Minimalist router for React apps
- Concise API
- Unopinionated route structure: routes are not necessarily hierarchical or otherwise tightly coupled
- Same route matching for components and prop values
- Decoupled from data loading: the router only manages routing
- Consistent with native APIs:
- route links are similar to HTML links
- route navigation interface is similar to
window.location
- SSR-compatible
Installation: npm i routescape
The route link components <A>
and <Area>
enabling SPA navigation have the same props as their HTML counterparts: <a>
and <area>
. Apart from reducing some cognitive load, this allows to quickly migrate from plain HTML links to route links (or the other way around).
import {A, useRoute} from 'routescape';
let App = () => {
let [route, withRoute] = useRoute();
return (
<>
<nav>
<A href="/intro" className={withRoute('/intro', 'active')}>
Intro
</A>
</nav>
{withRoute('/intro', (
<main>
<h1>Intro</h1>
</main>
))}
</>
);
};
The functional route matching by means of withRoute()
offers a simple and consistent way to render both components and props based on the current location.
Note that both the intro link's className
and <main>
are rendered in a similar fashion using the same route-matching function. withRoute('/intro', x)
returns x
only if the current location is /intro
.
(With the component-based route matching adopted by some routers, conditionally rendering a component and marking a link as active via its props have to be handled differently.)
Similarly to the ternary operator condition ? x : y
(often seen with the general conditional rendering pattern), withRoute()
accepts a fallback value as the optional third parameter: withRoute(routePattern, x, y)
.
let Nav = () => {
let [route, withRoute] = useRoute();
return (
<nav>
<A
href="/intro"
className={withRoute('/intro', 'active', 'inactive')}
>
Intro
</A>
</nav>
);
};
In the example above, the link is marked as active
if the current location is /intro
, and inactive
otherwise.
With the third parameter omitted, withRoute('/intro', 'active')
results in undefined
with locations other than /intro
(since the missing fallback parameter is effectively undefined
), which is perfectly fine as well.
By default, after the link navigation occurs, the user can navigate back by pressing the browser's back button. Optionally, by setting data-navigation-mode="replace"
a link component can be configured to replace the navigation history entry, which will prevent the user from returning to the previous location by clicking the browser's back button.
withRoute()
accepts route patterns of various types: string | RegExp | (string | RegExp)[]
. The parameters of a regular expression route pattern (or of the first match in the array) are passed to the second and the third parameter of withRoute()
if they are functions.
let App = () => {
let [, withRoute] = useRoute();
return (
<>
<nav>
<A href="/intro">Intro</A>
</nav>
{withRoute(/^\/section\/(?<id>\d+)\/?$/, ({id}) => (
<main>
<h1>Section #{id}</h1>
</main>
))}
</>
);
};
The fallback parameter of withRoute()
is also a way to handle unknown routes:
const routeMap = {
intro: '/intro',
sections: /^\/section\/(?<id>\d+)\/?$/,
};
const knownRoutes = Object.values(routeMap);
let App = () => {
let [, withRoute] = useRoute();
return (
<>
<nav>
<A href={routeMap.intro}>Intro</A>
</nav>
{withRoute(routeMap.intro, (
<main>
<h1>Intro</h1>
</main>
))}
{withRoute(routeMap.sections, ({id}) => (
<main>
<h1>Section #{id}</h1>
</main>
))}
{withRoute(knownRoutes, null, (
<main className="error">
<h1>404 Not found</h1>
</main>
))}
</>
);
};
Note that the last withRoute()
results in null
(that is no content) for all known routes and renders the error content for the rest unknown routes.
Although the routes are grouped together in the example above, that's not a requirement. withRoute()
calls are not coupled together, they can be split across separate components and files and arranged in any order (like any other conditionally rendered components).
To jump to another route programmatically, there's the route
object returned from the useRoute()
hook:
let ProfileButton = ({signedIn}) => {
let [route] = useRoute();
let handleClick = () => {
route.assign(signedIn ? '/profile' : '/login');
};
return <button onClick={handleClick}>Profile</button>;
};
This particular example is somewhat contrived since it could have been composed in a declarative fashion using the route link component <A>
. Still, it demonstrates how the route
object can be used in use cases where the imperative navigation is the only reasonable way to go.
The interface of the route
object consists of the following parts:
- SPA navigation via the History API:
.assign()
,.replace()
,.reload()
, and readonly properties:.href
,.pathname
,.search
,.hash
, semantically similar towindow.location
;.back()
,.forward()
,.go(delta)
, corresponding to thehistory
methods;
- route matching:
.matches(value)
, checking whether the current location matches the givenvalue
;.match(value)
, returning matched parameters if the givenvalue
is a regular expression andnull
if the current location doesn't match thevalue
.
Server-side rendering and unit tests are the examples of the environments lacking a global location (such as window.location
). They are the prime use cases for the location provider, <Router>
.
Let's consider an express application route as an example:
import {renderToString} from 'react-dom/server';
import {Router} from 'routescape';
app.get('/', (req, res) => {
let html = renderToString(
<Router location={req.originalUrl}>
<App/>
</Router>,
);
res.send(html);
});
The value passed to the router's location
prop can be accessed via the useRoute()
hook:
let [route, withRoute] = useRoute();
console.log(route.href); // returns the router's `location`
Both route
and withRoute()
returned from useRoute()
operate based on the router's location
.
<Router>
can be used with client-side rendering as well. In most cases, it is unnecessary since by default the route context takes the global location from window.location
if it's available.
The location provider component <Router>
can be used to redefine the route matching behavior.
import {NavigationLocation, getPath, Router} from 'routescape';
export class PathLocation extends NavigationLocation {
getHref(location) {
// disregard `search` and `hash`
return getPath(location, {search: false, hash: false});
}
}
let App = () => (
<Router location={new PathLocation(url)}>
<AppContent/>
</Router>
);
By default, routing relies on the entire URL. In this example, we've redefined this behavior to disregard the search
and hash
portions of the URL.
Extending the NavigationLocation
class gives plenty of room for customization. This approach allows in fact to go beyond the URL-based routing altogether.
A static chunk of HTML content is an example where the route link component <A>
can't be directly used but it still might be desirable to convert plain HTML links to SPA route links. The useRouteLinks()
hook can be helpful here:
import {useRef} from 'react';
import {useRouteLinks} from 'routescape';
let Content = ({value}) => {
let containerRef = useRef(null);
useRouteLinks(containerRef, 'a');
return (
<div ref={containerRef}>
{value}
</div>
);
};
In this example, the useRouteLinks()
hook converts all links matching the selector 'a'
inside the container referenced by containerRef
to SPA route links.