Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with
or
.
Download ZIP
A simple personal ifttt.com clone
Ruby HTML Shell CSS Other
Branch: master
Failed to load latest commit information.
app Remove unnecessary return statement
config Conflict fix.
db Initial commit
doc Initial commit
lib One liner 3. Added current local time return value.
log Initial commit
public Initial commit
script Initial commit
test Initial commit
vendor Initial commit
.gitignore Update Gemfile.lock
.rvmrc Initial commit
Gemfile Add Bootsrap and simple_form gems
Gemfile.lock Update Gemfile.lock
LICENSE Add MIT License
README No longer supported
Rakefile Initial commit
config.ru Initial commit

README

Whenbot - a personal open source ifttt.com clone. (check out
http://ifttt.com)

Whenbot is no longer supported!

Wiki page with more info: https://github.com/ottawaruby/whenbot/wiki/_pages

simple luuunchbot app: https://github.com/redronin/luuunchbot


The idea: A simple web service that runs on Heroku that lets you
listen on certain triggers and perform actions to other web services
based on that.

For example: Whenever I post an Instagram photo, create a Tumblr blog post.

It would be personal - in that it's meant for each person who wants to
run it to clone the repo, and host it on their own Heroku instance.
Heroku provides enough free addons and services that you can get it up
and running and functional without paying for anything (or at the
most, a few bucks a month).

Basic premise is that you have:

Triggers - things that you listen on (new instagram post, a tweet, a
new email, a new RSS feed entry, etc.). This can be a scheduled
polling to a certain API endpoint (heroku offers regular schedule
tasks every 10 minutes), or webhook endpoints.

Parsers - things that take the output from a trigger, and
massages/normalizes the data into a known format that can be consumed
by...

Publishers - things that you publish to (send a tweet, create a Tumblr
post, send an email, etc.)

The basic engine could be a very simple app, and the Triggers,
Parsers, and Publishers can be individual modules/classes. This would
allow people to work on various triggers, publishers, etc. on their
own while still contributing to the project as a whole. There'd also
be UI work to create the actions, configure services, etc.

Most triggers and publishers would just be using various API wrappers
for web services like Flickr, Tumblr, Posterous, etc. This would be a
great way for people to get exposed to APIs from various web services.
Some are complicated, some are pretty simple, so there'd be lots to
choose from.

I created a simple Instagram => Tumblr / Posterous app for
http://luuunch.com which I could show. It uses a ridiculously simple
Tumblr publisher, Posterous publisher and an Instagram webhook
trigger.

An immediate use case for something like this would be to create an
instagram / flickr / etc. feed to add to the OttawaRuby  website.
Something went wrong with that request. Please try again.