Print the tweets in a Twitter archive that are younger than a year, useful for heroku_ebooks

twitter2ebooks.py

#!/usr/bin/env python

from cStringIO import StringIO
from csv import DictReader
from sys import argv
from zipfile import ZipFile

from arrow import utcnow
import dateutil.parser


def twitter2ebooks(path):
    archive = ZipFile(path)

    csv = archive.read('tweets.csv')
    csv = StringIO(csv)

    rows = DictReader(csv)

    now = utcnow()

    for row in rows:
        created_at = row['timestamp']
        created_at = dateutil.parser.parse(created_at)

        if (created_at - now).days < 365:
            print repr(row['text'])


if __name__ == '__main__':
    twitter2ebooks(argv[1])