tweet_fetch.py

#!/usr/bin/env python3
# encoding: utf-8

# Based on https://gist.github.com/yanofsky/5436496
# Made by @jacksquaredson

import argparse
import csv
import time
import tweepy #https://github.com/tweepy/tweepy


def main(args):
	#authorize twitter, initialize tweepy
	auth = tweepy.OAuthHandler(args.consumer_key, args.consumer_secret)
	auth.set_access_token(args.access_key, args.access_secret)
	api = tweepy.API(auth)

	no_retweets = args.no_retweets

	# fetch the tweet ids
	with open('tweets.csv', 'r') as f_in:
		with open('tweet_texts', 'w') as f_out:
			reader = csv.reader(f_in)
			writer = csv.writer(f_out)
			next(reader) # Skip the header
			for idx, tweet in enumerate(reader):
				tweet_text = get_tweet_text(no_retweets, api, tweet[0])
				if tweet_text:
					writer.writerow([tweet_text])
				if not idx%10 and idx>0:
					print('Handled {idx} tweets'.format(idx=idx))

def get_tweet_text(no_retweets, api, id):
	tweet = get_tweet_by_id(api, id)
	if tweet.retweeted:
		if no_retweets:
			return None
		else:
			return tweet.retweeted_status.full_text
	else:
		return tweet.full_text

def get_tweet_by_id(api, id):
	delay_power = 0
	while True:
		try:
			return api.get_status(id, tweet_mode='extended')
		except tweepy.error.RateLimitError:
			delay_power += 1
			time.sleep(2**delay_power)


if __name__ == '__main__':

	parser = argparse.ArgumentParser()
	parser.add_argument('--no-retweets', '-n', default=False, action='store_true',
		help='Only store "original" tweets - no RTs')
	parser.add_argument('--consumer-key', required=True,
		help='Twitter API Consumer Key')
	parser.add_argument('--consumer-secret', required=True,
		help='Twitter API Consumer Secret')
	parser.add_argument('--access-key', required=True,
		help='Twitter API Access Key')
	parser.add_argument('--access-secret', required=True,
		help='Twitter API Access Secret')
	args = parser.parse_args()

	main(args)