Routing
Astro uses file-based routing to generate your build URLs based on the file layout of your project src/pages/ directory.
Navigating between pages
Astro uses standard HTML <a> elements to navigate between routes. There is no framework-specific <Link> component provided.
<p>Read more <a href="/about/">about</a> Astro!</p>
Static routes
.astro page components as well as Markdown and MDX Files (.md, .mdx) within the src/pages/ directory automatically become pages on your website. Each page’s route corresponds to its path and filename within the src/pages/ directory.
# Example: Static routes
src/pages/index.astro -> mysite.com/
src/pages/about.astro -> mysite.com/about
src/pages/about/index.astro -> mysite.com/about
src/pages/about/me.astro -> mysite.com/about/me
src/pages/posts/1.md -> mysite.com/posts/1
:::tip
There is no separate “routing config” to maintain in an Astro project! When you add a file to the src/pages/ directory, a new route is automatically created for you. In static builds, you can customize the file output format using the build.format configuration option.
:::
Dynamic routes
An Astro page file can specify dynamic route parameters in its filename to generate multiple, matching pages. For example, src/pages/authors/[author].astro generates a bio page for every author on your blog. author becomes a parameter that you can access from inside the page.
In Astro’s default static output mode, these pages are generated at build time, and so you must predetermine the list of authors that get a corresponding file. In SSR mode, a page will be generated on request for any route that matches.
Static (SSG) Mode
Because all routes must be determined at build time, a dynamic route must export a getStaticPaths() that returns an array of objects with a params property. Each of these objects will generate a corresponding route.
[dog].astro defines the dynamic dog parameter in its filename, so the objects returned by getStaticPaths() must include dog in their params. The page can then access this parameter using Astro.params.
---
export function getStaticPaths() {
return [
{ params: { dog: "clifford" } },
{ params: { dog: "rover" } },
{ params: { dog: "spot" } },
];
}
const { dog } = Astro.params;
---
<div>Good dog, {dog}!</div>
This will generate three pages: /dogs/clifford, /dogs/rover, and /dogs/spot, each displaying the corresponding dog name.
The filename can include multiple parameters, which must all be included in the params objects in getStaticPaths():
---
export function getStaticPaths() {
return [
{ params: { lang: "en", version: "v1" } },
{ params: { lang: "fr", version: "v2" } },
];
}
const { lang, version } = Astro.params;
---
...
This will generate /en-v1/info and /fr-v2/info.
Parameters can be included in separate parts of the path. For example, the file src/pages/[lang]/[version]/info.astro with the same getStaticPaths() above will generate the routes /en/v1/info and /fr/v2/info.
📚 Learn more about getStaticPaths().
<RecipeLinks slugs={[“en/recipes/i18n”]} />
Rest parameters
If you need more flexibility in your URL routing, you can use a rest parameter ([...path]) in your .astro filename to match file paths of any depth:
---
export function getStaticPaths() {
return [
{ params: { path: "one/two/three" } },
{ params: { path: "four" } },
{ params: { path: undefined } },
];
}
const { path } = Astro.params;
---
...
This will generate /sequences/one/two/three, /sequences/four, and /sequences. (Setting the rest parameter to undefined allows it to match the top level page.)
Rest parameters can be used with other named parameters. For example, GitHub’s file viewer can be represented with the following dynamic route:
/[org]/[repo]/tree/[branch]/[...file]
In this example, a request for /withastro/astro/tree/main/docs/public/favicon.svg would be split into the following named parameters:
{
org: 'withastro',
repo: 'astro',
branch: 'main',
file: 'docs/public/favicon.svg'
}
Example: Dynamic pages at multiple levels
In the following example, a rest parameter ([...slug]) and the props feature of getStaticPaths() generate pages for slugs of different depths.
---
export async function getStaticPaths() {
const pages = [
{
slug: undefined,
title: "Astro Store",
text: "Welcome to the Astro store!",
},
{
slug: "products",
title: "Astro products",
text: "We have lots of products for you",
},
{
slug: "products/astro-handbook",
title: "The ultimate Astro handbook",
text: "If you want to learn Astro, you must read this book.",
},
];
return pages.map(({ slug, title, text }) => {
return {
params: { slug },
props: { title, text },
};
});
}
const { title, text } = Astro.props;
---
<html>
<head>
<title>{title}</title>
</head>
<body>
<h1>{title}</h1>
<p>{text}</p>
</body>
</html>
Server (SSR) Mode
In SSR mode, dynamic routes are defined the same way: include [param] or [...path] brackets in your file names to match arbitrary strings or paths. But because the routes are no longer built ahead of time, the page will be served to any matching route. Since these are not “static” routes, getStaticPaths should not be used.
