This package implements a bloom filter for a general type of usage, with some usefull defaults in place. The implementation follows the official algorithm, which can be seen here.
const BloomFilter = require("DynamicBloomFilter");
const filter = new BloomFilter({ estimatedElementCount: 100 });
filter.insert("element 1");
filter.insert("element 2");
filter.test("element 1"); // returns true
filter.test("element 3"); // returns falseBloomFilter accepts an option parameter that configures various aspects of the Bloom Filter.
The following options are available:
| Option | Default value | Description |
|---|---|---|
| bitCount | null | bit count (represents the m variable from the algorithm) - if missing it will be computed based on estimatedElementCount |
| hashFunctionCount | null | k hash functions count (represents the k variable from the algorithm) - if missing it will be computed based on estimatedElementCount |
| estimatedElementCount | 0 | estimated number of elements from the collection |
| falsePositiveTolerance | 0.000001 | allowed probability of false positives |
| hashFunction | linearFowlerNollVoJenkinsHashFunction (defined inside src/hash-functions/linear-fowler–noll–vo-jenkins-hash-function.js) - linear hash function based on combination of Fowler–Noll–Vo hash function - FNV-1a hash variant and Jenkins hash function | default function that returns the element's hash |
| cryptoHashFunction | sha2 | crypto hash function that returns the element's hash |
| cryptoHashFunctionCount | 0 | number of crypto hash functions to be used (will be used at first before the default hashFunction) |
| cryptoSecret | "secret" | crypto hash function secret |
| BitCollectionStrategy | InMemoryBitCollectionStrategy (defined in src/in-memory-bit-collection-strategy.js) - keeps the bit collection in memory and applies bit operations | strategy which interacts with the bit collection |
Needs to be a function that returns a positive number representing the bit position that will be set to 1. It accepts 3 arguments: (data, index, options):
- data: the value that needs to be hashed
- index: the index of the current k hash function, ranges [0, k)
- options: the options object that is used by the BloomFilter
Needs to be a function constructor that accepts an single parameters representing the option object that is used by the BloomFilter. The function must have 2 methods defined:
- getIndex - function that receives an index: number and returns the value of the index's bit of the collection
- setIndex - function that receives an index: number and sets the value of the index's bit of the collection to 1