西马什

使用各种散列函数,如SipHash、MD5和SHA256探测近复制文本的Simhash执行

安装

npm i @counterrealist/simhash

使用量

import {SimHash, BitArray, HashFunction} from "@counterrealist/simhash"

const simHash = new SimHash({
    ngramSize: 3, // Default to 3
    hashFunction: HashFunction.SIPHASH // Default to SIPHASH, Options: SIPHASH, MD5, SHA256
});

const text1: string = "Hello, world!";
const text2: string = "Hell's world";

const text1_bitarray: BitArray = simHash.compute_bitarray(text1); // [0, 1, 0, 1, 1, 1, 1, 1, 0, 1, 0, 0, ...]
const text1_buffer: Buffer = simHash.compute_buffer(text1); // <Buffer 5f 4c d1 d8 77 30 f4 e5 fd af ec b7 58 c7 9c 5b>
const text1_hex: string = simHash.compute_hex(text1); // 5f4cd1d87730f4e5fdafecb758c79c5b

const text2_bitarray: BitArray = simHash.compute_bitarray(text2); // [0, 0, 0, 1, 1, 0, 0, 1, 0, 1, 1, 0, ...]
const text2_buffer: Buffer = simHash.compute_buffer(text2); // <Buffer 19 64 42 58 f7 48 04 e4 5d 03 0c 2f 50 d5 8e 4d>
const text2_hex: string = simHash.compute_hex(text2); // 19644258f74804e45d030c2f50d58e4d

const textSimilarity: number = simHash.similarity(text1, text2); // 0.6875
const textSimilarityFromHex: number = simHash.similarityFromHex(text1_hex, text2_hex); // 0.6875
const textSimilarityFromBuffers: number = simHash.similarityFromBuffers(text1_buffer, text2_buffer); // 0.6875
const textSimilarityFromBits: number = simHash.similarityFromBits(text1_bitarray, text2_bitarray); // 0.6875

实例

@ 反现实主义者/ simhash- crawler- 实例 ——一个简单的机器人,计算维基文章的修订之间的相似性.