GitHub/kill-the-news
1
0
Fork 0
mirror of https://github.com/juherr/kill-the-news.git synced 2026-06-20 22:03:48 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix(feed): correct feed link, canonical id, and strip html wrapper from content

Browse Source 
- link: computed as /admin/feeds/:id/emails instead of stale site_url from KV
- id: computed dynamically from baseUrl instead of stale feed_url from KV
- item description/content: strip <html><head><body> wrapper via extractBodyContent()
  so feed readers receive a body fragment, not a full HTML document

Fixes RSS validator warnings: SelfDoesntMatchLocation (stale KV domain) and
InvalidHTML (full HTML document inside <description>/<content:encoded>).
Adds 8 tests covering extractBodyContent and the new feed/atom link assertions.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Julien Herr
2026-05-22 18:28:13 +02:00
parent 780bf6c190
commit bcc9640591
2 changed files with 94 additions and 7 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/utils/feed-generator.ts
+17 -6
View File
@@ -13,6 +13,13 @@ function parseFromAddress(from: string): { name: string; email?: string } {
return { name: from.trim() };
}
// Email content is stored as a full HTML document. Feed readers expect only
// the body fragment in <description>/<content:encoded>, not a full document.
export function extractBodyContent(html: string): string {
const match = html.match(/<body[^>]*>([\s\S]*?)<\/body>/i);
return match ? match[1] : html;
}
function buildFeed(
feedConfig: FeedConfig,
emails: EmailData[],
@@ -22,11 +29,14 @@ function buildFeed(
const feed = new Feed({
title: feedConfig.title,
description: feedConfig.description || "",
id: feedConfig.feed_url,
link: feedConfig.site_url,
// Computed dynamically so the id is always canonical regardless of what
// was stored in KV at feed-creation time (which may have used a stale domain).
id: `${baseUrl}/rss/${feedId}`,
// Link points to the admin emails page — the "website" this feed represents.
link: `${baseUrl}/admin/feeds/${feedId}/emails`,
language: feedConfig.language,
updated: new Date(),
generator: "Email-to-RSS",
generator: "kill-the-news",
copyright: `Copyright © ${new Date().getFullYear()} ${feedConfig.title}`,
feedLinks: {
rss: `${baseUrl}/rss/${feedId}`,
@@ -35,7 +45,7 @@ function buildFeed(
author: feedConfig.author
? {
name: feedConfig.author,
email: `noreply@${new URL(feedConfig.site_url).hostname}`,
email: `noreply@${new URL(baseUrl).hostname}`,
}
: undefined,
});
@@ -43,12 +53,13 @@ function buildFeed(
for (const email of emails) {
const uniqueId = `${email.receivedAt}-${Buffer.from(email.subject).toString("base64").substring(0, 10)}`;
const firstAttachment = email.attachments?.[0];
const bodyContent = extractBodyContent(email.content);
feed.addItem({
title: email.subject,
id: uniqueId,
link: `${baseUrl}/entries/${feedId}/${email.receivedAt}`,
description: email.content,
content: email.content,
description: bodyContent,
content: bodyContent,
author: [parseFromAddress(email.from)],
date: new Date(email.receivedAt),
enclosure: firstAttachment