Archive your tweets in a relaxing way.

1. Request and download your Twitter Archive file, from the Twitter Settings page.

2. Extract the files somewhere e.g. ~/Desktop/tweets.

3. Git clone or download this repo.

4. In a terminal, go to repo directory, and do this:

   npm install
   

5. To test this locally, install CouchDB and start it. Else, get a hosted CouchDB instance. I use Cloudant, create a new database called tweets and set its permissions to read-only for Everyone else.

6. To import the Archive data into CouchDB, do the steps below.

   node scripts/archive2couch.js -h # Shows help
   node scripts/archive2couch.js -f <data/js/tweets folder> -c <https://user:pass@localhost:5984>
   

7. After the import is done, do this to create the design views:

   node scripts/create-views.js -c <https://user:pass@localhost:5984>
   

8. You're basically done. You can manually update CouchDB on a periodical basis by going through these steps again. Optionally, you can continue with the steps below to host this app on Heroku and run the updater on Scheduler to sync the latest tweets from Twitter to CouchDB.

1. Fill up the config required. This app uses nconf, thus there can be multiple configuration sources e.g. process.argv, process.env and config.json. For Heroku, it's recommended to use environment variables.

   heroku config:set user_id=<1234>
   heroku config:set couchdb_url=https://user:pass@localhost:5984
   heroku config:set consumer_key=XXX
   heroku config:set consumer_secret=XXX
   heroku config:set access_token=XXX
   heroku config:set access_token_secret=XXX
   

   • To get your Twitter user ID, use idfromuser.com.
   • Set up a new app to get your consumer key and secret. Create the access token and secret from the app's details page.

2. Deploy this app to Heroku.

3. Check if things are working by doing this:

   heroku run node twitter2couch.js
   

4. Add Heroku Scheduler.

5. Go to add-on page, add a job, set task to node twitter2couch.js, at an hourly frequency. Save.

6. You're done. Make sure it works properly by doing heroku logs.

Go to github.com/cheeaun/tweets.

1. I assume the archived tweets are from a Public account. The steps above might differ slightly for private account users. Pull requests are welcomed.
2. In Cloudant, you can set up a different username and password to be used for couchdb_url. Go to Permissions page and click Generate API key. The API key is also the username.
3. The JSON responses from Twitter API are more detailed than the ones provided by Twitter Archive. Everything should still work nevertheless.

Some tweets from the Twitter Archive have the wrong created_at timestamps. This is probably a known bug that's already reported in the developers forum.

To fix those tweets, run:

node scripts/fix-tweets.js -c <https://user:pass@localhost:5984>

This script will go through all tweets that contains 00:00:00 +0000 timestamps and query the Twitter API to get the real timestamps to replace them.

There could be some tweets that can't be fixed, due to reasons:

• Those tweets are no longer available. Usually retweeted tweets from a deleted account.
• The timestamp of that tweet is actually legit.

Licensed under the MIT License.

• Simplebird (Node.js) https://github.com/josephschmitt/simplebird
• Tweet Nest (PHP + MySQL) https://github.com/graulund/tweetnest - with many forks that support Twitter Archive
• Archive My Tweets (PHP + MySQL) https://github.com/amwhalen/archive-my-tweets
• Tweetarchive (Go + PostgreSQL) https://github.com/paulsmith/tweetarchive
• ElasticTweets (ElasticSearch) https://github.com/AdaTheDev/ElasticTweets
• GrailbirdUpdater (Ruby) https://github.com/DeMarko/grailbird_updater
• Twimo (Node.js + MongoDB) https://github.com/rowanmanning/twimo
• Twarch (PHP + SQLite) https://github.com/TomNomNom/twarch
• Twive (PHP + MySQL) https://github.com/robhogg/twive
• TwitterToWord (Java) https://github.com/odessa2/TwitterToWord
• Parse Twitter Archive (Python) https://github.com/mshea/Parse-Twitter-Archive
• Couch-tweet-archiver (Node.js + CouchDB) https://github.com/mlc/couch-tweet-archiver
• TweetDen (Ruby + MongoDB) https://github.com/philoye/tweetden
• Keep your Twitter Archive fresh on Google Drive using a bit of Google Apps Script