sunlightlabs/cub

Name: cub

Owner: Sunlight Labs

Description: Slack interface for Sunlight Foundation tools.

Created: 2015-08-14 16:20:21.0

Updated: 2015-10-09 16:54:03.0

Pushed: 2015-08-16 15:28:57.0

Homepage: null

Size: 136

Language: Python

GitHub Committers

User	Most Recent Commit	# Commits

Other Committers

User	Email	Most Recent Commit	# Commits

README

Cub

Slack integrations for Sunlight's campaign finance data

Cub is designed to poll a variety of Sunlight Foundation tools – Political Ad Sleuth, Realtime Influence Explorer, and Openstates – to deliver a bundle of raw intelligence about noteworthy movement in the world of politics, campaigns, and the state legislature.

When configured and connected to a Postgres database to keep its own copy of certain campaign finance information pulled from Realtime, Cub pushes alerts to a Slack channel on demand using Slack's webhook functionality.

Configuration

• There is a conf.json file. Edit it to include your Sunlight API key (free!) and your Slack Webhooks URL.
• There are shell scripts and SQL statements in admin/ and sql/. Edit them to point to the directories and databases you want. Note that the SQL statements currently use relative paths in a COPY FROM statement. This is almost certainly a terrible idea that will not work for you. Change those to absolute paths.

Cub is designed to run on the same computer as a Postgres server that houses its own copy of some campaign finance information pulled from Realtime. The Realtime app completes a variety of data cleaning and standardization tasks that make it a desirable source for campaign finance data as opposed to working with the raw FEC data. If you want to run Cub from one computer and host a database on another, or if you don't want to store your own data in Postgres, you'll have to do more work on the code.

Running Cub

The scripts that make all the magic happen are shell scripts in admin/ and run.py in the repository root.

There are more elegant ways to run Cub, for example, you could modify it to run as a service. However, a straightforward way that works is to run it as a cronjob.

On a *nix system, type: crontab -e

to edit your cron.

For folks running Cub from a virtualenv, you'll want lines in cron that look like this:

SHELL=/bin/bash

30 03 * * * cd $HOME/cub_path/admin && ./build_sked_tables.sh > /home/your_user_name/path_to_logfile.log

30 04 * * * source $HOME/.bash_profile && workon cub_virtualenv && python run.py > home/your_user_name/path_to_other_logfile.log