Cacheble Utils

@cacheable/utils is a collecton of utility functions, helpers, and types for cacheable and other caching libraries. It provides a robust set of features to enhance caching capabilities, including:

Data Types for Caching Items
Hash Functions for Key Generation
Coalesce Async for Handling Multiple Promises
Statistics for Tracking Cache Metrics
Sleep / Delay for Testing and Timing
Memoization for wraping or get / set options
Time to Live (TTL) Helpers
Tag-Based Cache Invalidation

Getting Started
Cacheable Types
Coalesce Async
Hash Functions
Shorthand Time Helpers
Sleep Helper
Statistics
Time to Live (TTL) Helpers
Run if Function Helper
Less Than Helper
Is Object Helper
Wrap / Memoization for Sync and Async Functions
Get Or Set Memoization Function
Cache Tags
How to Contribute
License and Copyright

Getting Started

npm install @cacheable/utils --save

Cacheable Types

The @cacheable/utils package provides various types that are used throughout the caching library. These types help in defining the structure of cached items, ensuring type safety and consistency across your caching operations.


/**
 * CacheableItem
 * @typedef {Object} CacheableItem
 * @property {string} key - The key of the cacheable item
 * @property {any} value - The value of the cacheable item
 * @property {number|string} [ttl] - Time to Live - If you set a number it is miliseconds, if you set a string it is a human-readable
 * format such as `1s` for 1 second or `1h` for 1 hour. Setting undefined means that it will use the default time-to-live. If both are
 * undefined then it will not have a time-to-live.
 */
export type CacheableItem = {
	key: string;
	value: any;
	ttl?: number | string;
};

/**
 * CacheableStoreItem
 * @typedef {Object} CacheableStoreItem
 * @property {string} key - The key of the cacheable store item
 * @property {any} value - The value of the cacheable store item
 * @property {number} [expires] - The expiration time in milliseconds since epoch. If not set, the item does not expire.
 */
export type CacheableStoreItem = {
	key: string;
	value: any;
	expires?: number;
};

Coalesce Async

The coalesceAsync function is a utility that allows you to handle multiple asynchronous operations efficiently. It was designed by Douglas Cayers https://github.com/douglascayers/promise-coalesce. It helps in coalescing multiple promises into a single promise, ensuring that only one operation is executed at a time for the same key.

import { coalesceAsync } from '@cacheable/utils';

const fetchData = async (key: string) => {
  // Simulate an asynchronous operation
  return new Promise((resolve) => setTimeout(() => resolve(`Data for ${key}`), 1000));
};

const result = await Promise.all([
	coalesceAsync('my-key', fetchData),
	coalesceAsync('my-key', fetchData),
	coalesceAsync('my-key', fetchData),
]);
console.log(result); // Data for my-key only executed once

Hash Functions

The @cacheable/utils package provides hash functions that can be used to generate unique keys for caching operations. These functions are useful for creating consistent and unique identifiers for cached items.

The hashing API provides both async (for cryptographic algorithms) and sync (for non-cryptographic algorithms) methods.

Async Hashing (Cryptographic Algorithms)

Use hash() and hashToNumber() for cryptographic algorithms like SHA-256, SHA-384, and SHA-512:

import { hash, hashToNumber, HashAlgorithm } from '@cacheable/utils';

// Hash using SHA-256 (default)
const key = await hash('my-cache-key');
console.log(key); // Unique hash for 'my-cache-key'

// Hash with specific algorithm
const sha512Hash = await hash('my-data', { algorithm: HashAlgorithm.SHA512 });

// Convert hash to number within range
const min = 0;
const max = 10;
const result = await hashToNumber({foo: 'bar'}, { min, max, algorithm: HashAlgorithm.SHA256 });
console.log(result); // A number between 0 and 10 based on the hash value

Sync Hashing (Non-Cryptographic Algorithms)

Use hashSync() and hashToNumberSync() for faster, non-cryptographic algorithms like DJB2, FNV1, MURMER, and CRC32:

import { hashSync, hashToNumberSync, HashAlgorithm } from '@cacheable/utils';

// Hash using DJB2 (default for sync)
const key = hashSync('my-cache-key');
console.log(key); // Unique hash for 'my-cache-key'

// Hash with specific algorithm
const fnv1Hash = hashSync('my-data', { algorithm: HashAlgorithm.FNV1 });

