Skip to content

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.

776 B
Raw Blame
import { expect, getActorTestDir, initialize, runActor, validateDataset } from '../tools.mjs';

const testActorDirname = getActorTestDir(import.meta.url);
await initialize(testActorDirname);

const { datasetItems } = await runActor(testActorDirname, 16384);

await expect(datasetItems.length > 15, 'Number of dataset items');
await expect(validateDataset(datasetItems, ['url', 'heading', 'requestHandlerMode']), 'Dataset items validation');

await expect(
    datasetItems.filter((it) => it.requestHandlerMode === 'browser').length >= 1,
    'The crawler should handle at least one request in the browser',
);

await expect(
    datasetItems.filter((it) => it.requestHandlerMode === 'httpOnly').length >= 5,
    'The crawler should handle some requests in http-only mode',
);