Download all your tweets and pickle them

twitter_download.py

# This script will download all your (or the logged in user's) tweets and save it as a serialized pickle object

import tweepy
import pickle

# Go to developer.twitter.com, apply to use the API, create a project and get the authorisation details below
API_KEY = ''
API_KEY_SECRET = ''
BEARER_TOKEN = ''
CLIENT_ID = ''
CLIENT_SECRET = ''
ACCESS_TOKEN = ''
ACCESS_TOKEN_SECRET = ''

def authorize_twitter():
    # Create a client with auth params to access API V2
    client = tweepy.Client(BEARER_TOKEN,API_KEY, API_KEY_SECRET, ACCESS_TOKEN, ACCESS_TOKEN_SECRET)

    # Create an api object with auth params to access API V1
    auth = tweepy.OAuth1UserHandler(API_KEY,API_KEY_SECRET,ACCESS_TOKEN,ACCESS_TOKEN_SECRET)
    api = tweepy.API(auth)

    return client, api

def get_all_tweets_for_user_and_pickle_it():
    # Authorise Twitter
    client, api = authorize_twitter()
    all_tweets = []

    # Get the current logged in user's timeline
    tweets = api.user_timeline(exclude_replies=True, include_rts=False, count=200)
    all_tweets.extend(tweets)

    # Grab oldest tweet in current list to use in the pagination loop
    oldest = all_tweets[-1].id - 1

    # Loop to retrieve older tweets till no more exist
    while len(tweets) > 0:
        print(f'Getting tweets before {oldest}')

        tweets = api.user_timeline(exclude_replies=True, include_rts=False, max_id=oldest, count=200)
        all_tweets.extend(tweets)
        oldest = all_tweets[-1].id - 1

        print(f"...{len(all_tweets)} tweets downloaded so far")

    # Pickle it
    with open('all_tweets.pkl', 'wb') as f:
        pickle.dump(all_tweets, f)

get_all_tweets_for_user_and_pickle_it()