Crawlee—A web scraping and browser automation library for Node.js to build reliable crawlers. In JavaScript and TypeScript. Extract data for AI, LLMs, RAG, or GPTs. Download HTML, PDF, JPG, PNG, and other files from websites. Works with Puppeteer, Playwright, Cheerio, JSDOM, and raw HTTP. Both headful and headless mode. With proxy rotation.
TypeScript apifyautomationcrawlercrawlingheadlessheadless-chromejavascriptnodejsnpmplaywrightpuppeteerscraperscrapingtypescriptweb-crawlerweb-crawlingweb-scraping
import { expect, getActorTestDir, initialize, runActor, validateDataset } from '../tools.mjs';
const testActorDirname = getActorTestDir(import.meta.url);
await initialize(testActorDirname);
const { stats, datasetItems } = await runActor(testActorDirname, 16384);
await expect(stats.requestsSucceeded === 3, 'All 3 requests finished');
await expect(stats.requestsFailed === 0, 'No requests failed');
await expect(datasetItems.length === 3, 'Three dataset items');
await expect(validateDataset(datasetItems, ['url', 'title', 'browserId']), 'Dataset items validation');
// Verify we got unique browser IDs (multiple browsers were used)
const browserIds = new Set(datasetItems.map((item) => item.browserId));
console.log(`[Test] Unique browser IDs: ${browserIds.size}`);
await expect(browserIds.size >= 2, 'Multiple browser instances used');
// Verify all URLs were processed
const urls = datasetItems.map((item) => item.url);
await expect(
urls.some((u) => u.includes('crawlee.dev')),
'crawlee.dev processed',
);
await expect(
urls.some((u) => u.includes('apify.com')),
'apify.com processed',
);
await expect(
urls.some((u) => u.includes('blog.apify.com')),
'blog.apify.com processed',
);
// Verify titles were extracted
for (const item of datasetItems) {
await expect(item.title && item.title.length > 0, `Title extracted for ${item.url}`);
}