import logging
import os.path

import cv2


def crop_picture(picture, width=720, height=1280):
    """
    根据屏幕分辨率裁剪图片，去除状态栏
    :param picture: 图片路径
    :param width: 屏幕宽度
    :param height: 屏幕高度
    """
    img = cv2.imread(picture)
    if img is None:
        return
    
    img_height, img_width = img.shape[:2]
    
    # 顶部状态栏、底部导航栏需要去除的高度
    status_bar_height = 90
    
    # 裁剪去除状态栏，保留完整屏幕内容
    x1, y1 = 0, status_bar_height
    x2, y2 = img_width, img_height
    
    img = img[y1:y2, x1:x2]
    cv2.imwrite(picture, img)


def compare_image_similarity(image1, image2):
    logging.info('{} is exist? [{}]'.format(image1, os.path.exists(image1)))
    logging.info('{} is exist? [{}]'.format(image2, os.path.exists(image2)))
    if not os.path.exists(image1) or not os.path.exists(image2):
        logging.info('file not found, set similarity as 0%')
        return 0
    image1 = cv2.imread(image1, 0)
    image2 = cv2.imread(image2, 0)

    # 初始化特征点检测器和描述符
    orb = cv2.ORB_create(edgeThreshold=5, patchSize=30)
    keypoints1, descriptors1 = orb.detectAndCompute(image1, None)
    keypoints2, descriptors2 = orb.detectAndCompute(image2, None)

    # 初始化匹配器
    bf = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)

    # 匹配特征点
    matches = bf.match(descriptors1, descriptors2)
    if not matches:
        logging.info('no fixture point found, set similarity as 0%')
        return 0
    if not keypoints1:
        if not keypoints2:
            logging.info('similarity is 100%')
            return 1
        logging.info('similarity is 0%')
        return 0
    # 计算相似度
    similarity = len(matches) / len(keypoints1)
    logging.info('similarity is {}%'.format(similarity * 100))
    return similarity