python解析xml实例

如下，一个银行卡打标签后导出的数据

<?xml version="1.0" encoding="ISO-8859-1"?>
<annotation>
<filename>a001.jpg</filename>
<folder>users/three33//card</folder>
<source>
<submittedBy>three</submittedBy>
</source>
<imagesize>
<nrows>2240</nrows>
<ncols>3968</ncols>
</imagesize>
<object>
<name>numbers</name>
<deleted>0</deleted>
<verified>0</verified>
<occluded>no</occluded>
<attributes>6228480808055442079</attributes>
<parts>
<hasparts/>
<ispartof/>
</parts>
<date>12-May-2019 06:21:39</date>
<id>0</id>
<type>bounding_box</type>
<polygon>
<username>anonymous</username>
<pt>
<x>927</x>
<y>1278</y>
</pt>
<pt>
<x>3269</x>
<y>1278</y>
</pt>
<pt>
<x>3269</x>
<y>1475</y>
</pt>
<pt>
<x>927</x>
<y>1475</y>
</pt>
</polygon>
</object>
</annotation>

（上面的代码无法保留格式，还是截张图吧

现要将其中的标记的四个坐标和银行卡号读取出来，并保存到文本文件。由于有几百张图片，需要批处理。

代码：

import os
import sys
import xml.etree.cElementTree as ET


from_path = "./card"     //输入文件夹
to_path = "./cardout"    //输出文件夹
files = os.listdir(from_path)
files.sort()      #按字典序排序


i = 1
for filename in files:
    
    dir1 = os.path.join(from_path, filename)
    tree = ET.ElementTree(file=dir1)
    root = tree.getroot()

    new_filename = filename[:-4] + ".txt"
    dir2 = os.path.join(to_path,new_filename)

    fobj = open(dir2,'w+')

    print("time: %d, from_filename: %s, to_filename: %s" % (i, dir1, dir2))

    for elem in tree.iterfind('object/polygon/pt'):
        fobj.write((elem[0].text + ',' + elem[1].text + ','))
        #print(elem[0].text + ',' + elem[1].text + ',')

    for elem in tree.iterfind('object/attributes'):
        fobj.write(elem.text)

    fobj.close()
    i = i + 1
    

效果：