED)
  mask = cv2.cvtColor(mask, cv2.COLOR_BGR2GRAY)
  rgba = np.dstack((frontmm, mask))
  im = utilss.overlay_image_alpha(im,rgba)
  # drawline
  cv2.drawContours(im, conts, -1, color=fcolor, thickness=stroke)
  return im
def image_resize(image, width = None, height = None, inter = cv2.INTER_AREA):
    dim = None
    (h, w) = image.shape[:2]
    if width is None and height is None:
        return image

    if width is None:
        r = height / float(h)
        dim = (int(w * r), height)
    else:
        r = width / float(w)
        dim = (width, int(h * r))
    resized = cv2.resize(image, dim, interpolation = inter)
    return resized
def download_file(url, local_filename):
    with requests.get(url, stream=True) as r:
        with open(local_filename, 'wb') as f:
            shutil.copyfileobj(r.raw, f)

    return local_filename
def load_and_preprocess_img(img_path, bbox=None):
  img = Image.open(img_path).convert('RGB')
  img = np.array(img).astype(np.uint8)
  return img


# @st.cache()
def predictim(im , model):
  if 'model_inference' not in sys.modules:
    # if not model_inference:
    from mmocr.apis import model_inference
  result = model_inference(model,im )
  # result[]
  return result
  # return result
# minfer=None
# minit=None
@st.cache(allow_output_mutation=True)
def loading_model():
  # global minfer
  # global minit
  
  from mmdet.apis import set_random_seed
  set_random_seed(0, deterministic=False)
  cfg = Config.fromfile('./cfgsn.py')
  # checkpoint = "./200s.pth"
  checkpoint = "./100s.pth"
  model1 = init_detector(cfg, checkpoint, device="cpu")
  # if model1.cfg.data.test['type'] == 'ConcatDataset':
  #   model1.cfg.data.test.pipeline = model1.cfg.data.test['datasets'][0].pipeline
  return model1
model = loading_model()



def main():
  st.sidebar.info('Images are deleted instantly')
  form0 =st.sidebar.form("my_form0")

  f = form0.file_uploader("Upload an Image", type=['png', 'jpg', 'jpeg', 'tiff', 'gif'])
  btnpredicrupload = form0.form_submit_button("PREDICT")


  # st.sidebar.write(" ------ ")
  form =st.sidebar.form("my_form")
  photos = ['1.jpg','2.jpg','3.jpg','4.jpg','5.jpg','6.jpg',]
  option = form.selectbox('Or choose a sample image', photos)
  submitted = form.form_submit_button("PREDICT")

  st.sidebar.write(" ------ ")
  st.sidebar.info('Options')
  extra_postprocess = st.sidebar.checkbox('perform extra postprocessing',value=True)
  fcolor = st.sidebar.selectbox('boundary color', ['red','green','blue'],index =0)
  fstroke = st.sidebar.selectbox('boundary thickness', ['1','2','3'] ,index=0 )
  # do_thresh_box = st.sidebar.checkbox('adaptive threshold')

  if btnpredicrupload and f is not None:
    tfile = tempfile.NamedTemporaryFile(delete=True)
    tfile.write(f.read())
    imgpath=tfile.name
    run_app(imgpath ,fcolor,extra_post=extra_postprocess,fstroke=fstroke)
  if submitted:
    st.empty()
    directory ='./imgs/'
    pic = os.path.join(directory, option)
    imgpath=directory+option
    run_app(imgpath ,fcolor,extra_post=extra_postprocess,fstroke=fstroke)
  # inp = st.text_input('t2','')
  # if st.button('run'):
  #   subprocess.run(inp.split(' '))


# def run_app(imgpath,do_thresh,fcolor):
def run_app(imgpath,fcolor,**kwargs):
  d0 =dict(red=[255,0,0] ,green=[0,255,0] ,blue=[0,0,255]  )
  fcolor = d0[fcolor]
  # st.sidebar.write(imgpath)
  # r = np.random.randint(1e3,1e7)
  tfile = tempfile.NamedTemporaryFile(delete=True,suffix='.jpg')
  dst = tfile.name
  # tfile = open()
  img = load_and_preprocess_img(imgpath)
  # img = utilss.resize_with_pad(img,800,800)
  # if do_thresh:
  #   gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
  #   threshim = cv2.adaptiveThreshold(gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,cv2.THRESH_BINARY, 51, 20)
  #   img= cv2.cvtColor(threshim, cv2.COLOR_GRAY2RGB)
  
  cv2.imwrite(dst,cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
  # imgo = image_resize(img , width = 640)
  st.write('input:')
  st.image(img, caption = "Selected Input" , width =640)

  # if st.button('predict2'):
  
  # try:
  # except: print('errr--------------')
  ''
  ##################
  preds,img_metas,downsample_ratio =  model_inference(model,dst )
  ff= fixer(img,preds,img_metas,downsample_ratio)
  polys= ff.polysfin3
  if kwargs['extra_post']:
    try:
      polys =find_polys4(ff) 
    except:
      print('error polys4')
      polys= ff.polysfin3
  

  # with open('./0.json','r') as f:
  #   res = json.load(f)
  ################
  mm = draw_conts(img,[poly2cont(p) for p in polys],fcolor,stroke=int(kwargs['fstroke']))
  ## mm=image_resize(mm, width = 640)
  #######
  # mm=preds[1]*preds[0]
  # mm=cv2.cvtColor(mm, cv2.COLOR_GRAY2RGB)
  st.image(mm,caption='output image' , width = 640)
  # del polys
  del ff
  del mm 
  del preds
  del polys
  del img
  del img_metas
  # os.remove(dst)
  # x = predictim(img,model)
  # st.write(x)

main()