Push2heroku

Makes it easy to push to heroku.

Installation

Add this to Gemfile:

gem 'push2heroku'

Usage

After installing the gem copy this file to config/push2heroku.yml .

Change the contents of config/push2heroku.yml as per your needs after reading rest of README.

What problem it solves

Here at BigBinary we create a separate branch for each feature we work on. Let's say that I am working on authentication with facebook. When I am done with the feature then I send pull request to my team members to review. However in order to review the work all the team members need to pull down the branch and fire up rails server and then review.

We like to see things working. So we developed push2heroku to push a feature branch to heroku instantly with one command. Executing push2heroku prints a url and we put that url in the pull request so that team members can actually test the feature.

Here is how it works

push2heroku reads the push2heroku.yml and executes those commands. It's that simple.

Lets say that I am working in a branch called 76-facebook-authentication. When I execute push2heroku then the application name under which it will be deployed to heroku will be nimbleshop-76-facebook-neeraj.

nimbleshop is the name of the project. 76-facebook is the first 10 letters of the branch name. neeraj is the first 5 letters of my github user name.

So in this case the url of the application will be http://nimbleshop-76-facebook-neeraj.herokuapp.com .

In the push2heroku.yml file the keys production and staging are branch names. And these branches are special branches. For these branches the url generated will be just the application name and the branch name. For example if I execute rake push2heroku from staging branch then the heroku url will be http://nimbleshop-staging.herokuapp.com.

However if I delete staging key from push2heroku.yml then staging is no longer a special branch and the heroku url would be http://nimbleshop-staging-neeraj.herokuapp.com .

License

push2heroku is released under MIT License.

The new design of push2heroku is very flexible. Let's say that Artem wants to test something then he can add to push2heroku.yml something like this

regenerate_images:
  - bundle exec heroku run rake db:regenerate_images --app <%=ENV['HEROKU_APP_NAME']%> --trace
  - bundle exec heroku run rake db:post_image_cleanup --app <%=ENV['HEROKU_APP_NAME']%> --trace

Now to execute all the commands under key regenrate_images all he has to do is

rake push2heroku CALLBACKS=reset_db_using_fixtures,regenerate_images

Just comma separate all the tasks. It's that simple. Now push2heroku gives all the control to the developer.