// Convert hash to number within range
const min = 0;
const max = 10;
const result = hashToNumberSync({foo: 'bar'}, { min, max, algorithm: HashAlgorithm.DJB2 });
console.log(result); // A number between 0 and 10 based on the hash value

Available Hash Algorithms

Cryptographic (Async):

HashAlgorithm.SHA256 - SHA-256 (default for async methods)
HashAlgorithm.SHA384 - SHA-384
HashAlgorithm.SHA512 - SHA-512

Non-Cryptographic (Sync):

HashAlgorithm.DJB2 - DJB2 (default for sync methods)
HashAlgorithm.FNV1 - FNV-1
HashAlgorithm.MURMER - Murmur hash
HashAlgorithm.CRC32 - CRC32

Shorthand Time Helpers

The @cacheable/utils package provides a shorthand function to convert human-readable time strings into milliseconds. This is useful for setting time-to-live (TTL) values in caching operations.

You can also use the shorthandToMilliseconds function:

import { shorthandToMilliseconds } from '@cacheable/utils';

const milliseconds = shorthandToMilliseconds('1h');
console.log(milliseconds); // 3600000

You can also use the shorthandToTime function to get the current date plus the shorthand time:

import { shorthandToTime } from '@cacheable/utils';

const currentDate = new Date();
const timeInMs = shorthandToTime('1h', currentDate);
console.log(timeInMs); // Current date + 1 hour in milliseconds since epoch

Sleep Helper

The sleep function is a utility that allows you to pause execution for a specified duration. This can be useful in testing scenarios or when you need to introduce delays in your code.

import { sleep } from '@cacheable/utils';

await sleep(1000); // Pause for 1 second
console.log('Execution resumed after 1 second');

Statistics

The Stats class provides a unified, event-driven way to track caching metrics such as hits, misses, hit rate, item counts, and approximate memory usage. It can be driven two ways:

Imperatively — call increment / decrement (or the named helpers) directly from your code.
Event-driven — subscribe it to an event emitter (such as @cacheable/node-cache or a Node EventEmitter) and let matching events update the counters automatically.

Statistics are opt-in: a new Stats instance is disabled by default and ignores every update until enabled, so there is zero tracking overhead unless you ask for it.

import { Stats } from '@cacheable/utils';

const stats = new Stats({ enabled: true });

stats.incrementHits();
stats.incrementMisses();
stats.incrementGets();

console.log(stats.hits);    // 1
console.log(stats.misses);  // 1
console.log(stats.hitRate); // 0.5

Available Statistics

Every counter is exposed as a read-only property:

Property	Type	Description
`hits`	`number`	Number of cache hits.
`misses`	`number`	Number of cache misses.
`gets`	`number`	Number of get operations.
`sets`	`number`	Number of set operations.
`deletes`	`number`	Number of delete operations.
`clears`	`number`	Number of clear operations.
`count`	`number`	Number of items currently tracked.
`ksize`	`number`	Approximate size of all keys, in bytes.
`vsize`	`number`	Approximate size of all values, in bytes.

Computed Properties

Property	Type	Description
`hitRate`	`number`	`hits / (hits + misses)`, or `0` when there have been no lookups.
`missRate`	`number`	`misses / (hits + misses)`, or `0` when there have been no lookups.

Metadata

Property	Type	Description
`enabled`	`boolean`	Whether tracking is currently on.
`lastUpdated`	`number \| undefined`	Timestamp (ms since epoch) of the last update while enabled.
`lastReset`	`number \| undefined`	Timestamp (ms since epoch) of the last `reset()` / `clear()`.

Enabling, Disabling, and Clearing

const stats = new Stats(); // disabled by default

stats.enable();          // start tracking (or: stats.enabled = true)
stats.incrementHits();
console.log(stats.hits); // 1

stats.disable();         // stop tracking (or: stats.enabled = false)
stats.incrementHits();
console.log(stats.hits); // still 1

stats.clear();           // reset every counter back to 0 (alias of reset())
console.log(stats.hits); // 0

reset() / clear() — set every counter back to 0 and record lastReset.
resetStoreValues() — reset only count, ksize, and vsize, leaving the hit/miss history intact.

