Files
kochwas/tests/integration/searxng.test.ts
hsiegeln 6a784488f5
All checks were successful
Build & Publish Docker Image / build-and-push (push) Successful in 55s
fix(search): enrich missing SearXNG thumbnails with og:image
SearXNG liefert je nach Seite mal ein thumbnail/img_src mit, mal nicht —
bei Chefkoch-Treffern hatten deshalb zufällig die Hälfte der Kacheln
einen Platzhalter, obwohl die Vorschau dann sehr wohl ein Bild fand.

searchWeb() holt jetzt für jeden Treffer ohne Thumbnail parallel
(max. 6 gleichzeitig, 4 s Timeout pro Request) die Seite und extrahiert
das og:image- oder twitter:image-Meta-Tag. Ergebnis wird 30 min
in-memory gecacht, damit wiederholte Suchen nicht wieder die gleichen
Seiten laden.

Tests:
- Neuer Test: Treffer ohne Thumbnail wird via og:image angereichert.
- Neuer Test: Treffer mit Thumbnail bleibt unverändert (keine Fetch).
- Bestehende Tests deaktivieren Enrichment via enrichThumbnails:false,
  damit sie keine echten Chefkoch-URLs aufrufen.
2026-04-17 17:55:53 +02:00

128 lines
4.7 KiB
TypeScript

import { describe, it, expect, beforeEach, afterEach } from 'vitest';
import { createServer, type Server } from 'node:http';
import type { AddressInfo } from 'node:net';
import { openInMemoryForTest } from '../../src/lib/server/db';
import { addDomain } from '../../src/lib/server/domains/repository';
import { searchWeb } from '../../src/lib/server/search/searxng';
let server: Server;
let baseUrl: string;
function respondWith(results: Record<string, unknown>[]) {
server.on('request', (_req, res) => {
res.writeHead(200, { 'content-type': 'application/json' });
res.end(JSON.stringify({ results }));
});
}
beforeEach(async () => {
server = createServer();
await new Promise<void>((r) => server.listen(0, '127.0.0.1', r));
const addr = server.address() as AddressInfo;
baseUrl = `http://127.0.0.1:${addr.port}`;
});
afterEach(async () => {
await new Promise<void>((r) => server.close(() => r()));
});
describe('searchWeb', () => {
it('filters results by whitelist', async () => {
const db = openInMemoryForTest();
addDomain(db, 'chefkoch.de');
respondWith([
{
url: 'https://www.chefkoch.de/rezepte/123/a.html',
title: 'Carbonara',
content: 'Pasta'
},
{
url: 'https://fake.de/x',
title: 'Not allowed',
content: 'blocked'
}
]);
const hits = await searchWeb(db, 'carbonara', { searxngUrl: baseUrl, enrichThumbnails: false });
expect(hits.length).toBe(1);
expect(hits[0].domain).toBe('chefkoch.de');
expect(hits[0].title).toBe('Carbonara');
});
it('dedupes identical URLs', async () => {
const db = openInMemoryForTest();
addDomain(db, 'chefkoch.de');
respondWith([
{ url: 'https://www.chefkoch.de/a', title: 'A', content: '' },
{ url: 'https://www.chefkoch.de/a', title: 'A dup', content: '' }
]);
const hits = await searchWeb(db, 'a', { searxngUrl: baseUrl, enrichThumbnails: false });
expect(hits.length).toBe(1);
});
it('returns empty list when no domains configured', async () => {
const db = openInMemoryForTest();
const hits = await searchWeb(db, 'x', { searxngUrl: baseUrl, enrichThumbnails: false });
expect(hits).toEqual([]);
});
it('returns empty for empty query', async () => {
const db = openInMemoryForTest();
addDomain(db, 'chefkoch.de');
const hits = await searchWeb(db, ' ', { searxngUrl: baseUrl, enrichThumbnails: false });
expect(hits).toEqual([]);
});
it('enriches missing thumbnails from og:image', async () => {
const pageServer = createServer((_req, res) => {
res.writeHead(200, { 'content-type': 'text/html; charset=utf-8' });
res.end(
'<html><head><meta property="og:image" content="https://cdn.example/foo.jpg" /></head><body></body></html>'
);
});
await new Promise<void>((r) => pageServer.listen(0, '127.0.0.1', r));
const addr = pageServer.address() as AddressInfo;
const pageUrl = `http://127.0.0.1:${addr.port}/rezept`;
try {
const db = openInMemoryForTest();
addDomain(db, '127.0.0.1');
respondWith([{ url: pageUrl, title: 'Kuchen', content: '' }]);
const hits = await searchWeb(db, 'kuchen', { searxngUrl: baseUrl });
expect(hits.length).toBe(1);
expect(hits[0].thumbnail).toBe('https://cdn.example/foo.jpg');
} finally {
await new Promise<void>((r) => pageServer.close(() => r()));
}
});
it('leaves existing thumbnails untouched (no enrichment fetch)', async () => {
const db = openInMemoryForTest();
addDomain(db, 'chefkoch.de');
respondWith([
{
url: 'https://www.chefkoch.de/rezepte/1/x.html',
title: 'X',
thumbnail: 'https://cdn.chefkoch/x.jpg'
}
]);
// enrichment enabled, but thumbnail is set → no fetch expected, no hang
const hits = await searchWeb(db, 'x', { searxngUrl: baseUrl });
expect(hits[0].thumbnail).toBe('https://cdn.chefkoch/x.jpg');
});
it('filters out forum/magazine/listing URLs', async () => {
const db = openInMemoryForTest();
addDomain(db, 'chefkoch.de');
respondWith([
{ url: 'https://www.chefkoch.de/rezepte/123/Ravioli.html', title: 'Ravioli' },
{ url: 'https://www.chefkoch.de/forum/2,17,89865/ravioli.html', title: 'Forum Ravioli' },
{ url: 'https://www.chefkoch.de/magazin/artikel/x.html', title: 'Magazin' },
{ url: 'https://www.chefkoch.de/suche/ravioli', title: 'Suche' },
{ url: 'https://www.chefkoch.de/themen/ravioli/', title: 'Themen' },
{ url: 'https://www.chefkoch.de/rezepte/', title: 'Rezepte Übersicht' }
]);
const hits = await searchWeb(db, 'ravioli', { searxngUrl: baseUrl, enrichThumbnails: false });
expect(hits.length).toBe(1);
expect(hits[0].title).toBe('Ravioli');
});
});