含Unicode字符的消息哈希：NodeJS与Python结果不一致的解决方法

我需要对包含特殊字节（如\x1a、\xab）的消息进行SHA256哈希运算，但Node.js与Python的计算结果不一致。期望Node.js能输出和Python相同的哈希值：947270b0b8041a92ba82ef37661b692a4a150532b88de59bf95e965ceb5c07f8。

现有代码 #

Node.js（使用CryptoJS） #

const CryptoJS = require('crypto-js');

let message = '\x1aSmartCash Signed Message:\n\xabCTxIn(COutPoint(7bb8ad134928a003752beb098471af5a66fc5475ff96b5ba4c2e1c4cbac3aa13, 0), scriptSig=)000000000002c5c2ef4afc588492773e6bbb18e4f2374b6dc159ef257667bd881667906410';
let hash = CryptoJS.SHA256(message).toString();

console.log('hash', hash); // 989c004534c6962293c95a9438bdb926c92c1d8b4dec0f4f1e535defa171e5fe

Python #

import hashlib

def to_bytes(something, encoding='utf8'):
    """
    cast string to bytes() like object, but for python2 support it's bytearray copy
    """
    if isinstance(something, bytes):
        return something
    if isinstance(something, str):
        return something.encode(encoding)
    elif isinstance(something, bytearray):
        return bytes(something)
    else:
        raise TypeError("Not a string or bytes like object")

def sha256(x):
    x = to_bytes(x, 'utf8')
    return bytes(hashlib.sha256(x).digest())

def Hash_Sha256(x):
    x = to_bytes(x, 'utf8')
    out = bytes(sha256(x))
    return out

message = b'\x1aSmartCash Signed Message:\n\xabCTxIn(COutPoint(7bb8ad134928a003752beb098471af5a66fc5475ff96b5ba4c2e1c4cbac3aa13, 0), scriptSig=)000000000002c5c2ef4afc588492773e6bbb18e4f2374b6dc159ef257667bd881667906410'

print('hash', Hash_Sha256(message).hex()) # 947270b0b8041a92ba82ef37661b692a4a150532b88de59bf95e965ceb5c07f8

问题根源 #

• Python里的b'\x1a'、b'\xab'是单字节原始数据，直接对应0x1A、0xAB的字节值，没有编码转换。
• Node.js中，字符串'\x1a'、'\xab'被CryptoJS默认按UTF-8编码处理：其中\xab不属于合法UTF-8字符，会被转译为UTF-8的替代字符序列（0xEF 0xBF 0xBD），导致最终参与哈希计算的字节序列和Python完全不同。

解决方案 #

要让两边使用完全一致的字节序列计算哈希，需要让CryptoJS按单字节编码规则处理字符串，或者直接传入原始字节数组：

方案1：用CryptoJS的enc.Latin1编码字符串 #

Latin1（ISO-8859-1）编码会将每个字符映射为单字节，和Python的bytes行为完全匹配：

const CryptoJS = require('crypto-js');

let message = '\x1aSmartCash Signed Message:\n\xabCTxIn(COutPoint(7bb8ad134928a003752beb098471af5a66fc5475ff96b5ba4c2e1c4cbac3aa13, 0), scriptSig=)000000000002c5c2ef4afc588492773e6bbb18e4f2374b6dc159ef257667bd881667906410';
// 将字符串按Latin1解析为字节数组后计算哈希
let hash = CryptoJS.SHA256(CryptoJS.enc.Latin1.parse(message)).toString();

console.log('hash', hash); // 947270b0b8041a92ba82ef37661b692a4a150532b88de59bf95e965ceb5c07f8

方案2：直接用Node.js Buffer传递原始字节 #

如果消息本身是原始字节数据，用Buffer定义更直观，和Python的b''语法逻辑一致：

const CryptoJS = require('crypto-js');
const { Buffer } = require('buffer');

// 用Buffer拼接原始字节，和Python的字节序列完全对齐
const messageBuffer = Buffer.concat([
  Buffer.from([0x1a]),
  Buffer.from('SmartCash Signed Message:\n'),
  Buffer.from([0xab]),
  Buffer.from('CTxIn(COutPoint(7bb8ad134928a003752beb098471af5a66fc5475ff96b5ba4c2e1c4cbac3aa13, 0), scriptSig=)000000000002c5c2ef4afc588492773e6bbb18e4f2374b6dc159ef257667bd881667906410')
]);

// 将Buffer转换为CryptoJS可处理的WordArray
let hash = CryptoJS.SHA256(CryptoJS.lib.WordArray.create(messageBuffer)).toString();

console.log('hash', hash); // 947270b0b8041a92ba82ef37661b692a4a150532b88de59bf95e965ceb5c07f8

验证 #

两种方案都能输出和Python完全一致的哈希值，解决了特殊字节编码不一致的问题。

内容的提问来源于stack exchange，提问作者CODEHUB