This is an errbot (http://errbot.net) backend for Cisco Spark (https://www.ciscospark.com/)

Requirements

Python Requests (https://github.com/kennethreitz/requests)

Python Websocket-client (https://github.com/liris/websocket-client)

Installation

git clone https://github.com/panholt/err-backend-cisco-spark.git

Configuration:

BACKEND = 'Cisco Spark'
BOT_EXTRA_BACKEND_DIR = '/path_to/err-backend-cisco-spark'

to your config.py

Authentication

The BOT_IDENTITY needs to be updated to contain the personId, Bearer Token, and email address of the bot Create bot accounts at https://developer.ciscospark.com/add-bot.html

BOT_IDENTITY = {'token' : ''} # Bearer Token belonging to the bot

The BOT_ADMINS key takes the email address associated with the Cisco Spark Account

BOT_ADMINS = ('userid@email.com')    # Email Address of bot admins

Websocket Proxy

The WEBSOCKET_PROXY key is used to setup the Websocket.

##Extra Configuration Options Cisco Spark has a message limit of 7439 characters for the messages API. Errbot should be configured to respect this limit:

MESSAGE_SIZE_LIMIT = 7439

Cisco Spark Webhooks for bots only trigger in 1:1 rooms with the bot, or if the bot is @mentioned in a group room. This backend will automatically register alternate prefixes for the @mention patterns so no additional prefix would be needed. For example "@bot status" in a group room would trigger the webhook and the message payload would read "bot status", and an alternate prefix of "bot" is injected to the Errbot configuration.

Contributing

Pull requests welcome