DEV Community

Manu Gond
Manu Gond

Posted on

Deleting old tweets using Python & Twitter API for a date range

If you wanted to delete your old tweets within a given date range, this post will walk you through the steps using python-twitter & data obtained from your twitter profile
GitHub Link of Jupyter Notebook


1. Twitter data archive:

We will require twitter data to get two attributes of each tweet:-
I) Tweet Id and
II) Tweet date
If you already have these two items in any csv file or any other form then you can skip this step.

To get you twitter data follow this link using desktop site
alternatively you can reach it from Settings & Privacy -> Account -> Download an archive of your twitter data (After request it will take up to 48 Hrs to make your data available for download)
Alt Text
Alt Text

2. Twitter API Credentials

Go to if you have no existing app then create one.
After your app has been created go to it's "Details", make sure the permissions are assigned to Read, Write and Direct Messages , otherwise edit and make the changes

Alt Text

After that you need to generate the Consumer API key and Access Tokens

Alt Text

Handling the Downloaded Twitter Data:-

After extracting the twitter data you will find two folders one is data and other is assets. The HTML file can be used to browse your tweets offline.
Alt Text

Before we use the code we need to make some changes in the tweets.js file located inside assets folder

Alt Text

Current twitter data download service lists all your tweet assets inside a tweet.js file as an array of objects. Since we will want this file to be read in python, so we need to make it a JSON file, hence remove the first line

   window.YTD.tweet.part0 = [ {
Enter fullscreen mode Exit fullscreen mode

and replace it with

   {"data": [ {
Enter fullscreen mode Exit fullscreen mode

for last line you need to add an extra "}" to make it a JSON object, hence your file's last line will look like this:-

  } ]}
Enter fullscreen mode Exit fullscreen mode

save it as a JSON file, for example "editedTweet.js"


The python jupyter notebook can be found at GitHub with this link

1) First make sure to uninstall standalone twitter package and install the python version of that (since the normal twitter package doesn't includes the "twitter.Api()" method), you can do it directly from Jupyter Notebook (you will need to restart notebook after installation)

!pip uninstall twitter
!pip install python-twitter
Enter fullscreen mode Exit fullscreen mode

OR from terminal using

pip uninstall twitter
pip install python-twitter
Enter fullscreen mode Exit fullscreen mode

variables with your own twitter API credentials, passing tweet id string into "deleteTweet(tweetId)" function will delete that tweet. The tweet id is located in the JSON file as "id_str" for each tweet.

# ==================================================================
# Import statements
# ==================================================================

import sys
import time
from datetime import datetime
import os
import twitter
from dateutil.parser import parse
import numpy as np
import pandas as pd
import json

# ==================================================================
# API Credentials
# ==================================================================


# ==================================================================
# Initialize
# ==================================================================

api = twitter.Api(consumer_key = CONSUMER_KEY,
                  consumer_secret = CONSUMER_SECRET,
                  access_token_key = ACCESS_TOKEN_KEY,
                  access_token_secret = ACCESS_TOKEN_SECRET)

# ======================================================================================
# Function to delete tweet by ID
# ======================================================================================

def deleteTweet(tweetId):
     print("Deleting tweet #{0})".format(tweetId))

   except Exception as err:
      print("Exception: %s\n" % err)
Enter fullscreen mode Exit fullscreen mode

3) Read the JSON file into a json variable in python

myData = None
with open('editedTweet.json') as json_file:
    myData = json.load(json_file)
Enter fullscreen mode Exit fullscreen mode

you can now browse the tweet attributes for each tweet in the array, for reference I have printed contents of element located at index 0, here we are only interested in "created_at" and "id_str" attributes:-

{'tweet': {'created_at': 'Thu Sep 11 12:26:39 +0000 2014',
  'display_text_range': ['0', '137'],
  'entities': {'hashtags': [],
   'symbols': [],
   'urls': [{'display_url': '',
     'expanded_url': '',
     'indices': ['115', '137'],
     'url': ''}],
   'user_mentions': []},
  'favorite_count': '0',
  'favorited': False,
  'full_text': 'for galaxy y , galaxy pocket, galaxy ace, galaxy music, galaxy y dous lite and any other low end android device...',
  'id': '510041733099712513',
  'id_str': '510041733099712513',
  'lang': 'en',
  'possibly_sensitive': False,
  'retweet_count': '0',
  'retweeted': False,
  'source': '<a href="" rel="nofollow">Facebook</a>',
  'truncated': False}}
Enter fullscreen mode Exit fullscreen mode

4) We now just need to create an array with id of tweets which needs to be deleted, hence select the date range for tweets which should be deleted, remember to keep the UTC offset into consideration. Use "range_start" and "range_end".

# Range (in UTC offset) within which tweets will be deleted
# =================================================================

range_start = datetime.strptime('Sep 10 00:00:00 +0000 2012','%b %d %H:%M:%S %z %Y')
range_end = datetime.strptime('Sep 10 00:00:00 +0000 2017','%b %d %H:%M:%S %z %Y')

# ==================================================================
# I am creating a list of tweet IDs for consideration, where tweetsToBeDeleted will be
# used for deleting tweet
# ==================================================================

tweetsToBeDeleted = []
tweetsToBeIgnored = []

for element in myData["data"]:
  tweet_post_time = datetime.strptime(element["tweet"]["created_at"],'%a %b %d %H:%M:%S %z %Y')
  if (tweet_post_time>= range_start and tweet_post_time<= range_end ):

Enter fullscreen mode Exit fullscreen mode

5) Finally you can iterate over the array and pass each id from "tweetsToBeDeleted" array into delete function for the tweets to be removed.

for id in tweetsToBeDeleted:
Enter fullscreen mode Exit fullscreen mode

Top comments (4)

shenzen19 profile image
Samir Zidat

Hello @3zadessg I tried it, then I have : 0 9967 when I launch code, do you know what is the problem? thanks

3zadessg profile image
Manu Gond

Hello Samir, if code is throwing exception, can you share the whole stack trace? or maybe screenshot of error.

tofunmi profile image

This is fantastic Manu, thank you so much for sharing! Allowed me to finally clean up my Twitter account. :)

jessicagarson profile image
Jessica Garson

This is great! Thanks for writing this!