利用python读取xml中的数据

悠闲的小鱼LIU

xml例子

xml = '''<?xml version='1.0' encoding='utf-8'?>
   <shape>square</shape>
   <degrees>360</degrees>
   <sides>4.0</sides>
   <shape>circle</shape>
   <degrees>360</degrees>
   <shape>triangle</shape>
   <degrees>180</degrees>
   <sides>3.0</sides>
</data>'''

方法一：利用cElementTree

from xml.etree import cElementTree as ET
import pandas as pd
# 读取xml字符串
root = ET.fromstring(text=xml)
# 读取xml文件
# tree = ET.ElementTree(file="text.xml")  
# root = tree.getroot() 
data = list()
for child in root:
    data1 = list()
    for son in child:
        data1.append(son.text)
    data.append(data1)
df = pd.DataFrame(data, columns=['shape', 'degrees', 'sides'])
print(df)

输出结果：

    shape  degrees  sides
0    square      360    4.0
1    circle      360    NaN
2  triangle      180    3.0

如果 shape 、degrees、sides 不是按照一定规律排列，这样取数据容易出错。

比如将最后一组按照degrees、 shape 、sides 排列，输出结果便会变成：

    shape   degrees sides
0  square       360   4.0
1  circle       360  None
2     180  triangle   3.0

方法二：利用read_xml()

import pandas as pd
df = pd.read_xml(xml)
print(df)

输出结果：

    shape  degrees  sides
0    square      360    4.0
1    circle      360    NaN
2  triangle      180    3.0

方法三：利用pd.json_normalize()

将xml转为类似json的格式
利用pd.json_normalize() 读到dataframe

def fun1(root):
    dic1 = dict()
    for child in root:
        if bool(child) is True:  # 有下一层
            print(child.tag)
            dic2 = fun1(child)  # 自己调用自己
            value = dic1.get(child.tag)  # 存在返回，不存在返回None
            if value:  # 存在
                value.append(dic2)
                dic1[child.tag] = value
            else:
                dic1[child.tag] = [dic2]
        else:
            dic1[child.tag] = child.text
    return dic1
if __name__ == '__main__':
    from xml.etree import cElementTree as ET
    import pandas as pd
    root = ET.fromstring(text=xml)