AsyncBots

Features

AsyncBots provides an asyncio based interface for writing Slack Real Time Messaging () chat bots with almost no code other than the core functionality. All user defined bots run through a single bot user on Slack, with AsyncBots directing messages to the appropriate bots.

Bots consist of a user defined trigger (e.g. !mycommand or !simulate <user>) and a function which will be called when a user says the trigger in chat. When activated, the bot can then send messages, add reactions, upload files, and more. Commands are defined using , which allows for filtering of messages without adding complexity.

In channels the bot is present in, channel history is (optionally) saved, including beyond the 10k message limit imposed on free teams.

Examples

Defining a simple bot

The simplest bot consists of a class with a constructor and one method.

from asyncbots.bot import SlackBot, register
from asyncbots.command import MessageCommand
from pyparsing import alphas, Word

class MyBot(SlackBot)
    def __init__(self, slack=None):
        # Call parent's constructor
        super(MyBot, self).__init__(self, slack=slack)
        self.name = 'My Bot'
        self.expr = 'greet' + Word(alphas).setResultsName('user')

    @register()
    async def handler(self, sender, channel, parsed_message):
        # Send a reply in the channel this command was received in
        return MessageCommand('Hello ' + parsed_message['user'])

The fields self.name and self.expr are used to register the command with the Slack core. To use other field names (for instance if you want multiple commands on one class), simply pass them to @register:

...
class MyBot(SlackBot)
    def __init__(self, slack=None):
        ...
        self.another_name = 'Airspeed measurement'

        self.another_expr = 'airspeed'

    @register(name='another_name', expr='another_expr')
    async def get_airspeed(self, sender, channel, parsed_message):
        return MessageCommand('African or European?')

Connecting to Slack

In order to connect to Slack, construct an instance of asyncbots.slack_api.Slack:

import asyncio
from asyncbots.slack_api import Slack
...
def main():
    slack = Slack()
    MyBot(slack=slack)

    loop = asyncio.get_event_loop()
    loop.run_until_complete(slack.run())

The above method of constructing the Slack object assumes that the SLACK_TOKEN and SLACK_BOT_NAME environment variables are set to the Slack API token and bot's name in Slack respectively.

Enabling logging

To enable message logging, create a connection to a mongoDB server using mongoengine.

from mongoengine import connect
from asyncbots.slack_api import Slack, SlackConfig
...
def main():
    connect('mydbname')
    config = SlackConfig(db=True)
    slack = Slack(config)
    ...