feat: improved sitemap

.changeset/popular-cherries-float.md

@@ -0,0 +1,18 @@
+---
+'@astrojs/sitemap': minor
+---
+
+# Key features
+
+- Split up your large sitemap into multiple sitemaps by custom limit.
+- Ability to add sitemap specific attributes such as `lastmod` etc.
+- Final output customization via JS function.
+- Localization support.
+- Reliability: all config options are validated.
+
+## Important changes
+
+The integration always generates at least two files instead of one:
+
+- `sitemap-index.xml` - index file;
+- `sitemap-{i}.xml` - actual sitemap.

examples/integrations-playground/astro.config.mjs

@@ -9,5 +9,6 @@ import solid from '@astrojs/solid-js';
 
 // https://astro.build/config
 export default defineConfig({
+	site: 'https://example.com',
 	integrations: [lit(), react(), tailwind(), turbolinks(), partytown(), sitemap(), solid()],
 });

packages/integrations/sitemap/README.md

@@ -64,7 +64,35 @@ export default {
 }
 ```
 
-Now, [build your site for production](https://docs.astro.build/en/reference/cli-reference/#astro-build) via the `astro build` command. You should find your sitemap under `dist/sitemap.xml`!
+Now, [build your site for production](https://docs.astro.build/en/reference/cli-reference/#astro-build) via the `astro build` command. You should find your _sitemap_ under `dist/sitemap-index.xml` and `dist/sitemap-0.xml`!
+
+Generated sitemap content for two pages website:
+
+**sitemap-index.xml**
+
+```xml
+<?xml version="1.0" encoding="UTF-8"?>
+  <sitemapindex xmlns="http://www.sitemaps.org/schemas/sitemap/0.9">
+  <sitemap>
+    <loc>https://stargazers.club/sitemap-0.xml</loc>
+  </sitemap>
+</sitemapindex>
+```
+
+**sitemap-0.xml**
+<?xml version="1.0" encoding="UTF-8"?>
+
+```xml
+<?xml version="1.0" encoding="UTF-8"?>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9" xmlns:news="http://www.google.com/schemas/sitemap-news/0.9" xmlns:xhtml="http://www.w3.org/1999/xhtml" xmlns:image="http://www.google.com/schemas/sitemap-image/1.1" xmlns:video="http://www.google.com/schemas/sitemap-video/1.1">
+  <url>
+    <loc>https://stargazers.club/</loc>
+  </url>
+  <url>
+    <loc>https://stargazers.club/second-page/</loc>
+  </url>
+</urlset>
+```
 
 You can also check our [Astro Integration Documentation][astro-integration] for more on integrations.
 
@@ -111,5 +139,158 @@ export default {
 }
 ```
 
