A script that helps to import Twitter's official archive.

import.py

#!/usr/bin/env python

USAGE = """
This script helps to import Twitter's official archive.

Usage: ./import.py tweets.csv

"""
"""
Original data structure:
['tweet_id',
 'in_reply_to_status_id',
 'in_reply_to_user_id',
 'timestamp',
 'source',
 'text',
 'retweeted_status_id',
 'retweeted_status_user_id',
 'retweeted_status_timestamp',
 'expanded_urls']

Processed data structure:
['tweet_id',
 'timestamp',
 'text',
 'expanded_urls']

"""

import sys
import csv
import time
from datetime import datetime


def main(inputfile):
    # Get rows from input file
    reader = csv.reader(inputfile)
    rows = list(reader)
    del rows[0] # Deletes header

    # Rearrange columns
    timeformat = '%Y-%m-%d %H:%M:%S %z'
    for row in rows:
        row[0] = int(row[0])
        row[1] = datetime.astimezone(datetime.strptime(row[3], timeformat)).strftime(timeformat)
        row[2] = row[5].replace('\n', '\\n').replace('\r', '\\r')
        try:
            row[3] = row[9]
        except IndexError:
            row[3] = ''
        del row[4:]

    rows.sort()

    # Prepare for writing
    lines = []
    for row in rows:
        lines.append(' '.join(map(str, row)).strip())

    timestamp = str(int(time.time()))
    outputfile = open('tweets-output-%s' % timestamp, 'a')
    for line in lines:
        print(line, file=outputfile)
    outputfile.close()


if __name__ == '__main__':
    try:
        inputfile = open(sys.argv[1], 'r')
    except IndexError:
        print(USAGE)
        sys.exit(1)
    main(inputfile)