worker_map

worker_map

An simple abstraction for Node.js worker_threads, allowing you to create and share Map (hash table) between worker threads and the main process. This simplifies the process of managing shared data and communication between worker threads.

Under the hood, the library uses SharedArrayBuffer to create shared memory, enabling seamless data sharing between threads. Additionally, it uses Atomics mechanism to implement a mutex, ensuring thread safety and preventing race conditions during data access and manipulation.

Installation

npm i threadshare

Basic Example

First, let's create a simple hash map structure in main process, then create a worker thread and share the hash.

// main.js
const { Worker } = require('worker_threads');
const { WorkerMap } = require('worker_map');

const map = new WorkerMap();
map.set('balance', 100); // sync operation

new Worker('./worker.js', {
  workerData: {
    mapBuffer: map.toSharedBuffer(),
  },
});

setTimeout(() => {
    console.log(map.get('balance')); // 200
}, 50);

Now, let's access the shared hash map structure in the worker thread.

// worker.js
const { WorkerMap } = require('worker_map');
const { workerData } = require('worker_threads');

const map = new WorkerMap(workerData.mapBuffer);
console.log(map.get('balance')); // 100

// The change will be reflected in the main process as well
map.set('balance', 200);

Instance methods

map.get(key):

const name = map.get('name');

map.set(key, value)

map.set('name', 'John');

map.delete(key):

map.delete('name'); // true
map.delete('something'); // false because doesn't exist

map.has(key)

map.has('name'); // true
map.has('country'); // false

map.size()

map.has('size'); // 1

map.keys()

map.keys(); // [ 'name' ]

map.toSharedBuffer()

const buffer = map.toSharedBuffer();
const sameMap = new WorkerMap(buffer);

map.toObject()

const mapObject = map.toObject();
mapObject.name; // 'John'

TODO

• map.clear()
• map.entries()
• map.forEach()

Limitations

Please be aware of the following limitations when using our library:

1. Functions: Function types are not supported.
2. NaN Values: NaN values are not supported.