Skip to content

HTTPS clone URL

Subversion checkout URL

You can clone with HTTPS or Subversion.

Download ZIP
ActiveRecord scenarios
Ruby
branch: master

Fetching latest commit…

Cannot retrieve the latest commit at this time

Failed to load latest commit information.
bin
examples
lib
rails_generators/scenario
spec
.gitignore
LICENSE
NOTICE
README.rdoc
Rakefile
VERSION.yml
run-specs
scenarios.gemspec

README.rdoc

scenarios

The idea behind 'scenarios' is simple. Scenarions are basically just organized bits of arbitrary ruby code.

why?

Here's why scenarios were originally written … you may use them however you wish!

I had a Rails application and I wanted to be able to do something like this:

$ rake scenarios:load SCENARIO=one_thousand_users

… which would put 1,000 users in the database.

Sometimes, while testing out my application's functionality in a browser, I would want to setup some scenario to test. Maybe I'm working on X feature and it would be helpful if I had Y data in the database to work with. Typically, I would open up `script/console` and create abunchof Y data manually. But … wouldn't it be useful if we could automate that?

Also, our application requires some data to be in the database for the app to load properly. So we created a scenario that loads up the minimum amount of data required to run the app!

documentation

Read the RDoc

how?

$ sudo gem install remi-scenarios -s http://gems.github.com

then, if you're in a Rails project …

$ ./script/generate scenario Foo

to see all defined scenarios …

$ rake scenarios

to load a particular scenario …

$ rake scenarios:load NAME=my_scenario
$ rake scenarios:load NAME=my_scenario,and_some_more_scenarios

what else?

more documentation to come soon!

TODO

  • add a 'scenario' bin script so you don't necessarily have to use rake? potentially? maybe check for local .scenariorc?

  • should be able to 'show' a full, long description of what a scenario does (full header comment)

  • document the context that scenarios are loaded within … right now we're using #load, which is nice because it gets line numbers for exceptions … altho eval'ing the code gives us more control over the context its loaded in

  • add dependencies so one scenario can call another one

  • make the scenario directory for the generator configurable (and that dir should always be included in the Scenario loadpath

  • make all scenarios load within a specific context and add hooks so people can update/modify that context

  • write a spec to make sure that the load paths work properly and multiple directories can have scenarios with the same name … the directories added last should override the previous ones. also make sure paths are uniq … if i add 'foo' and 'foo' is already there, it should make foo the last path (move it to the end)

Something went wrong with that request. Please try again.