This module provides an API handler and a plugin that enables your Next.js application to have working sitemaps so that search engine crawlers can index all your pages.
npm install @moxy/next-sitemaps1. Add the plugin to next.config.js
const withSitemap = require('@moxy/next-sitemaps/plugin');
module.exports = (phase) => withSitemap(phase, 'https://moxy.studio')({ ...nextConfig });ℹ️ If you have multiple plugins, you may simply compose them or use
next-compose-pluginsfor better readability.
2. Add the API handler on pages/api/sitemap.xml.js
import createSitemapApiHandler from '@moxy/next-sitemaps';
export default createSitemapApiHandler({
mapDynamicRoutes: {
'/[id]': () => ['id1', 'id2'],
},
});3. Add robots.txt file
Sitemap: ${siteUrl}/api/sitemap.xml
User-agent:*
Disallow:
# ...Please note that ${siteUrl} will be replaced automatically by the siteUrl specified when instantiating the plugin. However, the replacement will not happen during development to avoid checking in the changed file into source-control (e.g.: git).
⚠️ During development, new pages will not be present in the XML. Restart the development server for page changes to take effect.
The Next.js plugin that should be installed in next.config.js. It will do two things:
- Iterate over the
/pages(orsrc/pages) folder and map all files into routes. These mappings will be used by the API Handler. - Replace the
${siteUrl}placeholder insiderobots.txtwithsiteUrl.
Type: string
The Next.js current context phase. This plugin will only be active if phase is PHASE_DEVELOPMENT_SERVER or PHASE_PRODUCTION_BUILD.
Type: string
Default: '/'
The website URL.
Type: object
Type: boolean
Default: true if process.env.NODE_ENV is set to production
This option will replace ${siteUrl} inside robots.txt. During development, you want this option disabled to avoid checking in the resulting file into source-control (e.g.: git).
Defines an API handler that will respond with a valid sitemap XML file containing the website pages.
Type: object
Type: object
Default: {}
An object that indicates the possible values for each dynamic route. More info.
Type: string
Default: public, max-age=3600 (max-age is set to 0 if not in production)
A string defining the Cache-Control header for HTTP responses. Since sitemaps can be rather expensive to generate, you may leverage a Reverse Proxy cache or a pull CDN such as CloudFlare (via a page rule), to cache the sitemap response.
While a reverse-proxy or pull CDN is preferable, you may cache the sitemap programmatically like so:
import pMemoize from 'p-memoize';
import createSitemapApiHandler from '@moxy/next-sitemaps';
const apiHandler = createSitemapApiHandler('https://moxy.studio', {
mapDynamicRoutes: {
'/[id]': () => ['id1', 'id2'],
},
});
export default pMemoize(apiHandler, {
cacheKey: () => 'sitemap',
cachePromiseRejection: true,
maxAge: 5 * 60 * 1000, // 5 minutes
});The example above uses p-memoize, which caches the result in-memory for 5 minutes. However, you can use a distributed cache, such as Redis.
Type: function
Default: see logWarningDefault in src/api-handler/index.js
A function to log possible warnings. It has the following signature: (message) => {}.
Type: function
Default: see logErrorDefault in src/api-handler/index.js
A function to log possible errors. It has the following signature: (err) => {}.
createSitemapApiHandler('https://moxy.studio', {
mapDynamicRoutes: {
'/[project]': () => ['some-project', 'another-project'],
'/old/[post]': () => ['old-post', 'ancient-post'],
'/[post]/new': () => ['new-post', 'another-new-post'],
},
});The API route above would respond with the following XML:
<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
<url>
<loc>https://moxy.studio/another-project</loc>
</url>
<url>
<loc>https://moxy.studio/some-project</loc>
</url>
<url>
<loc>https://moxy.studio/old/old-post</loc>
</url>
<url>
<loc>https://moxy.studio/old/ancient-post</loc>
</url>
<url>
<loc>https://moxy.studio/new-post/new</loc>
</url>
<url>
<loc>https://moxy.studio/another-new-post/new</loc>
</url>
</urlset>ℹ️ Catch all routes are also supported.
You might have routes with several dynamic placeholders. In this case, first-level mappers will be called first, second-level mappers will be called secondly, one time for each result of the first mapper, and so on. In short, calls to mappers are permuted and they receive the current iteration values to aid you in data-fetching.
createSitemapApiHandler('https://moxy.studio', {
mapDynamicRoutes: {
'/[company]': async () => {
// Returns ['microsoft', 'apple']
const companies = await fetchCompanies();
return companies;
},
// The function below will be called two times.
// The first will have `company` set microsoft and the second one will have it set to `apple`.
'/[company]/[apps]': ({ company }) => {
// Returns ['vscode', 'vsstudio'] for microsoft and returns ['itunes'] for apple.
const apps = fetchCompanyApps(company);
return apps;
},
},
});Assuming the return values of the mappers above, the generated sitemap would look like:
<?xml version="1.0" encoding="UTF-8"?>
<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
<url>
<loc>https://moxy.studio/apple</loc>
</url>
<url>
<loc>https://moxy.studio/apple/itunes</loc>
</url>
<url>
<loc>https://moxy.studio/microsoft</loc>
</url>
<url>
<loc>https://moxy.studio/microsoft/vscode</loc>
</url>
<url>
<loc>https://moxy.studio/microsoft/vsstudio</loc>
</url>
</urlset>$ npm t
$ npm t -- --watch # To run watch mode