Handle malformed article URLs

components/ReaderView.tsx

@@ -2,6 +2,7 @@
 import React, { useEffect, useRef } from 'react';
 import { Article, ReaderSettings } from '../types';
 import { FileText, MousePointerClick } from 'lucide-react';
+import { getDisplayUrl } from '../utils/url';
 
 interface ReaderViewProps {
   article?: Article | null;
@@ -68,6 +69,8 @@ export const ReaderView: React.FC<ReaderViewProps> = ({ article, settings, onTog
     );
   }
 
+  const displayUrl = getDisplayUrl(article.url);
+
   return (
     <div className="bg-white dark:bg-slate-900 rounded-2xl border border-slate-200 dark:border-slate-800 shadow-sm overflow-hidden h-[calc(100vh-12rem)] flex flex-col transition-colors duration-300">
       <div className="p-6 border-b border-slate-100 dark:border-slate-800 bg-white dark:bg-slate-900 sticky top-0 z-10 flex justify-between items-start">
@@ -76,12 +79,12 @@ export const ReaderView: React.FC<ReaderViewProps> = ({ article, settings, onTog
             {article.title}
             </h2>
             <a
-            href={article.url} 
+            href={displayUrl.href}
             target="_blank"
             rel="noopener noreferrer"
             className="text-sm text-blue-600 dark:text-blue-400 hover:underline mt-2 inline-block"
             >
-            {new URL(article.url).hostname}
+            {displayUrl.hostname}
             </a>
         </div>
         

package.json

@@ -5,7 +5,8 @@
   "scripts": {
     "dev": "vite",
     "build": "vite build",
-    "preview": "vite preview"
+    "preview": "vite preview",
+    "test": "vitest"
   },
   "dependencies": {
     "@google/genai": "^1.30.0",
@@ -20,6 +21,7 @@
     "@types/uuid": "^10.0.0",
     "@vitejs/plugin-react": "^4.3.4",
     "typescript": "^5.7.2",
-    "vite": "^6.0.0"
+    "vite": "^6.0.0",
+    "vitest": "^4.0.14"
   }
 }

services/geminiService.ts

@@ -1,5 +1,6 @@
 import { GoogleGenAI, Modality } from '@google/genai';
 import { VoiceName } from '../types';
+import { normalizeUrl } from '../utils/url';
 
 const getAiClient = () => {
   const apiKey = process.env.API_KEY;
@@ -9,18 +10,6 @@ const getAiClient = () => {
   return new GoogleGenAI({ apiKey });
 };
 
-/**
- * Helper to ensure URL has protocol. 
- * Proxies often fail if 'http/https' is missing.
- */
-const normalizeUrl = (url: string) => {
-  let cleanUrl = url.trim();
-  if (!cleanUrl.startsWith('http://') && !cleanUrl.startsWith('https://')) {
-    return `https://${cleanUrl}`;
-  }
-  return cleanUrl;
-};
-
 /**
  * List of CORS proxies to try in order.
  * This improves reliability if one service is down or blocked.

utils/url.test.ts

@@ -0,0 +1,28 @@
+import { describe, expect, it } from 'vitest';
+import { getDisplayUrl, normalizeUrl } from './url';
+
+describe('normalizeUrl', () => {
+  it('adds https protocol when missing', () => {
+    expect(normalizeUrl('example.com/page')).toBe('https://example.com/page');
+  });
+});
+
+describe('getDisplayUrl', () => {
+  it('returns hostname and normalized href for valid URLs', () => {
+    const result = getDisplayUrl('https://example.com/path');
+    expect(result.hostname).toBe('example.com');
+    expect(result.href).toBe('https://example.com/path');
+  });
+
+  it('normalizes URLs without protocol for display', () => {
+    const result = getDisplayUrl('example.com/path');
+    expect(result.hostname).toBe('example.com');
+    expect(result.href).toBe('https://example.com/path');
+  });
+
+  it('falls back to raw URL when parsing fails', () => {
+    const result = getDisplayUrl('not a url');
+    expect(result.hostname).toBe('not a url');
+    expect(result.href).toBe('not a url');
+  });
+});

utils/url.ts

@@ -0,0 +1,23 @@
+export const normalizeUrl = (url: string) => {
+  let cleanUrl = url.trim();
+  if (!cleanUrl.startsWith('http://') && !cleanUrl.startsWith('https://')) {
+    return `https://${cleanUrl}`;
+  }
+  return cleanUrl;
+};
+
+export const getDisplayUrl = (url: string): { href: string; hostname: string } => {
+  const normalized = normalizeUrl(url);
+
+  try {
+    const parsed = new URL(normalized);
+    return { href: normalized, hostname: parsed.hostname };
+  } catch {
+    try {
+      const fallback = new URL(url);
+      return { href: url, hostname: fallback.hostname };
+    } catch {
+      return { href: url, hostname: url };
+    }
+  }
+};