Incrementing and Decrementing

Use the unified increment / decrement methods with any counter field, or the named helpers. All updates are ignored while disabled.

const stats = new Stats({ enabled: true });

// Unified API — optional amount (defaults to 1)
stats.increment('hits');
stats.increment('sets', 5);
stats.decrement('count', 2);

// Named helpers
stats.incrementHits();
stats.incrementMisses();
stats.incrementGets();
stats.incrementSets();
stats.incrementDeletes();
stats.incrementClears();
stats.incrementCount();
stats.decreaseCount();

// Approximate key/value sizes
stats.incrementKSize('my-key');  // adds the byte size of the key
stats.incrementVSize({ a: 1 });  // adds the byte size of the value
stats.decreaseKSize('my-key');
stats.decreaseVSize({ a: 1 });
stats.setCount(10);              // set the item count directly

Snapshot

toJSON() (aliased as snapshot()) returns a plain object of every counter, the computed rates, and the timestamps — handy for logging or sending to a metrics system.

const stats = new Stats({ enabled: true });
stats.incrementHits(3);
stats.incrementMisses();

console.log(stats.toJSON());
// {
//   enabled: true,
//   hits: 3, misses: 1, gets: 0, sets: 0, deletes: 0, clears: 0,
//   vsize: 0, ksize: 0, count: 0,
//   hitRate: 0.75, missRate: 0.25,
//   lastUpdated: 1749513600000, lastReset: undefined
// }

Event-Driven Tracking

Instead of calling the increment methods yourself, you can subscribe a Stats instance to an emitter and have events update the counters automatically. The emitter is duck-typed — anything with .on() (plus .off() or .removeListener() to detach) works, including Hookified-based classes and Node's EventEmitter.

An event map describes how each event name updates the stats. A map value can be:

a single field — "sets"
an array of fields — ["hits", "gets"]
a custom handler — (stats, ...args) => void

import { Stats, nodeCacheStatsEventMap } from '@cacheable/utils';
import { NodeCache } from '@cacheable/node-cache';

const cache = new NodeCache();
const stats = new Stats({ enabled: true });

// nodeCacheStatsEventMap maps set -> sets, del -> deletes, flush -> clears,
// and flush_stats -> reset.
stats.subscribe(cache, nodeCacheStatsEventMap);

cache.set('key', 'value');
console.log(stats.sets); // 1

stats.unsubscribe(); // detach all listeners (or pass an emitter to detach just one)

You can also provide your own map for any emitter:

import { EventEmitter } from 'node:events';
import { Stats } from '@cacheable/utils';

const emitter = new EventEmitter();
const stats = new Stats({ enabled: true });

stats.subscribe(emitter, {
  'cache:hit': ['hits', 'gets'],
  'cache:miss': ['misses', 'gets'],
  evicted: (s) => s.incrementDeletes(),
});

emitter.emit('cache:hit', { key: 'a' });
console.log(stats.hitRate); // 1

You can subscribe to multiple emitters from a single Stats instance, and pass an emitter to unsubscribe(emitter) to detach just that one. Counting is gated by enabled, so you can subscribe first and toggle tracking on later — handlers do not run at all while disabled.

Per-Key Tracking (Most and Least Used Keys)

To find your hottest and coldest keys, enable per-key tracking with trackKeys. Each recorded key keeps its own breakdown of hits, misses, gets, sets, and deletes, plus a computed total count and per-key hitRate.

import { Stats } from '@cacheable/utils';

const stats = new Stats({ enabled: true, trackKeys: true });

stats.recordKey('user:1', 'hits');
stats.recordKey('user:1', 'gets', 5);
stats.recordKey('user:2', 'misses');

// 100 most used keys by total operations (descending)
console.log(stats.mostUsedKeys(100));
// [
//   { key: 'user:1', count: 6, hits: 1, misses: 0, gets: 5, sets: 0, deletes: 0, hitRate: 1 },
//   { key: 'user:2', count: 1, hits: 0, misses: 1, gets: 0, sets: 0, deletes: 0, hitRate: 0 }
// ]

// 100 least used keys by total operations (ascending)
console.log(stats.leastUsedKeys(100));

// Rank by a single counter instead of the total
console.log(stats.mostUsedKeys(100, 'hits'));

// Inspect one key, or read the (read-only) tracked-keys map directly
console.log(stats.keyStats('user:1'));
console.log(stats.trackedKeys.size);

stats.clearKeys(); // clear per-key stats only (reset() clears these too)

Both mostUsedKeys and leastUsedKeys default to 100 entries, and trackedKeys is included in toJSON() snapshots.

