Chromium / Puppeteer site crawler

This crawler does a BFS starting from a given site entry point. It will not leave the entry point domain and it will not crawl a page more than once. Given a shared redis host/cluster this crawler can be distributed across multiple machines or processes. Discovered pages will be stored in mongo collection, each with a url, outbound urls, and a radius from the origin.

Installation

yarn

Usage

Basic

./crawl -u https://www.dadoune.com

Distributed

# Terminal 1
./crawl -u https://www.dadoune.com

# Terminal 2
./crawl -r

Debug

DEBUG=crawler:* ./crawl -u https://www.dadoune.com

Options

• --maxRadius or -m the maximum link depth the crawler will explore from the entry url.
• --resume or -r to resume crawling after prematurely exiting a process or to add additional crawlers to an existing crawl.
• --url or -u the entry point URL to kick the crawler off.