在列表中使用哈希函数时获取相同字符串的不同哈希摘要值

Question

我的程序中同一个词似乎得到了不同的摘要值。我不确定这是因为我将哈希函数保存在列表中（这样我就可以添加到列表中）

当我使用直接散列函数时，同一个词的散列摘要是相同的。当我使用列表中的哈希值时，情况有所不同。我做错了什么？

什么有效

import hashlib

bloom_len = 100

def bytes_to_int(hash_value):
    return int.from_bytes(hash_value, byteorder='big')  #big-endiang format

def bloom_index(hashint):
    return hashint % bloom_len


def hashIt(word):

    m1 = hashlib.md5()
    m2 = hashlib.sha1()
    m3 = hashlib.sha256()

    m4 = hashlib.sha3_512()
    m5 = hashlib.blake2s()

    m1.update(word)
    m2.update(word)
    m3.update(word)

    m4.update(word)
    m5.update(word)


    hash_values = [m1.digest(), m2.digest(), m3.digest(), m4.digest(), m5.digest()]
    hashints = list(map(bytes_to_int, hash_values))
    indices = list(map(bloom_index, hashints))

    print(indices)


inputWord = 'sent'
word = inputWord.encode('utf-8')
hashIt(word)

inputWord = 'blue'
word = inputWord.encode('utf-8')
hashIt(word)

inputWord = 'sent'
word = inputWord.encode('utf-8')
hashIt(word)

什么不起作用

import hashlib


class BloomFilter():
    def __init__(self, length = 100):

        self.bloomFilterLen = length
        self.bloomFilterArray = [0] * self.bloomFilterLen

        m1 = hashlib.md5()
        m2 = hashlib.sha3_512()
        m3 = hashlib.blake2s()        

        self.hashes = [m1, m2, m3]


    def encode(self, inputWord):
        encoded_word = inputWord.encode('utf-8')
        return encoded_word

    def bytes_to_int(self, hash_value):

        return int.from_bytes(hash_value, byteorder='big')  

    def bloom_index(self, hashint):

        return hashint % self.bloomFilterLen    

    def getIndices(self, inputWord):

        word = self.encode(inputWord)

        print(word)
        hashDigests = []

        for hashFunction in self.hashes:
            hashFunction.update(word)
            print('hashFunction ', hashFunction , '\n')
            print('hashDigest ', hashFunction.digest() , '\n')

            hashDigests.append(hashFunction.digest())


        hashInts = [self.bytes_to_int(h) for h in hashDigests]    
        #print('hashInts ', hashInts)

        bloomFilterIndices = [self.bloom_index(hInt) for hInt in hashInts]
        return bloomFilterIndices

    def insert(self, inputWord):

        bloomFilterIndices = self.getIndices(inputWord)

        for index in bloomFilterIndices:
            self.bloomFilterArray[index] = 1

        print(bloomFilterIndices)


    def lookup(self, inputWord):

        bloomFilterIndices = self.getIndices(inputWord)
        print('Inside lookup')
        print(bloomFilterIndices)

        for idx in bloomFilterIndices:
            print('idx value ', idx)
            print('self.bloomFilterArray[idx] value ', self.bloomFilterArray[idx])
            if self.bloomFilterArray[idx] == 0:
                # Indicates word not present in the bloom filter
                return False

        return True            


if __name__ == '__main__':

     word = 'sent'
     bloomFilter = BloomFilter()
     bloomFilter.insert(word)


     print(bloomFilter.lookup(word))

从第一个程序开始 - 我一直得到相同的整数索引

“已发送”的索引

[61, 82, 5, 53, 87]

“blue”的索引

[95, 25, 24, 69, 85]

“已发送”的索引

[61, 82, 5, 53, 87]

对于非工作程序，整数索引不同，当我打印出哈希摘要时也不同

“sent”的索引 - 第一次通过 add

[61, 53, 87]

HashDigest 来自 MD5 for 'sent'

hashDigest b'x\x91\x83\xb7\xe9\x86F\xc1\x1d_\x05D\xc8\xf3\xc4\xc9'

“sent”的索引 - 第二次通过 lookup

[70, 89, 8]

HashDigest 来自 MD5 for 'sent'

hashDigest b'\x95\x17bC\x17\x80\xb5\x9d]x\xca$\xda\x89\x06\x16'

Answer 1

所以我更改了__init中的代码__

来自

m1 = hashlib.md5()
m2 = hashlib.sha3_512()
m3 = hashlib.blake2s()        

self.hashes = [m1, m2, m3]

到

self.hashes = ['md5', 'sha3_512', 'blake2s']

然后在方法 getIndices() 的 for 循环内

从

更改而来

  for hashFunction in self.hashes:
        hashFunction.update(word)

到

for hashName in self.hashes:
    hashFunction = hashlib.new(hashName)
    hashFunction.update(word)

现在可以使用了！

Answer 2

哈希函数对象不能重复使用，您可以将这些函数对象移动到您的 getIndices 函数中：

import hashlib


class BloomFilter():
    def __init__(self, length = 100):

        self.bloomFilterLen = length
        self.bloomFilterArray = [0] * self.bloomFilterLen

    def encode(self, inputWord):
        encoded_word = inputWord.encode('utf-8')
        return encoded_word

    def bytes_to_int(self, hash_value):

        return int.from_bytes(hash_value, byteorder='big')  

    def bloom_index(self, hashint):

        return hashint % self.bloomFilterLen    

    def getIndices(self, inputWord):
        m1 = hashlib.md5()
        m2 = hashlib.sha3_512()
        m3 = hashlib.blake2s()        
        hashes = [m1, m2, m3]

        word = self.encode(inputWord)

        print(word)
        hashDigests = []

        for hashFunction in hashes:
            hashFunction.update(word)
            print('hashFunction ', hashFunction , '\n')
            print('hashDigest ', hashFunction.digest() , '\n')

            hashDigests.append(hashFunction.digest())


        hashInts = [self.bytes_to_int(h) for h in hashDigests]    
        #print('hashInts ', hashInts)

        bloomFilterIndices = [self.bloom_index(hInt) for hInt in hashInts]
        return bloomFilterIndices

    def insert(self, inputWord):

        bloomFilterIndices = self.getIndices(inputWord)

        for index in bloomFilterIndices:
            self.bloomFilterArray[index] = 1

        print(bloomFilterIndices)
        bloomFilterIndices = self.getIndices(inputWord)
        print(bloomFilterIndices)


    def lookup(self, inputWord):
        print('Inside lookup')
        bloomFilterIndices = self.getIndices(inputWord)
        print(bloomFilterIndices)

        for idx in bloomFilterIndices:
            print('idx value ', idx)
            print('self.bloomFilterArray[idx] value ', self.bloomFilterArray[idx])
            if self.bloomFilterArray[idx] == 0:
                # Indicates word not present in the bloom filter
                return False

        return True            


if __name__ == '__main__':

     word = 'sent'
     bloomFilter = BloomFilter()
     bloomFilter.insert(word)

     print(bloomFilter.lookup(word))

在列表中使用哈希函数时获取相同字符串的不同哈希摘要值

Getting different hash digest values for the same string when using hash functions in a list

python

hashlib