import numpy as np
import cv2
import os
from .get_image_info import main

rate = 0.5

def predict(path, model):
    ''' predict
    Args: path - image path
    Returns: image path, image info
    '''
    global img_y
    file_name = os.path.split(path)[1].split('.')[0]
    
    x = path.replace('\\', '/')
    file_name = file_name
    print(x)
    print(file_name)
    img_y = model.predict(x)['label_map']
    img_y = img_y * 255
    img_y = img_y.astype(np.int)
    cv2.imwrite(f'./tmp/mask/{file_name}_mask.png', img_y,
                (cv2.IMWRITE_PNG_COMPRESSION, 0))
    last_process(file_name)
    image_info = main(file_name)

    return file_name + '.png', image_info

def data_in_one(inputdata):
    if not inputdata.any():
        return inputdata
    inputdata = (inputdata - inputdata.min()) / (inputdata.max() - inputdata.min())
    return inputdata

def pre_process(data_path):
    file_name = os.path.split(data_path)[1].split('.')[0]
    return data_path, file_name

def last_process(file_name):
    ''' last process'''
    image = cv2.imread(f'./tmp/ct/{file_name}.png')
    mask = cv2.imread(f'./tmp/mask/{file_name}_mask.png', 0)
    contours, hierarchy = cv2.findContours(mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
    cv2.drawContours(image, contours, -1, (0, 255, 0), 2)
    cv2.imwrite('./tmp/draw/{}.png'.format(file_name), image)