---
const { resource, id } = Astro.params;
---
<h1>{resource}: {id}</h1>
This page will be served for any value of resource and id: resources/users/1, resources/colors/blue, etc.
Modifying the [...slug] example for SSR
Because SSR pages can’t use getStaticPaths(), they can’t receive props. The previous example can be adapted for SSR mode by looking up the value of the slug param in an object. If the route is at the root (”/”), the slug param will be undefined. If the value doesn’t exist in the object, we redirect to a 404 page.
---
const pages = [
{
slug: undefined,
title: "Astro Store",
text: "Welcome to the Astro store!",
},
{
slug: "products",
title: "Astro products",
text: "We have lots of products for you",
},
{
slug: "products/astro-handbook",
title: "The ultimate Astro handbook",
text: "If you want to learn Astro, you must read this book.",
},
];
const { slug } = Astro.params;
const page = pages.find((page) => page.slug === slug);
if (!page) return Astro.redirect("/404");
const { title, text } = page;
---
<html>
<head>
<title>{title}</title>
</head>
<body>
<h1>{title}</h1>
<p>{text}</p>
</body>
</html>
Redirects
Sometimes you will need to redirect your readers to a new page, either permanently because your site structure has changed or in response to an action such as logging in to an authenticated route.
You can define rules to redirect users to permanently-moved pages in your Astro config. Or, redirect users dynamically as they use your site.
Configured Redirects
You can specify a mapping of permanent redirects in your Astro config with the redirects value. For most redirects, this is a mapping of an old route to the new route:
import { defineConfig } from "astro/config";
export default defineConfig({
redirects: {
"/old-page": "/new-page",
},
});
These redirects follow the same rules as file-based routes. Dynamic routes are allowed as long as both the new and old routes contain the same parameters, for example:
{
"/blog/[...slug]": "/articles/[...slug]"
}
Using SSR or a static adapter, you can also provide an object as the value, allowing you to specify the status code in addition to the new destination:
import { defineConfig } from "astro/config";
export default defineConfig({
redirects: {
"/old-page": {
status: 302,
destination: "/new-page",
},
},
});
When running astro build, Astro will output HTML files with the meta refresh tag by default. Supported adapters will instead write out the host’s configuration file with the redirects.
The status code is 301 by default. If building to HTML files the status code is not used by the server.
Dynamic redirects
On the Astro global, the Astro.redirect method allows you to redirect to another page dynamically. You might do this after checking if the user is logged in by getting their session from a cookie.
---
import { isLoggedIn } from "../utils";
const cookie = Astro.request.headers.get("cookie");
// If the user is not logged in, redirect them to the login page
if (!isLoggedIn(cookie)) {
return Astro.redirect("/login");
}
---
<html>
<!-- Page here... -->
</html>
Route Priority Order
It’s possible for multiple routes to match the same URL path. For example each of these routes would match /posts/create:
Astro needs to know which route should be used to build the page. To do so, it sorts them according to the following rules:
- Static routes without path parameters will take precedence over all other routes
- Dynamic routes using named parameters take precedence over rest parameters
- Pre-rendered dynamic routes take precedence over server dynamic routes
- Rest parameters have the lowest priority
- Endpoints always take precedence over pages
- Ties are resolved alphabetically
Given the example above, here are a few examples of how the rules will match a requested URL to the route used to build the HTML:
pages/posts/create.astro- Will build/posts/createpages/posts/[pid].astro- Will build/posts/1,/posts/abc, etc. But not/posts/createpages/posts/[...slug].astro- Will build/posts/1/2,/posts/a/b/c, etc. But not/posts/create,/posts/1,/posts/abc
Redirects also follow the same rules, but are prioritized last; if there is a file-based route and a redirect with the same route priority level, the file-based route is chosen.
Pagination
Astro supports built-in pagination for large collections of data that need to be split into multiple pages. Astro will generate common pagination properties, including previous/next page URLs, total number of pages, and more.
