From 6e8b4f30c14c7210ba18d1e5d603c051ed448a59 Mon Sep 17 00:00:00 2001
From: jj <log@riseup.net>
Date: Sat, 8 Feb 2025 13:53:29 +0000
Subject: [PATCH] api/url: add function for resolving shortlinks

motivation: we frequently need to resolve shortlinks to full URLs
let's have a common standard function for doing this safely
instead of reinventing the wheel in every single service module
---
 api/src/misc/utils.js     | 15 ++++++++-------
 api/src/processing/url.js | 15 +++++++++++++++
 2 files changed, 23 insertions(+), 7 deletions(-)

diff --git a/api/src/misc/utils.js b/api/src/misc/utils.js
index 331528d4..ffa75433 100644
--- a/api/src/misc/utils.js
+++ b/api/src/misc/utils.js
@@ -1,12 +1,13 @@
+import { request } from 'undici';
 const redirectStatuses = new Set([301, 302, 303, 307, 308]);
 
-export async function getRedirectingURL(url, dispatcher) {
-    const location = await fetch(url, {
-        redirect: 'manual',
-        dispatcher,
-    }).then((r) => {
-        if (redirectStatuses.has(r.status) && r.headers.has('location')) {
-            return r.headers.get('location');
+export async function getRedirectingURL(url, dispatcher, userAgent) {
+    const location = await request(url, {
+        dispatcher, method: 'HEAD',
+        headers: { 'user-agent': userAgent }
+    }).then(r => {
+        if (redirectStatuses.has(r.statusCode) && r.headers['location']) {
+            return r.headers['location'];
         }
     }).catch(() => null);
 
diff --git a/api/src/processing/url.js b/api/src/processing/url.js
index cfbbecc0..5c4035eb 100644
--- a/api/src/processing/url.js
+++ b/api/src/processing/url.js
@@ -4,6 +4,7 @@ import { strict as assert } from "node:assert";
 import { env } from "../config.js";
 import { services } from "./service-config.js";
 import { friendlyServiceName } from "./service-alias.js";
+import { getRedirectingURL } from "../misc/utils.js";
 
 function aliasURL(url) {
     assert(url instanceof URL);
@@ -221,3 +222,17 @@ export function extract(url) {
 
     return { host, patternMatch };
 }
+
+export async function resolveRedirectingURL(url, dispatcher, userAgent) {
+    const originalService = getHostIfValid(normalizeURL(url));
+    if (!originalService) return;
+
+    const canonicalURL = await getRedirectingURL(url, dispatcher, userAgent);
+    if (!canonicalURL) return;
+
+    const { host, patternMatch } = extract(normalizeURL(canonicalURL));
+
+    if (host === originalService) {
+        return patternMatch;
+    }
+}