.xml文件批量修改

[python]代码库

#!/usr/bin/python
'''
此文件用于整理网络上搜集的数据集的xml文件，整理后方便用于训练。
'''
import os  # 文件操作相关
import xml.etree.ElementTree as ET  # xml文件操作相关
import cv2

# PRINT_FLAG=True
PRINT_FLAG = False


# 批量修改整个文件夹所有的xml文件
def change_all_xml(xmlfilepath, string1):
    total_xml = os.listdir(xmlfilepath)  # 用于返回指定的文件夹包含的文件或文件夹的名字的列表。

    num = len(total_xml)  # xml文件个数
    print(num)
    print(total_xml[0])
    for xmlfile in total_xml:
        # print("**********************************************************************************************************")
        # print(xmlfile)
        in_file = open('%s/%s' % (xmlfilepath, xmlfile), encoding='UTF-8')
        # print(in_file)

        tree = ET.parse(in_file)
        # print(tree)
        root = tree.getroot()
        # print(root)
        obj = root.find('path')  # 找到filename标签，
        # print(obj)
        path_text = obj.text
        # print(path_text)
        # end = "."
        end = "."
        string3 = path_text[path_text.rfind(end):]  # 在strint1中查找最后一个反斜杠\后面的字符,图片名称
        # print("string3:", string3)
        end = "."
        # string4 = string3[string3.rfind(end):]  # 在strint1中查找最后一个正斜杠/后面的字符,图片名称
        # print("string4:", string4)

        # print("xmlfile:", xmlfile)
        # print("string1:", string1)
        end = "."
        string2 = xmlfile[:xmlfile.rfind(end)]  # 在strint1中查找最后一个正斜杠/后面的字符,图片名称
        # print("string2:", string2)
        path_text_1 = string1 + string2 + string3
        try:
            # print("path_text_1:", path_text_1)
            image = cv2.imread(path_text_1, 1)
            image.shape
            # cv2.imshow("aa",image)
            # cv2.waitKey(2000)

            # print('Open image ok! ')
        except:
            print('1111:Open image Error! Try again!')
            print("path_text_1:", path_text_1)
            string3 = '.jpg'
            path_text_1 = string1 + string2 + string3
            print("path_text_1:", path_text_1)
            try:
                # print("path_text_1:", path_text_1)
                image = cv2.imread(path_text_1, 1)
                image.shape
                # cv2.imshow("aa",image)
                # cv2.waitKey(2000)
                print('Open image ok! ')
                print('*****************************************************************************')
            except:
                print('222 : Open image Error! Try again!')
                string3 = ".jpg"
                path_text_1 = string1 + string2 + string3
                print("path_text_1:", path_text_1)

        # print("path_text_1:",path_text_1)
        obj.text = path_text_1  # 修改标签内容
        # tree.write('%s/%s' % (xmlfilepath, xmlfile))  # 保存修改
        obj_2 = root.find('filename')  # 找到filename标签
        #
        path_text_2 = string2 + string3
        #
        obj_2.text = path_text_2  # 修改标签内容
        if PRINT_FLAG == True:
            print("string1:", string1)
            print("string2:", string2)
            print("string3:", string3)
            print("path_text_1:", path_text_1)
            print("obj_2:", obj_2)
            print("path_text_2:", path_text_2)
        tree.write('%s/%s' % (xmlfilepath, xmlfile))  # 保存修改


# xmlfilepath = 'temp'  # xml文件保存地址
xmlfilepath = 'helmet_xml'  # xml文件保存地址
# 要修改的内容
string1 = 'E:\\1_Training_picture\\6_helmet\\helmet_train\\'

change_all_xml(xmlfilepath, string1)
print("ok")