+### entryLimit
+
+Non-negative `Number` of entries per sitemap file. Default value is 45000. A sitemap index and multiple sitemaps are created if you have more entries. See explanation on [Google](https://developers.google.com/search/docs/advanced/sitemaps/large-sitemaps).
+
+__astro.config.mjs__
+
+```js
+import sitemap from '@astrojs/sitemap';
+
+export default {
+  site: 'https://stargazers.club',
+  integrations: [
+    sitemap({
+      entryLimit: 10000,
+    }),
+  ],
+}
+```
+
+### changefreq, lastmod, priority
+
+`changefreq` - How frequently the page is likely to change. Available values: `always` \| `hourly` \| `daily` \| `weekly` \| `monthly` \| `yearly` \| `never`.    
+
+`priority` - The priority of this URL relative to other URLs on your site. Valid values range from 0.0 to 1.0.  
+
+`lastmod` - The date of page last modification.  
+
+`changefreq` and `priority` are ignored by Google.  
+
+See detailed explanation of sitemap specific options on [sitemap.org](https://www.sitemaps.org/protocol.html).  
+
+
+:exclamation: This integration uses 'astro:build:done' hook. The hook exposes generated page paths only. So with present version of Astro the integration has no abilities to analyze a page source, frontmatter etc. The integration can add `changefreq`, `lastmod` and `priority` attributes only in a batch or nothing.
+
+__astro.config.mjs__
+
+```js
+import sitemap from '@astrojs/sitemap';
+
+export default {
+  site: 'https://stargazers.club',
+  integrations: [
+    sitemap({
+      changefreq: 'weekly',
+      priority: 0.7,
+      lastmod: new Date('2022-02-24'),
+    }),
+  ],
+}
+```
+
+### serialize
+
+Async or sync function called for each sitemap entry just before writing to a disk.  
+
+It receives as parameter `SitemapItem` object which consists of `url` (required, absolute page URL) and optional `changefreq`, `lastmod`, `priority` and `links` properties.  
+
+Optional `links` property contains a `LinkItem` list of alternate pages including a parent page.  
+`LinkItem` type has two required fields: `url` (the fully-qualified URL for the version of this page for the specified language) and `hreflang` (a supported language code targeted by this version of the page).
+
+`serialize` function should return `SitemapItem`, touched or not.
+
+The example below shows the ability to add the sitemap specific properties individually.
+
+__astro.config.mjs__
+
+```js
+import sitemap from '@astrojs/sitemap';
+
+export default {
+  site: 'https://stargazers.club',
+  integrations: [
+    sitemap({
+      serialize(item) {
+        if (/your-special-page/.test(item.url)) {
+          item.changefreq = 'daily';
+          item.lastmod = new Date();
+          item.priority = 0.9;
+        }
+        return item;
+      },
+    }),
+  ],
+}
+```
+
+### i18n
+
+To localize a sitemap you should supply the integration config with the `i18n` option. The integration will check generated page paths on presence of locale keys in paths.
+
+`i18n` object has two required properties:
+
+- `defaultLocale`: `String`. Its value must exist as one of `locales` keys.
+- `locales`:  `Record<String, String>`, key/value - pairs. The key is used to look for a locale part in a page path. The value is a language attribute, only English alphabet and hyphen allowed. See more about language attribute on [MDN](https://developer.mozilla.org/en-US/docs/Web/HTML/Global_attributes/lang).
+
+
+Read more about localization on Google in [Advanced SEO](https://developers.google.com/search/docs/advanced/crawling/localized-versions#all-method-guidelines).
+
+__astro.config.mjs__
+
+```js
+import sitemap from '@astrojs/sitemap';
+
+export default {
+  site: 'https://stargazers.club',
+  integrations: [
+    sitemap({  
+      i18n: {
+        defaultLocale: 'en',   // All urls that don't contain `es` or `fr` after `https://stargazers.club/` will be treated as default locale, i.e. `en`
+        locales: {
+          en: 'en-US',         // The `defaultLocale` value must present in `locales` keys
+          es: 'es-ES',
+          fr: 'fr-CA',
+        },
+      },
+    }),
+  ],
+};
+...
+
+```
+
+The sitemap content will be:
+
+```xml
+...
+  <url>
+    <loc>https://stargazers.club/</loc>
+    <xhtml:link rel="alternate" hreflang="en-US" href="https://stargazers.club/"/>
+    <xhtml:link rel="alternate" hreflang="es-ES" href="https://stargazers.club/es/"/>
+    <xhtml:link rel="alternate" hreflang="fr-CA" href="https://stargazers.club/fr/"/>
+  </url>
+  <url>
+    <loc>https://stargazers.club/es/</loc>
+    <xhtml:link rel="alternate" hreflang="en-US" href="https://stargazers.club/"/>
+    <xhtml:link rel="alternate" hreflang="es-ES" href="https://stargazers.club/es/"/>
+    <xhtml:link rel="alternate" hreflang="fr-CA" href="https://stargazers.club/fr/"/>
+  </url>
+  <url>
+    <loc>https://stargazers.club/fr/</loc>
+    <xhtml:link rel="alternate" hreflang="en-US" href="https://stargazers.club/"/>
+    <xhtml:link rel="alternate" hreflang="es-ES" href="https://stargazers.club/es/"/>
+    <xhtml:link rel="alternate" hreflang="fr-CA" href="https://stargazers.club/fr/"/>
+  </url>
+  <url>
+    <loc>https://stargazers.club/es/second-page/</loc>
+    <xhtml:link rel="alternate" hreflang="es-ES" href="https://stargazers.club/es/second-page/"/>
+    <xhtml:link rel="alternate" hreflang="fr-CA" href="https://stargazers.club/fr/second-page/"/>
+    <xhtml:link rel="alternate" hreflang="en-US" href="https://stargazers.club/second-page/"/>
+  </url>
+...
+```
+
 [astro-integration]: https://docs.astro.build/en/guides/integrations-guide/
 [astro-ui-frameworks]: https://docs.astro.build/en/core-concepts/framework-components/#using-framework-components

