Pure-Python robots.txt parser with support for modern conventions


Keywords
robots, txt, parser, rep, hacktoberfest, python, robots-parser, robots-txt
License
BSD-3-Clause
Install
pip install Protego==0.3.1

Documentation

Protego

Supported Python Versions

CI

Protego is a pure-Python robots.txt parser with support for modern conventions.

Install

To install Protego, simply use pip:

pip install protego

Usage

>>> from protego import Protego >>> robotstxt = """ ... User-agent: * ... Disallow: / ... Allow: /about ... Allow: /account ... Disallow: /account/contact$ ... Disallow: /account/*/profile ... Crawl-delay: 4 ... Request-rate: 10/1m # 10 requests every 1 minute ... ... Sitemap: http://example.com/sitemap-index.xml ... Host: http://example.co.in ... """ >>> rp = Protego.parse(robotstxt) >>> rp.can_fetch("http://example.com/profiles", "mybot") False >>> rp.can_fetch("http://example.com/about", "mybot") True >>> rp.can_fetch("http://example.com/account", "mybot") True >>> rp.can_fetch("http://example.com/account/myuser/profile", "mybot") False >>> rp.can_fetch("http://example.com/account/contact", "mybot") False >>> rp.crawl_delay("mybot") 4.0 >>> rp.request_rate("mybot") RequestRate(requests=10, seconds=60, start_time=None, end_time=None) >>> list(rp.sitemaps) ['http://example.com/sitemap-index.xml'] >>> rp.preferred_host 'http://example.co.in'

Using Protego with Requests:

>>> from protego import Protego >>> import requests >>> r = requests.get("https://google.com/robots.txt") >>> rp = Protego.parse(r.text) >>> rp.can_fetch("https://google.com/search", "mybot") False >>> rp.can_fetch("https://google.com/search/about", "mybot") True >>> list(rp.sitemaps) ['https://www.google.com/sitemap.xml']

Comparison

The following table compares Protego to the most popular robots.txt parsers implemented in Python or featuring Python bindings:

Protego RobotFileParser Reppy Robotexclusionrulesparser
Implementation language Python Python C++ Python
Reference specification Google `Martijn Koster’s 1996 dra ft`_
Wildcard support ✓ ✓ ✓
Length-based precedence ✓ ✓
Performance +40% +1300% -25%

API Reference

Class protego.Protego:

Properties

  • sitemaps {list_iterator} A list of sitemaps specified in robots.txt.
  • preferred_host {string} Preferred host specified in robots.txt.

Methods

  • parse(robotstxt_body) Parse robots.txt and return a new instance of protego.Protego.
  • can_fetch(url, user_agent) Return True if the user agent can fetch the URL, otherwise return False.
  • crawl_delay(user_agent) Return the crawl delay specified for the user agent as a float. If nothing is specified, return None.
  • request_rate(user_agent) Return the request rate specified for the user agent as a named tuple RequestRate(requests, seconds, start_time, end_time). If nothing is specified, return None.
  • visit_time(user_agent) Return the visit time specified for the user agent as a named tuple VisitTime(start_time, end_time). If nothing is specified, return None.