nestjs-cache-mediator

nestjs-cache-mediator is a highly abstract, type‑safe NestJS module that implements a cache‑first strategy using BullMQ for distributed job scheduling and a pluggable cache driver (via an interface) for storing results (e.g. in Redis). This package is designed to optimize performance and concurrency by ensuring that multiple concurrent requests for the same cache key share the same job.

Features

• Abstract & Generic API:
  Define your own job handlers with full TypeScript support and plug them into the cache‑first workflow.

• BullMQ Integration:
  Uses BullMQ to schedule and execute expensive data retrieval tasks, using the cache key as a unique job identifier to prevent duplicate work.

• Pluggable Cache Driver:
  Implements an ICacheDriver interface so you can easily use any cache system (e.g. Redis via ioredis).

• Warm-Up Support:
  Provides a method to force a cache refresh (for instance, on application boot).

• Optimized for Concurrency:
  Designed for distributed environments where multiple instances share the same Redis/BullMQ configuration, so duplicate work is avoided.

Installation

Install the package via npm:

1npm install nestjs-cache-mediator

Setup

Provide a Cache Driver

Implement the ICacheDriver interface to integrate your chosen cache (e.g. Redis). For example, using ioredis:

1// redis-cache-driver.ts
2import { ICacheDriver } from 'nestjs-cache-mediator';
3import * as Redis from 'ioredis';
4
5export class RedisCacheDriver implements ICacheDriver {
6  private client: Redis.Redis;
7  constructor(options?: Redis.RedisOptions) {
8    this.client = new Redis(options);
9  }
10  async get(key: string): Promise<string | null> {
11    return this.client.get(key);
12  }
13  async set(key: string, value: string, ttl: number): Promise<void> {
14    await this.client.set(key, value, 'EX', ttl);
15  }
16  async del(key: string): Promise<void> {
17    await this.client.del(key);
18  }
19}

Module Integration

In your NestJS application's module, import CacheFirstModule and provide your cache driver:

1// app.module.ts
2import { Module } from '@nestjs/common';
3import { CacheFirstModule, ICacheDriver } from 'nestjs-cache-mediator';
4import { RedisCacheDriver } from './redis-cache-driver';
5
6@Module({
7  imports: [CacheFirstModule],
8  providers: [
9    {
10      provide: ICacheDriver,
11      useValue: new RedisCacheDriver({ host: 'localhost', port: 6379 }),
12    },
13  ],
14})
15export class AppModule {}

Usage

Register a Job Handler

Register your job handler for a given job type. This should be done early (for example, in a bootstrap function or in the module's initialization):

1import { CacheFirstService } from 'nestjs-cache-mediator';
2
3interface SomeDataType {
4  foo: string;
5  bar: number;
6}
7
8CacheFirstService.registerHandler<{ extraParam: string }, SomeDataType>(
9  'getSomeData',
10  async (params) => {
11    // Implement your data retrieval logic here.
12    return { foo: `Hello ${params.extraParam}`, bar: 42 };
13  }
14);

Use the Cache-First Service

Inject the CacheFirstService into your own service and call its generic method:

1import { Injectable } from '@nestjs/common';
2import { CacheFirstService } from 'nestjs-cache-mediator';
3
4@Injectable()
5export class MyDataService {
6  constructor(private readonly cacheFirstService: CacheFirstService) {}
7
8  async getData(extraParam: string): Promise<{ foo: string; bar: number }> {
9    const cacheKey = `myData:getSomeData:${extraParam}`;
10    return this.cacheFirstService.cacheFirst<{ foo: string; bar: number }, { extraParam: string }>(
11      cacheKey,
12      3600,  // Redis TTL: 1 hour
13      10000, // Job TTL: 10 seconds
14      'getSomeData',
15      { extraParam }
16    );
17  }
18}

Warm Up the Cache

You can force a refresh (warm-up) of a cache key during application boot or on demand:

1await this.cacheFirstService.warmCacheForKey<{ foo: string; bar: number }, { extraParam: string }>(
2  'myData:getSomeData:example',
3  3600,  // 1 hour
4  10000, // 10 seconds
5  'getSomeData',
6  { extraParam: 'example' }
7);

API Reference

CacheFirstService Methods

• cacheFirst<T, P>(cacheKey: string, redisTTL: number, jobTTL: number, jobType: string, params: P): Promise<T>
  Retrieves data using a cache-first strategy.

  • cacheKey: A unique cache key (also used as the Bull job id).
  • redisTTL: Time-to-live in seconds for the cached data.
  • jobTTL: Maximum wait time (in milliseconds) for the job to complete.
  • jobType: Identifier for the job handler.
  • params: Parameters passed to the job handler.

• warmCacheForKey<T, P>(cacheKey: string, redisTTL: number, jobTTL: number, jobType: string, params: P): Promise<T>
  Forces a refresh of the cache by deleting any existing entry and scheduling a new job.

Job Handler Registration

• CacheFirstService.registerHandler<P, T>(jobType: string, handler: CacheJobHandler<P, T>): void
  Register a function that will be used to fetch data when a cache miss occurs.

BullMQ Processor

The processor automatically executes jobs added to the cacheFirstQueue by calling the registered handler for the job type.

Contributing

Contributions, improvements, and bug fixes are welcome. Please open an issue or submit a pull request on GitHub.

License

This project is licensed under the MIT License.