News Twitter Bot
See the bot in action here: @ny_tames.
This code implements a twitter bot that replies to news tweets performing search and replace from a collaborative translation table. Users can contribute to the translation table via DM or @mentions.
The translation table pending_dic.txt is then human reviewed and approved entries are appended to the production table dic.txt.
Setup tweepy credentials
We will be using Tweepy as the python twitter API library as it's easy and convenient.
After creating your bot account, go to the Developer Twitter Page and apply for a developer account. Once granted generate your keys consumer_key, consumer_secret, access_token, access_token_secret and put them into the keys.pyfile.
Setup with Google Cloud Platform
Go to the Google Cloud Console to get started. Set up billing and create a new project. Don't worry about getting billed, the scale of this project should fall well inside the free tier. Once the project is set up, go to the Cloud Functions to begin importing the code that will be executed.
We will be using an HTTP trigger so whenever a request is made to the URL the function will be executed. Choose 128MiB of allocated memory since it's the smallest. Our bot is written on Python so select Python3.7 as the runtime type and send_tweet() as the entry point. The send_tweet() function inside bot.py will actually be the function executed. If this is your first time, you will have to enable Cloud Build API.
Next, upload the files main.py, mentions.py, requirements.txt, dic.txt, pending_dic.txt, select bot as the entry point and hit deploy. Now we have a function that gets executed everytime someone visits the HTTP trigger, it is time to create an automatic trigger.
First go into the trigger section of the function and copy the trigger URL to your clipboard. Now go into the Google Cloud Scheduler and Create a job, select a region. Give it a name, a UNIX crontab format and HTTP target.