Options

Here at the different possible options of BentoCache. Some of them are configurable, either globally, at the store level or at operation level ( when calling core methods like getOrSet, get, etc. ).

Order of precedence is as follows: 1. Operation level > 2. Store level > 3. Global level.

const bento = new BentoCache({
  default: 'memory',

  // Global level 👇
  ttl: '1h',
  gracePeriod: {
    enabled: true,
    duration: '6h',
    fallbackDuration: '5m'
  },

  stores: {
    memory: bentostore({
      // Store level 👇
      ttl: '30m',
      gracePeriod: { enabled: false }
    })
  }
})

bento.getOrSet('key', () => fetchFromDb(), {
  // Operation level 👇
  ttl: '1h',
  gracePeriod: { enabled: true }
})

TTL Formats

Quick note about TTLs and durations. Everywhere you are asked to provide a TTL or a duration, you can provide either:

• a number in milliseconds.
• a string, duration in human-readable format, see lukeed/ms for more details on the different formats accepted.
• In some cases : a null value. In the context of a TTL, this means that the item will never expire for example.

List of options

prefix

Default: undefined

Levels: global, store

This prefix will be added in front of all your cache keys. This can be useful, for example, in the case where you share a Redis with another application. Particularly when you need to call .clear(). Without a prefix, all keys will be deleted, even those not added by your application.

ttl

Default: 30s

Levels: global, store, operation

The TTL of the item to cache. See TTL formats.

suppressL2Errors

Default: false

Levels: global, store, operation

If false, then errors thrown by your L2 cache will be rethrown, and you will have to handle them yourself. Otherwise, they will just be ignored.

Note that in some cases, like when you use Grace Periods, errors will not be thrown, even if this option is set to false. Since this is the whole point of grace periods.

earlyExpiration

Default: undefined

Levels: global, store, operation

Percentage of the TTL (between 0 and 1) that corresponds to the moment from which bentocache should launch a background refresh of your key. For example, if your TTL is 10m, and your earlyExpiration is 0.8, then a background refresh will be launched when the key is requested and there is less than 2 minutes of TTL remaining.

gracePeriod

Default undefined

Levels: global, store, operation

An object to configure the grace period:

{
  enabled: true,
  duration: '6h',
  fallbackDuration: '5m'
}

timeouts

Default: undefined Levels: global, store, operation

An object to configure the timeouts

{
  soft: '1s',
  hard: '3s'
}

Basically, when invoking a factory :

• If the factory takes more than soft to resolve and a graced/stale value is still available, then it will be returned. Factory will continue to run in the background.
• If the factory takes more than hard to resolve, then an exception will be thrown. You need to handle it yourself.

lockTimeout

Default: undefined

Levels: global, store, operation

The maximum amount of time (in milliseconds) that the in-memory lock for stampeded protection can be held. If the lock is not released before this timeout, it will be released automatically.

This is usually not needed, but can provide an extra layer of protection against theoretical deadlocks.

logger

Default: undefined.

Levels: global

Only configurable at the BentoCache level.

See logger for more details.

emitter

Default: new EventEmitter().

Levels: global

Only configurable at the BentoCache level.

See events for more details.