refactor: update RSS feed generation

astro.config.ts

@@ -36,16 +36,12 @@ const { locale }: { locale: ThemeConfig['global']['locale'] } = themeConfig.glob
 export default defineConfig({
   site: url,
   base: '/',
-  trailingSlash: 'always',
   i18n: {
     locales: Object.entries(langMap).map(([path, codes]) => ({
       path,
       codes,
     })),
     defaultLocale: locale,
-    routing: {
-      prefixDefaultLocale: false,
-    },
   },
   integrations: [
     partytown({

src/content/posts/rashomon.md

@@ -4,8 +4,6 @@ published: 1984-09-17
 tags: ["黑泽明"]
 ---
 
-这里是文章内容...
-
 一天，天色已晚。 仆人纪宁正在罗生门下等待雨停。
 
 除了他，宽大的门下空无一人。 只有一只蟋蟀栖息在一根大柱子上，柱子上有的地方已经脱漆。 罗生门位于铃作大寺街上，除了这个人之外，应该还有两三个人，包括避雨的一笠和搓着乌鸦帽的真江吉。 但是，除了这个人之外，一个人也没有。

src/pages/[lang]/rss.xml.ts

@@ -8,23 +8,25 @@ import sanitizeHtml from 'sanitize-html'
 
 const parser = new MarkdownIt()
 const { title, description, url } = themeConfig.site
-const { locale: defaultLocale, moreLocale } = themeConfig.global
+const { moreLocale } = themeConfig.global
 const followConfig = themeConfig.seo?.follow
 
-// Extract first 100 chars from content as description
+// Returns first 200 chars with proper truncation
 function getExcerpt(content: string): string {
-  const plainText = sanitizeHtml(
-    parser.render(content),
-    {
-      allowedTags: [],
-      allowedAttributes: {},
-    },
-  )
+  if (!content)
+    return ''
 
-  return `${plainText.slice(0, 100).trim()}...`
+  // Convert markdown to plain text by removing all HTML tags
+  const plainText = sanitizeHtml(parser.render(content), {
+    allowedTags: [],
+    allowedAttributes: {},
+  })
+
+  const excerpt = plainText.slice(0, 200).trim()
+  return excerpt.length === 200 ? `${excerpt}...` : excerpt
 }
 
-// Return 404 response for invalid language routes
+// Return 404 for invalid language paths
 function return404() {
   return new Response(null, {
     status: 404,
@@ -32,21 +34,22 @@ function return404() {
   })
 }
 
-// Add getStaticPaths for dynamic routes
+// Type for supported non-default languages
+type SupportedLanguage = typeof moreLocale[number]
+
+// Generate static paths for all supported languages
 export function getStaticPaths() {
-  return moreLocale.map(lang => ({ params: { lang } }))
+  return moreLocale.map((lang: SupportedLanguage) => ({ params: { lang } }))
 }
 
-// Generate RSS feed for non-default languages
 export async function GET({ params }: APIContext) {
-  const lang = params.lang as string
+  const lang = params.lang as SupportedLanguage
 
-  // Only generate RSS for valid non-default languages
-  if (!moreLocale.includes(lang)) {
+  // Return 404 if language is not supported
+  if (!moreLocale.includes(lang))
     return return404()
-  }
 
