ragstoriches 0.2

Develop highly-concurrent web scrapers, easily.

ragstoriches is a combined library/framework to ease writing web scrapers using gevent and requests.

A simple example to tell the story:

#!/usr/bin/env python
# -*- coding: utf-8 -*-

from urlparse import urljoin
import re

from bs4 import BeautifulSoup
from ragstoriches.scraper import Scraper

rr = Scraper(__name__)

@rr.scraper
def index(requests, context,
          url='http://eastidaho.craigslist.org/search/act?query=+'):
    soup = BeautifulSoup(requests.get(url).text)

    for row in soup.find_all(class_='row'):
        yield 'posting', context, urljoin(url, row.find('a').attrs['href'])

    nextpage = soup.find(class_='nextpage')
    if nextpage:
        yield 'index', context, urljoin(url, nextpage.find('a').attrs['href'])


@rr.scraper
def posting(requests, context, url):
    soup = BeautifulSoup(requests.get(url).text)
    infos = soup.find(class_='postinginfos').find_all(class_='postinginfo')

    title = soup.find(class_='postingtitle').text.strip()
    id = re.findall('\d+', infos[0].text)[0]
    date = infos[1].find('date').text.strip()
    body = soup.find(id='postingbody').text.strip()

    print title
    print '=' * len(title)
    print 'post *%s*, posted on %s' % (id, date)
    print body
    print

Install the library and BeautifulSoup 4 using pip install ragstoriches beautifulsoup4, then save the above as craigs.py, finally run with ragstoriches craigs.py.

You will get a bunch of jumbled input, so next step is redirecting stdout to a file:

ragstoriches craigs.py > output.md

Try giving different urls for this scraper on the command-line:

ragstoriches craigs.py http://newyork.craigslist.org/mnh/acc/  > output.md  # hustle
ragstoriches craigs.py http://orangecounty.craigslist.org/wet/ > output.md  # writing OC
ragstoriches craigs.py http://seattle.craigslist.org/w4m/      > output.md  # sleepless in seattle

There are a lot of commandline-options available, see ragstoriches --help for a list.

Writing scrapers

A scraper module consists of some initialization code and a number of subscrapers. Scraping starts by calling the a scraper named index on the scraper rr in the moduel (see the example above).

The requests argument should be treated like the requests module (it actually is an instance of requests Pool). As long as you use it for fetching webpages, you never have to worry about blocking or exceeding concurrency limits.

The context variable is arbitrary, but by convention a dictionary. It’s a way of passing state from one scraper to another or sharing it. It is only passed on by ragstoriches and never touched otherwise.

The url is the url to scrape and parse.

Return values of scrapers are ignored. However, if a scraper is a generater (i.e. contains a yield statement), any value it yields must be a 3-tuple consisting of the name of a scraper, a context object and another url. These are added to the queue of jobs to scrape.

Good friends of ragstoriches are the urlparse.urljoin function and BeautifulSoup4.

Usage as a library

You can use ragstoriches as a library as well by not using the commandline tools but simply importing a scraper and running it with the scrape() method. Remember to monkey-patch using gevent first.

See the source files for details, as there is not that much documentation available at this point.