consume-rBeer

A spider/scraper for ratebeer.com

Goals:

• spider for all known breweries, beers, styles, and users (public profiles and histories)
• parse each of these four page types and output the parsed data into various formats, probably an sqlite db to start

Simple Use:

To grab all the ratings for a beer (Stone Ruination IPA) and the make a histogram (using matplotlib's pylab):

> import Beer
> import pylab
> ruin = Beer.Beer(14709)
> ruin.scrape_user_rating_list()
> pylab.hist([rating for (name, userid, rating) in ruin.ratings], bins=20)

And you should see a nice 20 bin histogram of these ratings, which as of 22 Jan 2012, peaked at just above 4.0, with a gaussian shape.

You can make a similar histogram but this time for every beer for a brewery (Redemption):

> import Brewery
> import pylab
> redemption = Brewery.Brewery(rb_id=11318)
> redemption.parse()
> for beer in redemption.beers:
>     beer.scrape_user_rating_list()
> hist([rating for (name, uid, rating) in beer.ratings for beer in redemption.beers], bins=20)

This one produces a few clusters of bin magnitudes, with the peak at around 3.5

Neat!

some further sample usage of data grabbed with the scrapper can be seen in notebooks/