IMDB Ratings DB Munging

munge.py

# First, get a clean version of just the ratings data

ratings = open('ratings.list').read()
_, ratings = ratings.split('MOVIE RATINGS REPORT\n\n')
ratings, _ = ratings.split('\n\n------------------------------------------------------------------------------')
open('ratings.clean.list', 'w').write(ratings)

# Now play
import pandas as pd
titles, rating_data = ratings.split('\n', 1)
titles = titles.split()
rating_data_lines = rating_data.splitlines()
rating_data_split = [re.split(r"\s+", l, maxsplit=len(titles)-1) for l in rating_data_lines]

ratings = pd.DataFrame(rating_data_split, columns=titles).convert_objects(convert_numeric=True)

ratings = pd.read_csv('ratings.clean.list', delimiter=r"\s\s+")