Skip to content

Latest commit

 

History

History
212 lines (152 loc) · 6.43 KB

README.md

File metadata and controls

212 lines (152 loc) · 6.43 KB
Raw

PyPI Build Status

A chat bot for Slack inspired by llimllib/limbo and will.

Features

Installation

sudo pip install slackbot

Usage

Generate the slack api token

First you need to get the slack api token for your bot. You have two options:

  1. If you use a bot user integration of slack, you can get the api token on the integration page.
  2. If you use a real slack user, you can generate an api token on slack web api page.

Configure the bot

First create a slackbot_settings.py and a run.py in your own instance of slackbot.

Configure the api token

Then you need to configure the API_TOKEN in a python module slackbot_settings.py, which must be located in a python import path. This will be automatically imported by the bot.

slackbot_settings.py:

API_TOKEN = "<your-api-token>"

Alternatively, you can use the environment variable SLACKBOT_API_TOKEN.

Run the bot
from slackbot.bot import Bot
def main():
    bot = Bot()
    bot.run()

if __name__ == "__main__":
    main()
Configure the default answer

Add a DEFAULT_REPLY to slackbot_settings.py:

DEFAULT_REPLY = "Sorry but I didn't understand you"
Configure the docs answer

The message attribute passed to your custom plugins has an special function message.docs_reply() that will parse all the plugins available and return the Docs in each of them.

Send all tracebacks directly to a channel, group, or user

Set ERRORS_TO in slackbot_settings.py to the desired recipient. It can be any channel, group, or user. Note that the bot must already be in the channel. If a user is specified, ensure that they have sent at least one DM to the bot first.

ERRORS_TO = 'some_channel'
# or...
ERRORS_TO = 'username'
Configure the plugins

Add your plugin modules to a PLUGINS list in slackbot_settings.py:

PLUGINS = [
    'slackbot.plugins',
    'mybot.plugins',
]

Now you can talk to your bot in your slack client!

Attachment Support

from slackbot.bot import respond_to
import re
import json


@respond_to('github', re.IGNORECASE)
def github():
    attachments = [
    {
        'fallback': 'Fallback text',
        'author_name': 'Author',
        'author_link': 'http://www.github.com',
        'text': 'Some text',
        'color': '#59afe1'
    }]
    message.send_webapi('', json.dumps(attachments))

Create Plugins

A chat bot is meaningless unless you can extend/customize it to fit your own use cases.

To write a new plugin, simply create a function decorated by slackbot.bot.respond_to, slackbot.bot.listen_to, or slackbot.bot.idle:

  • A function decorated with respond_to is called when a message matching the pattern is sent to the bot (direct message or @botname in a channel/group chat)
  • A function decorated with listen_to is called when a message matching the pattern is sent on a channel/group chat (not directly sent to the bot)
  • (development version only) A function decorated with idle is called whenever a message has not been sent for the past second
from slackbot.bot import respond_to, listen_to, idle
import re

@respond_to('hi', re.IGNORECASE)
def hi(message):
    message.reply('I can understand hi or HI!')
    # react with thumb up emoji
    message.react('+1')

@respond_to('I love you')
def love(message):
    message.reply('I love you too!')

@listen_to('Can someone help me?')
def help(message):
    # Message is replied to the sender (prefixed with @user)
    message.reply('Yes, I can!')

    # Message is sent on the channel
    # message.send('I can help everybody!')
    
last_bored = time.time()
@idle
def bored(client):
    if time.time() - last_bored >= 30:
        last_bored = time.time()
        
        # Messages can be sent to a channel
        client.rtm_send_message('some_channel', "I'm bored!")
        # Or directly to a user
        client.rtm_send_message('some_user', "Hey, entertain me!")
        
        # If a name is ambiguous:
        client.rtm_send_message(client.find_channel_by_name('ambiguous'), "To ambiguous the channel")
        client.rtm_send_message(client.find_user_by_name('ambiguous'), "To ambiguous the user")
        
        # Attachments can be sent with `client.rtm_send_message(..., attachments=attachments)`.

To extract params from the message, you can use regular expression:

from slackbot.bot import respond_to

@respond_to('Give me (.*)')
def giveme(message, something):
    message.reply('Here is {}'.format(something))

If you would like to have a command like 'stats' and 'stats start_date end_date', you can create reg ex like so:

from slackbot.bot import respond_to
import re


@respond_to('stat$', re.IGNORECASE)
@respond_to('stat (.*) (.*)', re.IGNORECASE)
def stats(message, start_date=None, end_date=None):

And add the plugins module to PLUGINS list of slackbot settings, e.g. slackbot_settings.py:

PLUGINS = [
    'slackbot.plugins',
    'mybot.plugins',
]

The @default_reply decorator

added in slackbot 0.4.1

Besides specifying DEFAULT_REPLY in slackbot_settings.py, you can also decorate a function with the @default_reply decorator to make it the default reply handler, which is more handy.

@default_reply
def my_default_hanlder(messsage):
    message.reply('...')

Here is another variant of the decorator:

@default_reply(r'hello.*)')
def my_default_hanlder(messsage):
    message.reply('...')

The above default handler would only handle the messages which must (1) match the specified pattern and (2) can't be handled by any other registered hanlder.

List of third party plugins

You can find a list of the available third party plugins on this page.