Paginated route names should use the same [bracket] syntax as a standard dynamic route. For instance, the file name /astronauts/[page].astro will generate routes for /astronauts/1, /astronauts/2, etc, where [page] is the generated page number.
You can use the paginate() function to generate these pages for an array of values like so:
---
// src/pages/astronauts/[page].astro
export async function getStaticPaths({ paginate }) {
const astronautPages = [
{
astronaut: "Neil Armstrong",
},
{
astronaut: "Buzz Aldrin",
},
{
astronaut: "Sally Ride",
},
{
astronaut: "John Glenn",
},
];
// Generate pages from our array of astronauts, with 2 to a page
return paginate(astronautPages, { pageSize: 2 });
}
// All paginated data is passed on the "page" prop
const { page } = Astro.props;
---
<!--Display the current page number. Astro.params.page can also be used!-->
<h1>Page {page.currentPage}</h1>
<ul>
<!--List the array of astronaut info-->
{page.data.map(({ astronaut }) => <li>{astronaut}</li>)}
</ul>
This generates the following pages, with 2 items to a page:
/astronauts/1- Page 1: Displays “Neil Armstrong” and “Buzz Aldrin”/astronauts/2- Page 2: Displays “Sally Ride” and “John Glenn”
The page prop
When you use the paginate() function, each page will be passed its data via a page prop. The page prop has many useful properties, but here are the highlights:
- page.data - array containing the page’s slice of data that you passed to the
paginate()function - page.url.next - link to the next page in the set
- page.url.prev - link to the previous page in the set
---
// src/pages/astronauts/[page].astro
// Paginate same list of { astronaut } objects as the previous example
export async function getStaticPaths({ paginate }) {
/* ... */
}
const { page } = Astro.props;
---
<h1>Page {page.currentPage}</h1>
<ul>
{page.data.map(({ astronaut }) => <li>{astronaut}</li>)}
</ul>
{page.url.prev ? <a href={page.url.prev}>Previous</a> : null}
{page.url.next ? <a href={page.url.next}>Next</a> : null}
Complete API reference
interface Page<T = any> {
/** result */
data: T[];
/** metadata */
/** the count of the first item on the page, starting from 0 */
start: number;
/** the count of the last item on the page, starting from 0 */
end: number;
/** total number of results */
total: number;
/** the current page number, starting from 1 */
currentPage: number;
/** number of items per page (default: 25) */
size: number;
/** number of last page */
lastPage: number;
url: {
/** url of the current page */
current: string;
/** url of the previous page (if there is one) */
prev: string | undefined;
/** url of the next page (if there is one) */
next: string | undefined;
};
}
Nested Pagination
A more advanced use-case for pagination is nested pagination. This is when pagination is combined with other dynamic route params. You can use nested pagination to group your paginated collection by some property or tag.
For example, if you want to group your paginated Markdown posts by some tag, you would use nested pagination by creating a /src/pages/[tag]/[page].astro page that would match the following URLS:
/red/1(tag=red)/red/2(tag=red)/blue/1(tag=blue)/green/1(tag=green)
Nested pagination works by returning an array of paginate() results from getStaticPaths(), one for each grouping.
In the following example, we will implement nested pagination to build the URLs listed above:
---
// src/pages/[tag]/[page].astro
export async function getStaticPaths({ paginate }) {
const allTags = ["red", "blue", "green"];
const allPosts = await Astro.glob("../../posts/*.md");
// For every tag, return a paginate() result.
// Make sure that you pass `{params: {tag}}` to `paginate()`
// so that Astro knows which tag grouping the result is for.
return allTags.flatMap((tag) => {
const filteredPosts = allPosts.filter(
(post) => post.frontmatter.tag === tag,
);
return paginate(filteredPosts, {
params: { tag },
pageSize: 10,
});
});
}
const { page } = Astro.props;
const params = Astro.params;
---
Excluding pages
You can exclude pages or directories from being built by prefixing their names with an underscore (_). Files with the _ prefix won’t be recognized by the router and won’t be placed into the dist/ directory.
You can use this to temporarily disable pages, and also to put tests, utilities, and components in the same folder as their related pages.
In this example, only src/pages/index.astro and src/pages/posts/post1.md will be built as page routes and HTML files.