Per-key tracking is fed two ways, just like the aggregate counters:

Imperatively — call recordKey(key, field, amount?) wherever you already increment stats.
Event-driven — nodeCacheStatsEventMap automatically records keys from set/del events when trackKeys is on, and custom event-map handlers can call recordKey with whatever the payload carries.

Memory is proportional to the number of unique keys tracked, so trackKeys is off by default. You can also set maxTrackedKeys as a safety cap — when exceeded, the lowest-count keys are pruned. Note that pruning keeps mostUsedKeys approximately accurate but makes leastUsedKeys unreliable (the pruned keys are the least used), so leave it unset if you need exact least-used-key rankings.

Note: a built-in map is provided only where a library's events map cleanly to stats. nodeCacheStatsEventMap is included because @cacheable/node-cache emits each lifecycle event exactly once. Libraries that emit per-store probes or omit events on a miss (such as cacheable and cache-manager) should be wired with a custom map or driven imperatively so the counts stay accurate.

Time to Live (TTL) Helpers

The @cacheable/utils package provides helpers for managing time-to-live (TTL) values for cached items.

You can use the calculateTtlFromExpiration function to calculate the TTL based on an expiration date:

import { calculateTtlFromExpiration } from '@cacheable/utils';

const expirationDate = new Date(Date.now() + 1000 * 60 * 5); // 5 minutes from now
const ttl = calculateTtlFromExpiration(Date.now(), expirationDate);
console.log(ttl); // 300000

You can also use getTtlFromExpires to get the TTL from an expiration date:

import { getTtlFromExpires } from '@cacheable/utils';

const expirationDate = new Date(Date.now() + 1000 * 60 * 5); // 5 minutes from now
const ttl = getTtlFromExpires(expirationDate);
console.log(ttl); // 300000

You can use getCascadingTtl to get the TTL for cascading cache operations:

import { getCascadingTtl } from '@cacheable/utils';
const cacheableTtl = 1000 * 60 * 5; // 5 minutes
const primaryTtl = 1000 * 60 * 2; // 2 minutes
const secondaryTtl = 1000 * 60; // 1 minute
const ttl = getCascadingTtl(cacheableTtl, primaryTtl, secondaryTtl);

Run if Function Helper

The runIfFn utility function provides a convenient way to conditionally execute functions or return values based on whether the input is a function or not. This pattern is commonly used in UI libraries and configuration systems where values can be either static or computed.

import { runIfFn } from '@cacheable/utils';

// Static value - returns the value as-is
const staticValue = runIfFn('hello world');
console.log(staticValue); // 'hello world'

// Function with no arguments - executes the function
const dynamicValue = runIfFn(() => new Date().toISOString());
console.log(dynamicValue); // Current timestamp

// Function with arguments - executes with provided arguments
const sum = runIfFn((a: number, b: number) => a + b, 5, 10);
console.log(sum); // 15

// Complex example with conditional logic
const getConfig = (isDevelopment: boolean) => ({
  apiUrl: isDevelopment ? 'http://localhost:3000' : 'https://api.example.com',
  timeout: isDevelopment ? 5000 : 30000
});

const config = runIfFn(getConfig, true);
console.log(config); // { apiUrl: 'http://localhost:3000', timeout: 5000 }

Less Than Helper

The lessThan utility function provides a safe way to compare two values and determine if the first value is less than the second. It only performs the comparison if both values are valid numbers, returning false for any non-number inputs.

import { lessThan } from '@cacheable/utils';

// Basic number comparisons
console.log(lessThan(1, 2)); // true
console.log(lessThan(2, 1)); // false
console.log(lessThan(1, 1)); // false

// Works with negative numbers
console.log(lessThan(-1, 0)); // true
console.log(lessThan(-2, -1)); // true

// Works with decimal numbers
console.log(lessThan(1.5, 2.5)); // true
console.log(lessThan(2.7, 2.7)); // false

// Safe handling of non-number values
console.log(lessThan("1", 2)); // false
console.log(lessThan(1, "2")); // false
console.log(lessThan(null, 1)); // false
console.log(lessThan(undefined, 1)); // false
console.log(lessThan(NaN, 1)); // false

// Useful in filtering and sorting operations
const numbers = [5, 2, 8, 1, 9];
const lessThanFive = numbers.filter(n => lessThan(n, 5));
console.log(lessThanFive); // [2, 1]

// Safe comparison in conditional logic
function processValue(a?: number, b?: number) {
  if (lessThan(a, b)) {
    return `${a} is less than ${b}`;
  }
  return 'Invalid comparison or a >= b';
}

This utility is particularly useful when dealing with potentially undefined or invalid numeric values, ensuring type safety in comparison operations.

