corrections
An extendable Python based Twitter bot which will reply to various keyworded tweets.
"Isn't this against the ToS?", I hear you ask. Well, sort of, but sort of not. Yes, it might be covered by the spamming or unsolicited mentions rules, but in my opinion, no more than some asshole going around correcting people manually, depending on your settings.
Example usage
Here's my favourite use of the Corrections bot (and why I actually wrote this): the a lot correction bot:
# -*- coding: utf-8 -*-
from corrections import Correct
from random import choice
class Alot(Correct):
auth = (OAUTH_TOKEN,
OAUTH_SECRET,
CONSUMER_KEY,
CONSUMER_SECRET)
phrases = ["alot"]
cooldown = (60 * 5, 60 * 3)
def reply(self, phrase, user):
return choice([
"@{user}: a lot*",
"@{user}: Surely you mean 'a lot', right?",
"@{user}: I think you meant 'a lot'!",
"@{user}: Come on, use 'a lot', lazy..",
"@{user}: Alot is sad you can't use 'a lot' right..",
"@{user}: There should be a space in there somewhere.. 'a lot'*",
]).format(user=user)
if __name__ == "__main__":
correct = Alot()
correct()Just for testing
Want to make sure your bot is functional but don't want to start spamming tweets yet? Easy! Set dry = True in your Corrections class and Corrections won't actually update your status.
class Alot(Correct):
auth = (OAUTH_TOKEN,
OAUTH_SECRET,
CONSUMER_KEY,
CONSUMER_SECRET)
phrases = [...]
cooldown = (...)
dry = True
def reply(...):
# ...
