cacheable-request

cacheable-request

Wrap native HTTP requests with RFC compliant cache support

RFC 7234 compliant HTTP caching for native Node.js HTTP/HTTPS requests. Caching works out of the box in memory or is easily pluggable with a wide range of storage adapters.

Note: This is a low level wrapper around the core HTTP modules, it's not a high level request library.

Features

• Only stores cacheable responses as defined by RFC 7234
• Fresh cache entries are served directly from cache
• Stale cache entries are revalidated with If-None-Match/If-Modified-Since headers
• 304 responses from revalidation requests use cached body
• Updates Age header on cached responses
• Can completely bypass cache on a per request basis
• In memory cache by default
• Official support for Redis, MongoDB, SQLite, PostgreSQL and MySQL storage adapters
• Easily plug in your own or third-party storage adapters
• If DB connection fails, cache is automatically bypassed (disabled by default)
• Adds cache support to any existing HTTP code with minimal changes
• Uses http-cache-semantics internally for HTTP RFC 7234 compliance

Install

npm install cacheable-request

Usage

const http = require('http');
const CacheableRequest = require('cacheable-request');

// Then instead of
const req = http.request('http://example.com', cb);
req.end();

// You can do
const cacheableRequest = new CacheableRequest(http.request);
const cacheReq = cacheableRequest('http://example.com', cb);
cacheReq.on('request', req => req.end());
// Future requests to 'example.com' will be returned from cache if still valid

// You pass in any other http.request API compatible method to be wrapped with cache support:
const cacheableRequest = new CacheableRequest(https.request);
const cacheableRequest = new CacheableRequest(electron.net);

Storage Adapters

cacheable-request uses Keyv to support a wide range of storage adapters.

For example, to use Redis as a cache backend, you just need to install the official Redis Keyv storage adapter:

npm install @keyv/redis

And then you can pass CacheableRequest your connection string:

const cacheableRequest = new CacheableRequest(http.request, 'redis://user:pass@localhost:6379');

View all official Keyv storage adapters.

Keyv also supports anything that follows the Map API so it's easy to write your own storage adapter or use a third-party solution.

e.g The following are all valid storage adapters

const storageAdapter = new Map();
// or
const storageAdapter = require('./my-storage-adapter');
// or
const QuickLRU = require('quick-lru');
const storageAdapter = new QuickLRU({ maxSize: 1000 });

const cacheableRequest = new CacheableRequest(http.request, storageAdapter);

View the Keyv docs for more information on how to use storage adapters.

API

new cacheableRequest(request, [storageAdapter])

Returns the provided request function wrapped with cache support.

request

Type: function

Request function to wrap with cache support. Should be http.request or a similar API compatible request function.

storageAdapter

Type: Keyv storage adapter
Default: new Map()

A Keyv storage adapter instance, or connection string if using with an official Keyv storage adapter.

Instance

cacheableRequest(opts, [cb])

Returns an event emitter.

opts

Type: object, string

• Any of the default request functions options.
• Any http-cache-semantics options.
• Any of the following:

opts.cache

Type: boolean
Default: true

If the cache should be used. Setting this to false will completely bypass the cache for the current request.

opts.strictTtl

Type: boolean
Default: false

If set to true once a cached resource has expired it is deleted and will have to be re-requested.

If set to false (default), after a cached resource's TTL expires it is kept in the cache and will be revalidated on the next request with If-None-Match/If-Modified-Since headers.

opts.maxTtl

Type: number
Default: undefined

Limits TTL. The number represents milliseconds.

opts.automaticFailover

Type: boolean
Default: false

When set to true, if the DB connection fails we will automatically fallback to a network request. DB errors will still be emitted to notify you of the problem even though the request callback may succeed.

opts.forceRefresh

Type: boolean
Default: false

Forces refreshing the cache. If the response could be retrieved from the cache, it will perform a new request and override the cache instead.

cb

Type: function

The callback function which will receive the response as an argument.

The response can be either a Node.js HTTP response stream or a responselike object. The response will also have a fromCache property set with a boolean value.

.on('request', request)

request event to get the request object of the request.

Note: This event will only fire if an HTTP request is actually made, not when a response is retrieved from cache. However, you should always handle the request event to end the request and handle any potential request errors.

.on('response', response)

response event to get the response object from the HTTP request or cache.

.on('error', error)

error event emitted in case of an error with the cache.

Errors emitted here will be an instance of CacheableRequest.RequestError or CacheableRequest.CacheError. You will only ever receive a RequestError if the request function throws (normally caused by invalid user input). Normal request errors should be handled inside the request event.

To properly handle all error scenarios you should use the following pattern:

cacheableRequest('example.com', cb)
  .on('error', err => {
    if (err instanceof CacheableRequest.CacheError) {
      handleCacheError(err); // Cache error
    } else if (err instanceof CacheableRequest.RequestError) {
      handleRequestError(err); // Request function thrown
    }
  })
  .on('request', req => {
    req.on('error', handleRequestError); // Request error emitted
    req.end();
  });

Note: Database connection errors are emitted here, however cacheable-request will attempt to re-request the resource and bypass the cache on a connection error. Therefore a database connection error doesn't necessarily mean the request won't be fulfilled.

License

MIT © Luke Childs