Is Object Helper

The isObject utility function provides a type-safe way to determine if a value is a plain object. It returns true for objects but false for arrays, null, functions, and primitive types. This function also serves as a TypeScript type guard.

import { isObject } from '@cacheable/utils';

// Basic object detection
console.log(isObject({})); // true
console.log(isObject({ name: 'John', age: 30 })); // true
console.log(isObject(Object.create(null))); // true

// Arrays are not considered objects
console.log(isObject([])); // false
console.log(isObject([1, 2, 3])); // false

// null is not considered an object (despite typeof null === 'object')
console.log(isObject(null)); // false

// Primitive types return false
console.log(isObject('string')); // false
console.log(isObject(123)); // false
console.log(isObject(true)); // false
console.log(isObject(undefined)); // false

// Functions return false
console.log(isObject(() => {})); // false
console.log(isObject(Date)); // false

// Built-in object types return true
console.log(isObject(new Date())); // true
console.log(isObject(/regex/)); // true
console.log(isObject(new Error('test'))); // true
console.log(isObject(new Map())); // true

// TypeScript type guard usage
function processValue(value: unknown) {
  if (isObject<{ name: string; age: number }>(value)) {
    // TypeScript now knows value is an object with name and age properties
    console.log(`Name: ${value.name}, Age: ${value.age}`);
  }
}

// Useful for configuration validation
function validateConfig(config: unknown) {
  if (!isObject(config)) {
    throw new Error('Configuration must be an object');
  }
  
  // Safe to access object properties
  return config;
}

// Filtering arrays for objects only
const mixedArray = [1, 'string', {}, [], null, { valid: true }];
const objectsOnly = mixedArray.filter(isObject);
console.log(objectsOnly); // [{}', { valid: true }]

This utility is particularly useful for:

Type validation - Ensuring values are objects before accessing properties
TypeScript type guarding - Narrowing types in conditional blocks
Configuration parsing - Validating that configuration values are objects
Data filtering - Separating objects from other data types

Wrap / Memoization for Sync and Async Functions

The @cacheable/utils package provides two main functions: wrap and wrapSync. These functions are used to memoize asynchronous and synchronous functions, respectively.

import { Cacheable } from 'cacheable';
const asyncFunction = async (value: number) => {
  return Math.random() * value;
};

const cache = new Cacheable();
const options = {
  ttl: '1h', // 1 hour
  keyPrefix: 'p1', // key prefix. This is used if you have multiple functions and need to set a unique prefix.
  cache,
}
const wrappedFunction = wrap(asyncFunction, options);
console.log(await wrappedFunction(2)); // 4
console.log(await wrappedFunction(2)); // 4 from cache

With wrap we have also included stampede protection so that a Promise based call will only be called once if multiple requests of the same are executed at the same time. Here is an example of how to test for stampede protection:

import { Cacheable } from 'cacheable';
const asyncFunction = async (value: number) => {
  return value;
};

const cache = new Cacheable();
const options = {
  ttl: '1h', // 1 hour
  keyPrefix: 'p1', // key prefix. This is used if you have multiple functions and need to set a unique prefix.
  cache,
}

const wrappedFunction = wrap(asyncFunction, options);
const promises = [];
for (let i = 0; i < 10; i++) {
  promises.push(wrappedFunction(i));
}

const results = await Promise.all(promises); // all results should be the same

console.log(results); // [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]

In this example we are wrapping an async function in a cache with a ttl of 1 hour. This will cache the result of the function for 1 hour and then expire the value. You can also wrap a sync function in a cache:

import { CacheableMemory } from 'cacheable';
const syncFunction = (value: number) => {
  return value * 2;
};

const cache = new CacheableMemory();
const wrappedFunction = wrap(syncFunction, { ttl: '1h', key: 'syncFunction', cache });
console.log(wrappedFunction(2)); // 4
console.log(wrappedFunction(2)); // 4 from cache

In this example we are wrapping a sync function in a cache with a ttl of 1 hour. This will cache the result of the function for 1 hour and then expire the value. You can also set the key property in the wrap() options to set a custom key for the cache.

