Files
flynn/src/tools/builtin/web-fetch.test.ts
T
William Valentin 6d9e27a591 feat: enhance web-fetch with HTML-to-markdown extraction (Phase 6)
Add turndown + readability for clean content extraction:
- HTML-to-markdown conversion with smart article extraction
- Format parameter (markdown/text/html)
- Response caching for repeated fetches
- 10 tests
2026-02-06 14:24:28 -08:00

223 lines
7.2 KiB
TypeScript

import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
import { webFetchTool, _clearCache, _MAX_CONTENT_LENGTH } from './web-fetch.js';
// ---------------------------------------------------------------------------
// Mock global fetch
// ---------------------------------------------------------------------------
const mockFetch = vi.fn();
vi.stubGlobal('fetch', mockFetch);
// ---------------------------------------------------------------------------
// HTML fixture used by extraction tests
// ---------------------------------------------------------------------------
const SAMPLE_HTML = `<!DOCTYPE html>
<html><head><title>Test Page</title>
<script>console.log('js')</script>
<style>body { color: red }</style>
</head><body>
<nav>Menu items</nav>
<article>
<h1>Article Title</h1>
<p>This is the main content of the article. It contains important information.</p>
<p>Second paragraph with more details about the topic.</p>
</article>
<footer>Copyright 2026</footer>
</body></html>`;
// ---------------------------------------------------------------------------
// Setup / teardown
// ---------------------------------------------------------------------------
beforeEach(() => {
mockFetch.mockReset();
_clearCache();
});
afterEach(() => {
vi.useRealTimers();
});
// ---------------------------------------------------------------------------
// Tests
// ---------------------------------------------------------------------------
describe('web.fetch', () => {
// ---- Metadata ----
it('has correct metadata', () => {
expect(webFetchTool.name).toBe('web.fetch');
expect(webFetchTool.inputSchema.required).toContain('url');
// New format property is advertised in the schema
expect(webFetchTool.inputSchema.properties).toHaveProperty('format');
});
// ---- Default behaviour (markdown extraction) ----
it('fetches a URL and returns markdown by default', async () => {
mockFetch.mockResolvedValue({
ok: true,
status: 200,
text: async () => SAMPLE_HTML,
headers: new Headers({ 'content-type': 'text/html' }),
});
const result = await webFetchTool.execute({ url: 'https://example.com' });
expect(result.success).toBe(true);
expect(result.output).toBeTruthy();
// Should contain markdown-ish content (heading or paragraph text)
expect(result.output).toContain('Article Title');
// Should NOT contain raw HTML tags
expect(result.output).not.toContain('<script>');
expect(result.output).not.toContain('<nav>');
expect(mockFetch).toHaveBeenCalledWith('https://example.com', expect.any(Object));
});
// ---- format=html ----
it('returns raw HTML when format is html', async () => {
mockFetch.mockResolvedValue({
ok: true,
status: 200,
text: async () => SAMPLE_HTML,
headers: new Headers({ 'content-type': 'text/html' }),
});
const result = await webFetchTool.execute({
url: 'https://example.com',
format: 'html',
});
expect(result.success).toBe(true);
// Raw HTML is returned unchanged
expect(result.output).toContain('<article>');
expect(result.output).toContain('<h1>Article Title</h1>');
expect(result.output).toContain('<script>');
});
// ---- format=text ----
it('returns extracted text when format is text', async () => {
mockFetch.mockResolvedValue({
ok: true,
status: 200,
text: async () => SAMPLE_HTML,
headers: new Headers({ 'content-type': 'text/html' }),
});
const result = await webFetchTool.execute({
url: 'https://example.com',
format: 'text',
});
expect(result.success).toBe(true);
expect(result.output).toContain('Article Title');
expect(result.output).toContain('main content');
// No HTML tags
expect(result.output).not.toContain('<');
});
// ---- JSON content-type ----
it('returns prettified JSON for application/json content', async () => {
const jsonPayload = '{"name":"Flynn","version":1}';
mockFetch.mockResolvedValue({
ok: true,
status: 200,
text: async () => jsonPayload,
headers: new Headers({ 'content-type': 'application/json' }),
});
const result = await webFetchTool.execute({ url: 'https://api.example.com/data' });
expect(result.success).toBe(true);
// Should be prettified (indented) JSON
const parsed = JSON.parse(result.output);
expect(parsed).toEqual({ name: 'Flynn', version: 1 });
expect(result.output).toContain('\n'); // multi-line
});
// ---- HTTP error ----
it('returns error on HTTP failure', async () => {
mockFetch.mockResolvedValue({
ok: false,
status: 404,
text: async () => 'Not Found',
headers: new Headers(),
});
const result = await webFetchTool.execute({ url: 'https://example.com/nope' });
expect(result.success).toBe(false);
expect(result.error).toContain('404');
});
// ---- Network error ----
it('returns error on network failure', async () => {
mockFetch.mockRejectedValue(new Error('network error'));
const result = await webFetchTool.execute({ url: 'https://down.example.com' });
expect(result.success).toBe(false);
expect(result.error).toContain('network error');
});
// ---- Caching ----
it('caches responses and reuses them on the second call', async () => {
mockFetch.mockResolvedValue({
ok: true,
status: 200,
text: async () => '<html><body><p>Cached content</p></body></html>',
headers: new Headers({ 'content-type': 'text/html' }),
});
const first = await webFetchTool.execute({ url: 'https://cached.example.com' });
const second = await webFetchTool.execute({ url: 'https://cached.example.com' });
expect(first.success).toBe(true);
expect(second.success).toBe(true);
expect(second.output).toBe(first.output);
// fetch should have been called exactly once — second call served from cache
expect(mockFetch).toHaveBeenCalledTimes(1);
});
it('expires cache entries after TTL', async () => {
vi.useFakeTimers();
mockFetch.mockResolvedValue({
ok: true,
status: 200,
text: async () => '<html><body><p>Stale content</p></body></html>',
headers: new Headers({ 'content-type': 'text/html' }),
});
// First call — populates cache
await webFetchTool.execute({ url: 'https://stale.example.com' });
expect(mockFetch).toHaveBeenCalledTimes(1);
// Advance time past the 5-minute TTL
vi.advanceTimersByTime(5 * 60 * 1_000 + 1);
// Second call — cache should be expired, triggers a new fetch
await webFetchTool.execute({ url: 'https://stale.example.com' });
expect(mockFetch).toHaveBeenCalledTimes(2);
});
// ---- Truncation ----
it('truncates content exceeding max length with a marker', async () => {
const longContent = 'A'.repeat(_MAX_CONTENT_LENGTH + 1_000);
mockFetch.mockResolvedValue({
ok: true,
status: 200,
text: async () => longContent,
headers: new Headers({ 'content-type': 'text/plain' }),
});
const result = await webFetchTool.execute({ url: 'https://big.example.com' });
expect(result.success).toBe(true);
expect(result.output).toContain('[content truncated]');
// Total length should be MAX + marker, not the original
expect(result.output.length).toBeLessThan(longContent.length);
});
});