solid-suggest is a UI component for SolidJS developers that renders a text input with dropdown suggestions. It can be used in scenarios such as search suggestions or results triggered by text input.
It's a super simple library without many batteries included. See following sections for what it can and can't do for you.
Because of its simplicity, I also expect it to be usable even without updates for years.
Install from the npm registry using npm
or your favorite alternate package manager (yarn, pnpm).
npm install solid-suggest --save
Import into your SolidJS project.
import Suggest from "solid-suggest";
const items = [
{ name: "Scout", class: "Offense" },
{ name: "Medic", class: "Support" },
{ name: "Heavy", class: "Defense" }
];
<Suggest
renderSuggestion={item => `${item.name} (${item.class})`}
onQuery={query =>
items.filter(
item =>
item.name.toLowerCase().includes(query.toLowerCase()) ||
item.class.toLowerCase().includes(query.toLowerCase())
)
}
onSelect={item => setSelected(item.name)}
/>
See more in the docs.
Exposes two main interfaces as functions that you implement:
onQuery
- solid-suggest runs it on initial render and on each user input change. For solid-suggest, your implementation answers the question "what suggestions should solid-suggest show in the current state?"onSelect
- solid-suggest invokes this to emit the user's selection to your code.
All this library does is provide JavaScript/TypeScript bones for a suggestions dropdown so that you don't have to clutter your application with the required internal states, keyboard and mouse handling, etc.
- Supports objects as suggestions (not just strings), with support for custom rendering of each suggestion.
- Supports optional debouncing of user input via the
debounceMs
prop (in milliseconds). If not set or zero, input is not debounced. - Styling is still supplied by you.
Some more questions are answered in the docs.
- Come with any styling. solid-suggest is a headless library that completely relies on the you to style. This way, it fully integrates with any branding and design system.
- Provide built-in network fetch capabilities. You are free to make network calls this in their
onQuery
implementations. Consider supplying adebounceMs
value to reduce network calls. - Wash your car.
Some more questions are answered in the docs.
Find documentation and tips here.
npm install
... before anything else.
npm run devOnly
... runs a continuous watcher for changes to library code that transpiles TypeScript to a dist/ dir at project top level.
npm run devDemo
... runs a local server that you can use to see and iterate on the demo page at http://localhost:8000/docs/index.htm. This is a basic HTML page that imports some libraries from CDNs. Hot reload/replacement is not enabled, so it needs manual refreshing.
npm run devWithDemo
... runs both of the above at the same time.