-  // Get posts for specific language (include universal posts)
+  // Get posts for specific language (including universal posts)
   const posts = await getCollection(
     'posts',
     ({ data }: { data: CollectionEntry<'posts'>['data'] }) =>
@@ -57,28 +60,27 @@ export async function GET({ params }: APIContext) {
     title: `${title} (${lang})`,
     description,
     site: url,
-    stylesheet: '/rss/styles.xsl',
-    // Map posts to RSS items with language-specific URLs
-    items: posts.map(post => ({
+    items: posts.map((post: CollectionEntry<'posts'>) => ({
       title: post.data.title,
       pubDate: post.data.published,
       description: post.data.description || getExcerpt(post.body),
-      link: `/${lang}/posts/${post.slug}/`,
-      content: sanitizeHtml(
-        parser.render(post.body),
-        {
-          allowedTags: sanitizeHtml.defaults.allowedTags.concat(['img']),
-        },
-      ),
+      // Generate absolute URL with language prefix
+      link: new URL(`${lang}/posts/${post.slug}/`, url).toString(),
+      // Convert markdown content to HTML, allowing img tags
+      content: post.body
+        ? sanitizeHtml(parser.render(post.body), {
+            allowedTags: sanitizeHtml.defaults.allowedTags.concat(['img']),
+          })
+        : '',
     })),
-    // Add language and follow challenge info
+    // Add XML namespaces for language and follow challenge
     customData: `
       <language>${lang}</language>
       ${followConfig?.feedID && followConfig?.userID
           ? `<follow_challenge>
-          <feedId>${followConfig.feedID}</feedId>
-          <userId>${followConfig.userID}</userId>
-        </follow_challenge>`
+            <feedId>${followConfig.feedID}</feedId>
+            <userId>${followConfig.userID}</userId>
+          </follow_challenge>`
           : ''
       }
     `.trim(),

src/pages/rss.xml.ts

@@ -1,4 +1,3 @@
-import type { APIContext } from 'astro'
 import type { CollectionEntry } from 'astro:content'
 import themeConfig from '@/config'
 import rss from '@astrojs/rss'
@@ -11,22 +10,23 @@ const { title, description, url } = themeConfig.site
 const { locale } = themeConfig.global
 const followConfig = themeConfig.seo?.follow
 
-// Extract first 100 chars from content as description
+// Returns first 200 chars with proper truncation
 function getExcerpt(content: string): string {
-  const plainText = sanitizeHtml(
-    parser.render(content),
-    {
-      allowedTags: [],
-      allowedAttributes: {},
-    },
-  )
+  if (!content)
+    return ''
 
-  return `${plainText.slice(0, 100).trim()}...`
+  // Convert markdown to plain text by removing all HTML tags
+  const plainText = sanitizeHtml(parser.render(content), {
+    allowedTags: [],
+    allowedAttributes: {},
+  })
+
+  const excerpt = plainText.slice(0, 200).trim()
+  return excerpt.length === 200 ? `${excerpt}...` : excerpt
 }
 
-// Generate RSS feed for default language
 export async function GET() {
-  // Only handle posts for default language
+  // Get all posts for default language (including universal posts)
   const posts = await getCollection(
     'posts',
     ({ data }: { data: CollectionEntry<'posts'>['data'] }) =>
@@ -37,28 +37,27 @@ export async function GET() {
     title,
     description,
     site: url,
-    stylesheet: '/rss/styles.xsl',
-    // Map posts to RSS items
     items: posts.map((post: CollectionEntry<'posts'>) => ({
       title: post.data.title,
       pubDate: post.data.published,
       description: post.data.description || getExcerpt(post.body),
-      link: `/posts/${post.slug}/`,
-      content: sanitizeHtml(
-        parser.render(post.body),
-        {
-          allowedTags: sanitizeHtml.defaults.allowedTags.concat(['img']),
-        },
-      ),
+      // Generate absolute URL for post
+      link: new URL(`posts/${post.slug}/`, url).toString(),
+      // Convert markdown content to HTML, allowing img tags
+      content: post.body
+        ? sanitizeHtml(parser.render(post.body), {
+            allowedTags: sanitizeHtml.defaults.allowedTags.concat(['img']),
+          })
+        : '',
     })),
-    // Add language and follow challenge info
+    // Add XML namespaces for language and follow challenge
     customData: `
       <language>${locale}</language>
       ${followConfig?.feedID && followConfig?.userID
           ? `<follow_challenge>
-          <feedId>${followConfig.feedID}</feedId>
-          <userId>${followConfig.userID}</userId>
-        </follow_challenge>`
+            <feedId>${followConfig.feedID}</feedId>
+            <userId>${followConfig.userID}</userId>
+          </follow_challenge>`
           : ''
       }
     `.trim(),