JackSzwergold
commented
2 years ago Yes! This can be done right now if you uninstall the default PIP package version that is out there right now and instead install the latest version that is right here in this GitHub repo. This GitHub repo has an import function that doesn’t exist in the standard pip3 install version.
I followed this process and in my case I had 11,000+ Tweets a few days ago and now I trimmed it down to 1,800-ish Tweets. Works like a charm!
First, uninstall the version of semiphemeral you initially installed via PIP like this:
pip3 uninstall semiphemeralThen reinstall it via this GitHub repo like this:
pip3 install git+ssh://git@github.com/micahflee/semiphemeral.gitThis will install the current master branch in this repo. To check and see if it was properly installed, just run the semiphemeral without anything else and the help text will read as follows; note the new import option:
The output should look like this:
Automatically delete your old tweets, except for the ones you want to keep
Options:
--help Show this message and exit.
Commands:
configure Start the web server to configure semiphemeral
delete Delete tweets that aren't automatically or manually
excluded, likes, and DMs
delete_dms Delete DMs that aren't available through the Twitter API
excluded_export Export tweets excluded that are excluded from deletion
excluded_import Import tweets excluded that are excluded from deletion
fetch Download all tweets/DMs
import Import tweets from a Twitter data export
stats Show stats about tweets in the database
unlike Delete old likes that aren't available through the Twitter
APINow if that all worked well and import is in there, you are ready to do the next step… Actually import your Tweets from the Twitter archive.
First, unZip the archive and now run the command as follows:
semiphemeral import /path/to/twitter/archiveNote that this might fail because semiphemeral is looking for the file data/tweet.js when it seems like the current Twitter archive format has that set as data/tweets.js.
The quick and dirty solution is to copy or rename data/tweets.js to data/tweet.js and run that command again.
If all goes well, the semiphemeral import process will show data just like the semiphemeral fetch process. Just note that while the import grinds through data, it will still be live-checking the Twitter API to fetch data for each Tweet. So it might take a while.
But once it is done, all of the Tweets will be stored locally on your system and ready for you to trim and cull them as you see fit.
dippedrusk
Alphakilo
I was one of the users who got hit by the weird Twitter API thing where only ~3000 tweets got downloaded and deleted.
I have my Twitter archive with all my tweet IDs and since you do a similar thing on the hosted version with DMs I thought maybe it would be fairly easy to set up a similar thing to delete all old tweets from the
data/tweet.jsfile of an archive.Update: meanwhile (since i didn't want to give twitter my phone number to change my API access level and then play with this repository) I wrote a horrible ugly selenium script to go through a list of tweet IDs from my archive and delete things for me in a browser, slowly, as i watch :P