Delete your tweet history
Twitforget uses the Twitter API to delete your tweet history.
You need to authorise the app with a Twitter access token for your Twitter account. Follow the instructions here: https://dev.twitter.com/oauth/overview/application-owner-access-tokens
Add the access token to a configuration file, (defaults to ~/.twitrc) like
this:
[twitter]
token = <token>
token_key = <token_key>
con_secret = <con_secret>
con_secret_key = <con_secret_key>
You can bootstrap twitforget by asking Twitter for an archive of all your tweets. Twitforget can then read in this archive and use it to keep track of your tweets from that point onwards, deleting your history as you go.
To request your Twitter archive, follow the instructions here.
Once you've downloaded the archive ZIP file, load it with twitforget like this:
./twitforget.py -i <zipfile> --nofetch --nodelete <your_twitter_handle>
This will just load the data from the archive and won't do anything else.
You're now ready to start deleting your tweets!
Twitforget fetches a list of your tweet history, and then deletes tweets using a few different methods, depending on how you want to maintain your history.
If you just want to test things without actually deleting any tweets,
use the --dryrun flag.
The default method twitforget uses is to keep all but your last n tweets. You
can tell twitforget how many tweets to keep using the -K flag. It defaults to keeping the last 5000 tweets.
Use the --before-days flag to tell twitforget to delete tweets sent more
than n days ago. Use it to keep anything more recent.
Use the -A/--after-days and the -B/--before-days flags
to define a date range to delete tweets. Twitforget will delete all the tweets
before --before-days and after --after-days.
You might particularly like certain tweets and would prefer to keep them. No
problem! Tell twitforget to keep those tweets using the -k flag and providing
the tweet id for the good tweet. You can provide the flag multiple times if you
have more than one good tweet you'd like to keep.
That can get a bit tedious when you have a lot of good tweets you want to keep,
so you can also add the tweet ids to your config file with the keeptweets
parameter and twitforget will always know to keep those tweets.
Something like this:
keeptweets = 825526299879829504
846940673173540864
846941755597574145
869115756755034112
875394335298142208
(These are real tweets of mine that I've decided to keep, so you can go check.)
If you make a new, good tweet and decide to keep it, just add it to the list.
You can also delete your likes with the likesforget.py command.
It's based on twitforget.py so it functions pretty much the same way, with
the same flags and approach to things, just adapted to work with likes.
Import your likes archive in the same way you did for your tweets:
./likesforget.py -i <zipfile> --nofetch --nodelete <your_twitter_handle>
Twitter doesn't provide much detail for likes, so the script doesn't currently populate the cache with creation times.
Just as with keeping tweets, you might want to keep certain likes. Use the
-k flag, or add the id of the tweets you've liked to your config file with
the keeplikes parameter and those likes will be kept.
Twitforget has a limitation imposed by Twitter: Twitter only lets you see your last 3200 tweets.
If you've sent more tweets than that since the last time twitforget added tweets to its database, you can't delete tweets older than the 3200 limit unless you get a new archive from twitter and augment your current tweetcache.
twitforget won't overwrite existing information in the tweetcache, so you can safely load in a Twitter archive file as many times as you like. It'll only add missing tweets.
If you're a verified user, it appears you can see more of your own history than a non-verified user, so twitforget can delete more of your history.
Twitforget deliberately runs a little bit slower than what the Twitter API limits allow (at time of writing) so the first run of deleting might take a little while, depending on how many tweets you have to delete.
twitforget is design to run in the background on a schedule once you've done the initial load and purge of setup.
twitforget stores a very basic summary of your tweet history in a SQLlite
database stored by default in your home directory in ~/.tweetcache.db.
It only stores the tweet id, handle, tweet creation datetime, the tweet text as returned by Twitter, and whether or not twitforget thinks it's deleted the tweet.
If you want to store a full archive of your tweets, keep a copy of what Twitter sends you when you first set up this tool, and then periodically request an updated archive before you delete tweets with twitforget.
One day we might start storing the full JSON payload returned from Twitter when we fetch tweet info, but keeping tweets isn't really what twitforget is about.