packages/integrations/sitemap/package.json

@@ -13,20 +13,27 @@
   },
   "keywords": [
     "astro-component",
-    "seo"
+    "seo",
+    "sitemap"
   ],
   "bugs": "https://github.com/withastro/astro/issues",
   "homepage": "https://astro.build",
   "exports": {
     ".": "./dist/index.js",
     "./package.json": "./package.json"
   },
+  "files": [
+    "dist"
+  ],
   "scripts": {
     "build": "astro-scripts build \"src/**/*.ts\" && tsc",
     "build:ci": "astro-scripts build \"src/**/*.ts\"",
     "dev": "astro-scripts dev \"src/**/*.ts\""
   },
-  "dependencies": {},
+  "dependencies": {
+    "sitemap": "^7.1.1",
+    "zod": "^3.17.3"
+  },
   "devDependencies": {
     "astro": "workspace:*",
     "astro-scripts": "workspace:*"

packages/integrations/sitemap/src/config-defaults.ts

@@ -0,0 +1,5 @@
+import type { SitemapOptions } from './index';
+
+export const SITEMAP_CONFIG_DEFAULTS: SitemapOptions & any = {
+	entryLimit: 45000,
+};

packages/integrations/sitemap/src/constants.ts

@@ -0,0 +1,9 @@
+export const changefreqValues = [
+	'always',
+	'hourly',
+	'daily',
+	'weekly',
+	'monthly',
+	'yearly',
+	'never',
+] as const;

packages/integrations/sitemap/src/generate-sitemap.ts

@@ -0,0 +1,55 @@
+import { SitemapItemLoose } from 'sitemap';
+
+import type { SitemapOptions } from './index';
+import { parseUrl } from './utils/parse-url';
+
+const STATUS_CODE_PAGE_REGEXP = /\/[0-9]{3}\/?$/;
+
+/** Construct sitemap.xml given a set of URLs */
+export function generateSitemap(pages: string[], finalSiteUrl: string, opts: SitemapOptions) {
+	const { changefreq, priority: prioritySrc, lastmod: lastmodSrc, i18n } = opts || {};
+	// TODO: find way to respect <link rel="canonical"> URLs here
+	const urls = [...pages].filter((url) => !STATUS_CODE_PAGE_REGEXP.test(url));
+	urls.sort((a, b) => a.localeCompare(b, 'en', { numeric: true })); // sort alphabetically so sitemap is same each time
+
+	const lastmod = lastmodSrc?.toISOString();
+	const priority = typeof prioritySrc === 'number' ? prioritySrc : undefined;
+
+	const { locales, defaultLocale } = i18n || {};
+	const localeCodes = Object.keys(locales || {});
+
+	const getPath = (url: string) => {
+		const result = parseUrl(url, i18n?.defaultLocale || '', localeCodes, finalSiteUrl);
+		return result?.path;
+	};
+	const getLocale = (url: string) => {
+		const result = parseUrl(url, i18n?.defaultLocale || '', localeCodes, finalSiteUrl);
+		return result?.locale;
+	};
+
+	const urlData = urls.map((url) => {
+		let links;
+		if (defaultLocale && locales) {
+			const currentPath = getPath(url);
+			if (currentPath) {
+				const filtered = urls.filter((subUrl) => getPath(subUrl) === currentPath);
+				if (filtered.length > 1) {
+					links = filtered.map((subUrl) => ({
+						url: subUrl,
+						lang: locales[getLocale(subUrl)!],
+					}));
+				}
+			}
+		}
+
+		return {
+			url,
+			links,
+			lastmod,
+			priority,
+			changefreq, // : changefreq as EnumChangefreq,
+		} as SitemapItemLoose;
+	});
+
+	return urlData;
+}