Question

My app has permissions, and certain tests need to not be run when a particular permission is on, and some tests need to be run when that same permission is on.

Is there a way to do this? or do I need to use a different framework?

Answer 1

The standard way of excluding ‘tests’ in Cucumber is to add a tag to them to identify them, then when invoking Cucumber you specify which tags to include/exclude. In your example, you could tag a specific scenario:

@needs_permission
Scenario: View users billing information

Or tag the whole feature:

@needs_permission
Feature: Administrative area
    Scenario: View users billing information

Or tag certain examples in a scenario outline:

Scenario Outline: Visit a page
    Given I visit "<page>"

Examples: Don't need permission
    | page    |
    | index   |
    | sitemap |

@needs_permission
Examples: Do need permission
    | page  |
    | admin |

Now, when you run Cucumber, you can exclude those tags if necessary:

When the permission is on and you want to run all tests:

cucumber .

When the permission is off and you want to exclude the tests that need it:

cucumber . -t ~@needs_permission

An alternative which I have used with mixed results, if you really don’t know ahead of time, is to mark a step as pending if it doesn’t apply given the current scenario, e.g.

Given /^I visit some page which needs permission$/ do
    pending "Permissions aren't enabled - skipping" unless permissions_enabled?
end

This will mark the step as ‘pending’ which really means ‘not fully implemented’, which isn’t ideal, especially if you have many such steps – it can be difficult to ensure that other unimplemented steps don’t accidentally creep in and get hidden by all the ones you’ve deliberately marked as such.