Preact Fetching

Preact hooks for asynchronous data fetching.

Features

• Promise-based data fetching. Unopinionated about the source of your data.
• Small bundle size. Less than 0.5kb minified and gzipped.
• TypeScript support. Respects the resolved type of your fetcher implementation.
• Configurable global cache. All components share the same data by distinct keys.
• Conditional data fetching. Supports advanced use-cases like form submission.

Example

import { useQuery } from 'preact-fetching';

function GitHubStars({ owner, repo }) {
	const { isLoading, isError, error, data } = useQuery(`${owner}/${repo}`, async () => {
		const response = await fetch(`https://api.github.com/repos/${owner}/${repo}`);
		const data = await response.json();
		return data.stargazers_count;
	});

	if (isError) {
		return `An error occurred! ${error.message}`;
	}

	if (isLoading) {
		return 'Loading...';
	}

	return `Count: ${data}`;
}

Installation

Install using NPM or Yarn:

npm install preact-fetching

yarn add preact-fetching

Usage

useQuery accepts two arguments:

• A key to uniquely identify the data being fetched.
• A function which fetches the data and returns either a promise or the resulting data.

Fetching

The fetching function will be called any time a component is mounted and there is not already a fetch in progress for that key. Requests are deduplicated if many components are mounted at the same time which reference the same data. Data is refetched if when future components are mounted using that same key, during which time the stale data will still be available.

Cache

The default cache behavior uses a simple Map object to reference cached values.

Cache Key

Because the default cache behavior uses a simple Map object, a key must be strictly equal on subsequent renders for a cache hit to occur (for example, using a string). For array or object values, you can consider one of the following:

• Using the experimental records and tuples language feature.
• Changing the cache to one which implements the Map interface, but supports deep object equivalency (for example, EquivalentKeyMap).

Conditional Fetching

To prevent the fetch behavior based on some condition, pass a nullish (null or undefined) value as the key. If a nullish value is passed as the key, fetching will not occur.

API

useQuery

function useQuery<Data>(
	key: any,
	fetcher: () => Data | Promise<Data>,
): LoadingResult<Data> | SuccessResult<Data> | ErrorResult<Data> | IdleResult<Data>;

Triggers a new fetch request as appropriate and returns a result of the current status.

CacheContext

let CacheContext: import('preact').Context<
	MapLike<
		any,
		CacheLoadingEntry<any> | CacheSuccessEntry<any> | CacheErrorEntry<any> | CacheIdleEntry<any>
	>
>;

Context serving as cache state container. For most usage, you shouldn't need to interface with the context object, but in advanced use-cases you can use this to substitute or scope caches.

Project Goals and Non-Goals

Goals:

• Lack of configurability as a feature, preferring smart defaults and a minimal-but-flexible API over a multitude of settings.
• Efficiency in bundle size, performance, and cache invalidations. Micro-optimizations and code golf are welcome.

Non-goals:

• Feature parity with similar libraries (swr, react-query, etc.).

License

Copyright 2024 Andrew Duthie

Released under the MIT License. See LICENSE.md.