When an error occurs in the function it will not cache the value and will return the error. This is useful if you want to cache the results of a function but not cache the error. If you want it to cache the error you can set the cacheError property to true in the wrap() options. This is disabled by default.

import { CacheableMemory } from 'cacheable';
const syncFunction = (value: number) => {
  throw new Error('error');
};

const cache = new CacheableMemory();
const wrappedFunction = wrap(syncFunction, { ttl: '1h', key: 'syncFunction', cacheError: true, cache });
console.log(wrappedFunction()); // error
console.log(wrappedFunction()); // error from cache

If you would like to generate your own key for the wrapped function you can set the createKey property in the wrap() options. This is useful if you want to generate a key based on the arguments of the function or any other criteria.

  const cache = new Cacheable();
  const options: WrapOptions = {
    cache,
    keyPrefix: 'test',
    createKey: (function_, arguments_, options: WrapOptions) => `customKey:${options?.keyPrefix}:${arguments_[0]}`,
  };

  const wrapped = wrap((argument: string) => `Result for ${argument}`, options);

  const result1 = await wrapped('arg1');
  const result2 = await wrapped('arg1'); // Should hit the cache

  console.log(result1); // Result for arg1
  console.log(result2); // Result for arg1 (from cache)

We will pass in the function that is being wrapped, the arguments passed to the function, and the options used to wrap the function. You can then use these to generate a custom key for the cache.

Get Or Set Memoization Function

The getOrSet method provides a convenient way to implement the cache-aside pattern. It attempts to retrieve a value from cache, and if not found, calls the provided function to compute the value and store it in cache before returning it. Here are the options:

export type GetOrSetFunctionOptions = {
	ttl?: number | string;
	cacheErrors?: boolean;
	throwErrors?: boolean;
	nonBlocking?: boolean;
};

The nonBlocking option allows you to override the instance-level nonBlocking setting for the get call within getOrSet. When set to false, the get will block and wait for a response from the secondary store before deciding whether to call the provided function. When set to true, the primary store returns immediately and syncs from secondary in the background.

Here is an example of how to use the getOrSet method:

import { Cacheable } from 'cacheable';
const cache = new Cacheable();
// Use getOrSet to fetch user data
const function_ = async () => Math.random() * 100;
const value = await getOrSet('randomValue', function_, { ttl: '1h', cache });
console.log(value); // e.g. 42.123456789

You can also use a function to compute the key for the function:

import { Cacheable, GetOrSetOptions } from 'cacheable';
const cache = new Cacheable();

// Function to generate a key based on options
const generateKey = (options?: GetOrSetOptions) => {
  return `custom_key_:${options?.cacheId || 'default'}`;
};

const function_ = async () => Math.random() * 100;
const value = await getOrSet(generateKey(), function_, { ttl: '1h', cache });

Cache Tags

The CacheTags service provides tag-based invalidation on top of any Keyv store. It is store-agnostic and does not require any adapter changes.

The service uses a lazy invalidation model. Instead of scanning and deleting keys, invalidateTag increments a per-tag version counter. Each cached key stores a snapshot of its tag versions at the time it was written, and isKeyFresh compares that snapshot to the current versions. If any tag version has been incremented since the snapshot was taken, the key is considered stale. Stale entries are not deleted explicitly and are expected to fall out of the cache via their TTL.

This approach keeps invalidation constant-time regardless of how many keys reference a tag. The trade-off is one additional isKeyFresh read per cache lookup.

import { Keyv } from 'keyv';
import { CacheTags } from '@cacheable/utils';

const store = new Keyv();
const cacheTags = new CacheTags({ store, namespace: 'app' });

await cacheTags.setKeyTags('user:42', ['users', 'org:7'], { ttl: 3600000 });
console.log(await cacheTags.isKeyFresh('user:42')); // true

await cacheTags.invalidateTag('users');
console.log(await cacheTags.isKeyFresh('user:42')); // false

The recommended pattern is to call isKeyFresh before trusting a value returned from your cache, and to refresh the tag snapshot whenever you write a new value:

import { Cacheable } from 'cacheable';
import { Keyv } from 'keyv';
import { CacheTags } from '@cacheable/utils';

const cache = new Cacheable();
const cacheTags = new CacheTags({ store: new Keyv() });

const getUser = async (id: string) => {
  const key = `user:${id}`;

  if (await cacheTags.isKeyFresh(key)) {
    const cached = await cache.get(key);
    if (cached !== undefined) {
      return cached;
    }
  }

  const fresh = await loadUser(id);
  await cache.set(key, fresh, '1h');
  await cacheTags.setKeyTags(key, ['users', `org:${fresh.orgId}`], { ttl: 3600000 });
  return fresh;
};

