This repo is a homegrown twitter bot developed for the harvest and dissemination of resources published by a DSpace RSS feed. While it could feasibly be used to tweet out other RSS-type resource data, this program was designed with DSpace in mind (in particular, MDSOAR).
The primary workflow is as follows:
source.sh set to run at a given time. Will check to see if database tables are appropriately created and will create if not. source.sh will harvest rss feed indicated in script will initiate tweet_store.py which will write out relevant tweet data to tweet.db if resource handle is not already present in archive (i.e. has not already passed through database).
tweet_write.py set to run at a given time. Takes entered data from RSS feed and generates tweet text based on resource handle and title. This relies on the tweepy library.
A few prerequisites & considerations:
Download all libraries indicated in the python files.
Existing database is titled tweet.db and is used running SQLITE3. tweet_store.py will create tables on first run.
tweet_write.py will require your Twitter application keys. If you are not familiar with creating Twitter apps, check out the following guide.
The original vision for this program is that it should be run via crontab with commands to the effect of:
30 4 * * * source /path/to/file/source.sh > dev/null
0 * * * * python3 /path/to/file/tweet_write.py > dev/null