将CryptoJS AES解密代码转为Python时遇UnicodeDecodeError求助

我正在把一段用CryptoJS实现的Node.js AES CBC解密代码转换成Python，已经写了部分实现，但运行时抛出UnicodeDecodeError: 'utf-8' codec can't decode byte 0xaf in position 0: invalid start byte错误。以下是相关代码和错误信息，求问题原因及解决方法。

原Node.js代码 #

const CryptoJS = require("crypto-js")

const ee = "#3vlsc6mdk2k400$"
const encrypt_data = "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"

var nn = CryptoJS.enc.Utf8.parse(ee)
var rr = CryptoJS.enc.Utf8.parse(ee.toUpperCase())
var ii = CryptoJS.AES.decrypt(encrypt_data, nn, {
    iv: rr,
    mode: CryptoJS.mode.CBC,
    padding: CryptoJS.pad.Pkcs7
})

var xx = ii.toString(CryptoJS.enc.Utf8)
var out = JSON.parse(xx);

我的Python代码 #

import json

from Cryptodome.Util.Padding import pad
from Cryptodome.Cipher import AES


def decrypt():
    ee = "#3vlsc6mdk2k400$"
    encrypt_data = "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"

    nn = ee.encode("utf-8")
    rr = ee.upper().encode("utf-8")

    cypher = AES.new(key=nn, mode=AES.MODE_CBC, iv=rr)
    encode_data = encrypt_data.encode("utf-8")
    padded_data = pad(
        data_to_pad=encode_data,
        block_size=16,
        style="pkcs7",
    )
    ii = cypher.decrypt(padded_data)
    xx = ii.decode("utf-8")
    out = json.loads(xx)
    return out


if __name__ == "__main__":
    decrypt()

错误信息 #

UnicodeDecodeError: 'utf-8' codec can't decode byte 0xaf in position 0: invalid start byte

问题原因及修复方案 #

核心错误点 #

1. 密文处理逻辑错误：Node.js的CryptoJS.AES.decrypt会自动识别Base64格式的密文并解码，但你在Python中直接将密文字符串转成UTF-8字节，相当于把Base64字符串当成了原始加密字节，完全不符合解密要求。
2. 填充操作完全搞反：PKCS7填充是加密阶段为了让数据长度符合AES块大小要求而添加的，解密时需要去除填充，但你反而给密文额外加了一层填充，导致解密结果完全混乱，无法解析为UTF-8字符串。

修复后的Python代码 #

import json
import base64
from Cryptodome.Util.Padding import unpad
from Cryptodome.Cipher import AES


def decrypt():
    ee = "#3vlsc6mdk2k400$"
    encrypt_data = "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"

    # 密钥和IV的处理和Node.js一致
    nn = ee.encode("utf-8")
    rr = ee.upper().encode("utf-8")

    cypher = AES.new(key=nn, mode=AES.MODE_CBC, iv=rr)
    # 1. 对Base64格式的密文进行解码，得到原始加密字节
    decode_data = base64.b64decode(encrypt_data)
    # 2. 解密得到带填充的明文
    encrypted_plaintext = cypher.decrypt(decode_data)
    # 3. 去除PKCS7填充
    unpadded_data = unpad(encrypted_plaintext, block_size=16, style="pkcs7")
    # 4. 解码为UTF-8字符串并解析JSON
    xx = unpadded_data.decode("utf-8")
    out = json.loads(xx)
    return out


if __name__ == "__main__":
    print(decrypt())

修复说明 #

• 使用base64.b64decode处理密文，还原成加密时的原始字节。
• 解密后调用unpad去除PKCS7填充，得到纯净的明文字节。
• 最后将明文字节解码为UTF-8字符串，再解析为JSON对象。

内容的提问来源于stack exchange，提问作者Jossy