You can invalidate one or many tags at a time. Both methods return the names of the tags that were bumped:

const bumped = await cacheTags.invalidateTags(['users', 'org:7']);
console.log(bumped); // ['users', 'org:7']

When integrating with a cache where most keys are untagged, use isKeyStale instead of isKeyFresh. It only reports true when a snapshot exists for the key and one of its tags has been invalidated, so keys that were never tagged are not treated as stale:

console.log(await cacheTags.isKeyStale('never-tagged')); // false
await cacheTags.setKeyTags('user:42', ['users']);
console.log(await cacheTags.isKeyStale('user:42')); // false
await cacheTags.invalidateTag('users');
console.log(await cacheTags.isKeyStale('user:42')); // true

The getStaleKeys method checks many keys at once using two batched store reads regardless of how many keys are passed — one for the snapshots and one for the union of their tag versions:

await cacheTags.setKeyTags('a', ['x']);
await cacheTags.setKeyTags('b', ['y']);
await cacheTags.invalidateTag('x');
console.log(await cacheTags.getStaleKeys(['a', 'b', 'untagged'])); // ['a']

The getTags method returns the tags currently associated with a key, or undefined if the key has no snapshot:

await cacheTags.setKeyTags('user:42', ['users', 'org:7']);
console.log(await cacheTags.getTags('user:42')); // ['users', 'org:7']
console.log(await cacheTags.getTags('missing')); // undefined

The removeKey and removeKeys methods delete tag snapshots when the cached values themselves are deleted. removeKeys performs a single batched delete:

await cacheTags.removeKeys(['user:1', 'user:2']);

The service can be disabled via the enabled option or property so integrations pay no extra store reads for untagged workloads. While disabled, every method is a no-op: read methods return their neutral value (isKeyFresh returns true, isKeyStale returns false, getStaleKeys returns [], and so on) and writes are skipped. The service never enables itself — you have to turn it on explicitly, which keeps behavior consistent across distributed instances sharing a store:

const cacheTags = new CacheTags({ store, enabled: false });
console.log(await cacheTags.isKeyStale('anything')); // false, no store read
await cacheTags.setKeyTags('user:42', ['users']); // no-op while disabled
cacheTags.enabled = true; // turn it on to use tags

setKeyTags, removeKey, and removeKeys accept a nonBlocking option to fire-and-forget the store write. Failures from non-blocking operations are reported to the onError constructor option since they cannot be thrown to the caller:

const cacheTags = new CacheTags({ store, onError: (error) => console.error(error) });
await cacheTags.setKeyTags('user:42', ['users'], { ttl: 3600000, nonBlocking: true });

The getKeysByTag method returns the keys currently referencing a given tag. It iterates the Keyv namespace and is therefore an O(N) operation. It is intended for debugging and tests rather than hot paths.

await cacheTags.setKeyTags('user:1', ['users']);
await cacheTags.setKeyTags('user:2', ['users']);
const keys = await cacheTags.getKeysByTag('users');
console.log(keys); // ['user:1', 'user:2']

The service stores its metadata under a reserved prefix so that it cannot collide with user keys:

--cacheable--tags--::tag:  → integer version counter
--cacheable--tags--::key:  → { tags: { [tag]: versionAtSetTime } }

Tag version counters are stored without a TTL because they must outlive any key that references them. Key entries respect the ttl passed to setKeyTags, which should be set to match the TTL of the cached value it tracks.

The namespace defaults to default and can be set via the constructor. Two services configured with different namespaces can share the same store without seeing each other's tags or keys.

The read-version then write-snapshot sequence in setKeyTags is not atomic across processes. A concurrent invalidateTag that runs between the read and the write can leave a freshly written key referencing a stale version. An atomic Redis fast path using MULTI or Lua is a planned future enhancement.

How to Contribute

You can contribute by forking the repo and submitting a pull request. Please make sure to add tests and update the documentation. To learn more about how to contribute go to our main README https://github.com/jaredwray/cacheable. This will talk about how to Open a Pull Request, Ask a Question, or Post an Issue.

Table of Contents