向列表追加XLField对象仅保留最后一个的问题求助

尝试将DataFrame的每行数据实例化为XLField类对象，追加到列表后，列表中所有元素都是最后一个对象的重复值。在追加前打印对象属性能得到预期不同结果，但遍历列表时全是重复的最后一个值。

核心代码 #

import pandas as pd

df_dict = pd.read_excel(excel_path).to_dict("records")    
xl_field_list = []
for field in df_dict:
    xl_field = XLField(
        field1=field["Field 1"],
        field2=field["Field 2"],
        field3=field["Field 3"],
        field4=field["Field 4"],
    )
    xl_field_list.append(xl_field)

for f in xl_field_list:
    print(f.field1)

运行结果 #

0052
0052
0052
0052
0052
0052

XLField类定义 #

import re

class XLField1:
    def __get__(self, obj, objtype=None):
        return self.value

    def __set__(self, obj, value):
        value = value.upper()
        if not re.fullmatch(r"[A-Z0-9]{2}", value):
            raise ValueError("Check Field 1")
        self.value = value


class XLField2:
    def __get__(self, obj, objtype=None):
        return self.value

    def __set__(self, obj, value):
        value = str(value).zfill(4)
        if not re.fullmatch(r"[0-9]{3,4}", value):
            raise ValueError("Check Field 2")
        self.value = value


class XLField3:
    Days_Of_The_Week = [
        "MONDAY",
        "TUESDAY",
        "WEDNESDAY",
    ]

    def __get__(self, obj, objtype=None):
        return self.value

    def __set__(self, obj, value):
        value = value.upper()
        if value not in XLField3.Days_Of_The_Week:
            raise ValueError("Check Field 3")
        self.value = value


class XLField4:
    INT_OR_COM = ["A", "D"]

    def __get__(self, obj, objtype=None):
        return self.value

    def __set__(self, obj, value):
        value = value.upper()
        if value not in XLField4.INT_OR_COM:
            raise ValueError("Check Field 4")
        self.value = value


class XLField:
    field1 = XLField1()
    field2 = XLField2()
    field3 = XLField3()
    field4 = XLField4()

    def __init__(self, field1, field2, field3, field4):
        self.field1 = field1
        self.field2 = field2
        self.field3 = field3
        self.field4 = field4

你实现的XLField1、XLField2等是描述符类，但这些描述符把值存在了自身实例（self.value）里，而不是存在被描述的对象（即XLField的实例）中。

XLField类里的field1、field2是类属性，所有XLField的实例共享同一个XLField1、XLField2实例。每次给self.field1赋值时，实际是修改了类属性中那个唯一的XLField1实例的value，后面的赋值会覆盖前面的，最终所有实例访问field1时，都会取到最后一次赋值的value。

修改描述符类，把值存在被描述的对象（obj）上，而不是描述符自身：

修正后的描述符类代码 #

import re

class XLField1:
    def __get__(self, obj, objtype=None):
        return obj._field1

    def __set__(self, obj, value):
        value = value.upper()
        if not re.fullmatch(r"[A-Z0-9]{2}", value):
            raise ValueError("Check Field 1")
        obj._field1 = value


class XLField2:
    def __get__(self, obj, objtype=None):
        return obj._field2

    def __set__(self, obj, value):
        value = str(value).zfill(4)
        if not re.fullmatch(r"[0-9]{3,4}", value):
            raise ValueError("Check Field 2")
        obj._field2 = value


class XLField3:
    Days_Of_The_Week = [
        "MONDAY",
        "TUESDAY",
        "WEDNESDAY",
    ]

    def __get__(self, obj, objtype=None):
        return obj._field3

    def __set__(self, obj, value):
        value = value.upper()
        if value not in XLField3.Days_Of_The_Week:
            raise ValueError("Check Field 3")
        obj._field3 = value


class XLField4:
    INT_OR_COM = ["A", "D"]

    def __get__(self, obj, objtype=None):
        return obj._field4

    def __set__(self, obj, value):
        value = value.upper()
        if value not in XLField4.INT_OR_COM:
            raise ValueError("Check Field 4")
        obj._field4 = value


class XLField:
    field1 = XLField1()
    field2 = XLField2()
    field3 = XLField3()
    field4 = XLField4()

    def __init__(self, field1, field2, field3, field4):
        self.field1 = field1
        self.field2 = field2
        self.field3 = field3
        self.field4 = field4

说明 #

• 每个描述符的__set__方法把处理后的值存入obj的私有属性（比如_field1），而不是描述符自身的self.value
• __get__方法从obj的私有属性中取值，这样每个XLField实例都有自己独立的属性值，不会互相覆盖

这样修改后，遍历xl_field_list时就能得到每个对象的真实属性值了。

内容的提问来源于stack exchange，提问作者redracer88