读取xml文件写到csv文件中

#xml数据提取，拼合为csv文件

import os
import csv
import xml.dom.minidom
import glob




def to_int(str_value):
    try:
        return int(str_value)
    except ValueError:
        try:
            float_value = float(str_value)
            return int(float_value)
        except ValueError:
            return None

def get_damage_code(damage_type):
    """
    根据损伤类型返回相应的代码。

    参数:
    damage_type (str): 损伤类型名称。

    返回:
    int: 对应的损伤代码，如果损伤类型未知则返回None。
    """
    if damage_type == "bolt":
        return 30
    elif damage_type == "bolt crack":
        return 31
    elif damage_type == "head_unknow_1":
        return 501
    elif damage_type == "head_unknow_2":
        return 502
    elif damage_type == "head_unknow_3":
        return 503
    elif damage_type == "joint":
        return 600
    elif damage_type == "joint_1":
        return 601
    elif damage_type == "joint_3":
        return 602
    elif damage_type == "jietou":
        return 300
    elif damage_type == "phantom wave":
        return 504
    elif damage_type == "linehole":
        return 32
    elif damage_type == "bolt_loss_white":
        return 701
    elif damage_type == "bolt_loss_red":
        return 702
    elif damage_type == "bolt_crack_h":
        return 703
    elif damage_type == "head crack":
        return 505
    elif damage_type == "bottom crack":
        return 800
    elif damage_type == "badMidBolt":
        return 704
    elif damage_type == "hanfengMidFlaw":
        return 900
    elif damage_type == "unknown_45_white":
        return 705
    elif damage_type == "unknown_45_red":
        return 706
    else:
        return None  # 如果损伤类型不匹配任何已知类型，则返回None

def xml_to_csv(xml_files, csv_path):
    last_max_xmax = 0  # 初始化前一个文件的xmax最大值

    # 打开一个新的CSV文件用于写入
    with open(csv_path, 'w', newline='') as csvfile:
        csvwriter = csv.writer(csvfile)
        # 写入CSV的表头
        csvwriter.writerow(['xmin', 'ymin', 'xmax', 'ymax','Injury'])

        # 遍历所有XML文件
        for xml_file in xml_files:
            dom = xml.dom.minidom.parse(xml_file)
            root = dom.documentElement

            local_max_xmax = 0  # 初始化当前文件的xmax最大值

            # 获取所有bndbox节点
            bndboxes = root.getElementsByTagName('object')

            # 遍历每个bndbox节点
            for bndbox in bndboxes:
                xmin = to_int(bndbox.getElementsByTagName('xmin')[0].firstChild.data)
                ymin = to_int(bndbox.getElementsByTagName('ymin')[0].firstChild.data)
                xmax = to_int(bndbox.getElementsByTagName('xmax')[0].firstChild.data)
                ymax = to_int(bndbox.getElementsByTagName('ymax')[0].firstChild.data)
                Injury = get_damage_code(bndbox.getElementsByTagName('name')[0].firstChild.data)
                # 如果这不是第一个文件，将前一个文件的xmax最大值加到当前文件的xmin值上
                if last_max_xmax > 0:
                    xmin += last_max_xmax+3000
                    xmax += last_max_xmax+3000
                 # 更新当前文件的xmax最大值
                if xmax is not None:
                    local_max_xmax = max(local_max_xmax, xmax)
                    print(local_max_xmax)
                # 写入每个节点的值到CSV
                csvwriter.writerow([xmin, ymin, xmax, ymax,Injury])
            # 更新全局xmax最大值
            last_max_xmax = local_max_xmax

def get_xml_files(path):
    # 使用glob模块获取所有XML文件
    return glob.glob(os.path.join(path, '*.xml'))

if __name__ == '__main__':
    # xml_path = 'F:\\zhengwangwork\\H\\样本\\完整的伤损样本库20240313\\完整的伤损样本库20240308\\xml'
    # xml_path = 'F:\\testcsv\\xml'
    xml_path = 'F:\\png'
    # csv_path = 'F:\\testcsv\\output_xml.csv'
    csv_path = 'F:\\testcsv\\6.csv'
    xml_files = get_xml_files(xml_path)
    xml_to_csv(xml_files, csv_path)