@requence/cache

Tag-based function caching with pluggable backends — memory, Redis, or both.

Install

bun add @requence/cache

Features

• Fluent API — define cached functions with .define() chaining
• Tag-based invalidation — invalidate groups of cached results by tag
• Scoped caches — isolate cache entries per tenant, branch, or any grouping
• Pluggable backends — in-memory (default), Redis, or combined multi-layer
• Cascade invalidation — nested cache calls propagate tags automatically
• Request deduplication — concurrent calls to the same function are coalesced
• TTL & LRU eviction — configurable time-to-live and max-size per function
• Post-processing — transform cached return values without re-executing

Quick Start

import { createCache, addCacheTag } from '@requence/cache'

const cache = createCache({ memoryTTL: 60_000 })
  .define('getUser', async (id: string) => {
    addCacheTag(['user', id])
    return db.users.findById(id)
  })
  .define('listUsers', async () => {
    addCacheTag('user')
    return db.users.findAll()
  })

// Cached
await cache.getUser('abc')

// Invalidates both getUser('abc') and listUsers()
await cache.invalidateTag('user')

Redis Backend

import { createCache } from '@requence/cache'
import createRedisBackend from '@requence/cache/redis'

const cache = createCache({
  backend: createRedisBackend({
    url: process.env.REDIS_URL,
    prefix: 'myapp',
    ttl: 300_000,
  }),
})

Documentation

Full documentation — including concept guides and API reference — is available at:

https://cache.docs.requence.cloud

License

MIT