Awesome
Cacho
A careless caching client optimized for scraping.
Description
Cacho is an HTTP client for scraping. It will do most of the things you want when scraping:
- Follow redirects.
- Set the
User-Agent
to a browser-like string. - Accept and process gzip encoding.
- Detect when it's been rate limited and wait.
- Retry on silly network errors.
- Use persistent HTTP connections.
Most importantly, Cacho will store responses on disk so that multiple runs of your script will not hit the endpoints you are scraping. This prevents being rate limited and also makes your script faster every time.
Usage
require "cacho"
client = Cacho.new
res = client.request(:get, "https://news.ycombinator.com")
Installation
$ gem install cacho
License
See the UNLICENSE
.