python script to archive your tweets

twitter_archiver.py

#!/usr/bin/env python

"""
twitter_archiver.py written by Jehiah Czebotar 2010 <jehiah@gmail.com> http://jehiah.cz/

this uses the great 'python twitter tools' library by Mike Verdone
http://mike.verdone.ca/twitter/

usage:
   $ pip install twitter
   $ python twitter_archiver.py
   $ view tweets.json

"""

import sys
import os.path
import simplejson as json
import time
import logging

logging.basicConfig(stream=sys.stdout, level=logging.DEBUG,
   format='%(asctime)s %(process)d %(filename)s %(lineno)d %(levelname)s #| %(message)s',
   datefmt='%H:%M:%S')

from twitter.api import Twitter, TwitterHTTPError
from twitter.oauth import OAuth, read_token_file
from twitter.oauth_dance import oauth_dance
from twitter.cmdline import CONSUMER_KEY, CONSUMER_SECRET

OPTIONS = {
    'oauth_filename': os.environ.get('HOME', '') + os.sep + '.twitter_oauth',
    'secure': True,
}


def save_user_timeline(twitter, outputfile):
    # docs are at http://dev.twitter.com/doc/get/statuses/user_timeline
    max_id = None
    data = []
    while True:
        try:
            kwargs = {}
            if max_id:
                kwargs['max_id'] = max_id
            temp_data = twitter.statuses.user_timeline(count=200, trim_user='t', include_rts='t', include_entities='t', **kwargs)
            if not temp_data:
                # yay we reached the end
                break
            max_id = min([x.get('id') for x in temp_data]) # bound to the min of this range
            print temp_data
            data += temp_data
        except TwitterHTTPError, e:
            logging.exception('twitter error')
            d = (e.e.code, e.uri, e.format, e.uriparts, e.e.fp.read())
            print d
            f = open('error.log', 'w')
            f.write(json.dumps(d))
            f.close()
            break
        except:
            logging.exception('failed')
            break
        time.sleep(15)
    
    f = open(outputfile, 'w')
    f.write(json.dumps(data))
    f.close()
    print 'saved %d records' % len(data)

def main(args=sys.argv[1:]):
    oauth_filename = os.path.expanduser(OPTIONS['oauth_filename'])

    if ('authorize' in args
        or not os.path.exists(oauth_filename)):
        oauth_dance(
            "the Command-Line Tool", CONSUMER_KEY, CONSUMER_SECRET,
            OPTIONS['oauth_filename'])

    oauth_token, oauth_token_secret = read_token_file(oauth_filename)

    twitter = Twitter(
        auth=OAuth(
            oauth_token, oauth_token_secret, CONSUMER_KEY, CONSUMER_SECRET),
        secure=OPTIONS['secure'],
        api_version='1',
        domain='api.twitter.com')

    save_user_timeline(twitter, outputfile='tweets.json')

if __name__ == "__main__":
    main()

Thanks for this, it still works with just a minor change to api_version='1.1':

https://gist.github.com/hugovk/d5d57086ed8685345f54a9cc5fcf95b9/revisions