RoboBrowser: Your friendly neighborhood web scraper

https://badge.fury.io/py/robobrowser.png https://travis-ci.org/jmcarp/robobrowser.png?branch=master https://coveralls.io/repos/jmcarp/robobrowser/badge.png?branch=master

Homepage: http://robobrowser.readthedocs.org/

RoboBrowser is a simple, Pythonic library for browsing the web without a standalone web browser. RoboBrowser can fetch a page, click on links and buttons, and fill out and submit forms. If you need to interact with web services that don’t have APIs, RoboBrowser can help.

import re
from robobrowser import RoboBrowser

# Browse to Rap Genius
browser = RoboBrowser(history=True)
browser.open('http://rapgenius.com/')

# Search for Queen
form = browser.get_form(action='/search')
form                # <RoboForm q=>
form['q'].value = 'queen'
browser.submit_form(form)

# Look up the first song
songs = browser.select('.song_name')
browser.follow_link(songs[0])
lyrics = browser.select('.lyrics')
lyrics[0].text      # \n[Intro]\nIs this the real life...

# Back to results page
browser.back()

# Look up my favorite song
browser.follow_link('death on two legs')

# Can also search HTML using regex patterns
lyrics = browser.find(class_=re.compile(r'\blyrics\b'))
lyrics.text         # \n[Verse 1]\nYou suck my blood like a leech...

RoboBrowser combines the best of two excellent Python libraries: Requests and BeautifulSoup. RoboBrowser represents browser sessions using Requests and HTML responses using BeautifulSoup, transparently exposing methods of both libraries:

import re
from robobrowser import RoboBrowser

browser = RoboBrowser(user_agent='a python robot')
browser.open('https://github.com/')

# Inspect the browser session
browser.session.cookies['_gh_sess']         # BAh7Bzo...
browser.session.headers['User-Agent']       # a python robot

# Search the parsed HTML
browser.select('div.teaser-icon')       # [<div class="teaser-icon">
                                        # <span class="mega-octicon octicon-checklist"></span>
                                        # </div>,
                                        # ...
browser.find(class_=re.compile(r'column', re.I))    # <div class="one-third column">
                                                    # <div class="teaser-icon">
                                                    # <span class="mega-octicon octicon-checklist"></span>
                                                    # ...

RoboBrowser also includes tools for working with forms, inspired by WebTest and Mechanize.

from robobrowser import RoboBrowser

browser = RoboBrowser()
browser.open('http://twitter.com')

# Get the signup form
signup_form = browser.get_form(class_='signup')
signup_form         # <RoboForm user[name]=, user[email]=, ...

# Inspect its values
signup_form['authenticity_token'].value     # 6d03597 ...

# Fill it out
signup_form['user[name]'].value = 'python-robot'
signup_form['user[user_password]'].value = 'secret'

# Serialize it to JSON
signup_form.serialize()         # {'data': {'authenticity_token': '6d03597...',
                                #  'context': '',
                                #  'user[email]': '',
                                #  'user[name]': 'python-robot',
                                #  'user[user_password]': ''}}

# And submit
browser.submit_form(signup_form)

Requirements

  • Python >= 2.6 or >= 3.3

License

MIT licensed. See the bundled LICENSE file for more details.

Table Of Contents

Related Topics

This Page

Fork me on GitHub