Failing test results in an error message from phantomjs about jasmine-rails

[doryphores]

Wouldn't this prevent failing tests to break a build in a CI environment? I investigated preventing this error output before but seem to remember that the exit status code is required for CI.

[jondkinney]

Here's a screenshot of the tmp/runner.html file

[RyanV]

@jondkinney , the error isn't an issue with your setup. The exception is raised when the Kernel#system call returns false on receipt of a non-zero exit status code. The PhantomJS runner returns a exit status code 1 if any specs fail (https://github.com/searls/jasmine-rails/blob/master/lib/assets/javascripts/jasmine-runner.js#L52). Here is a quick patch you can use if it bothers you enough. I also chose to comment out the line that logs the PhantomJS command string before running the specs. Uncomment the line if you want to use the default logging.

# config/initializers/jasmine_rails.rb
if Rails.env.in? %w(test development)
  require 'jasmine_rails/runner'

  module JasmineRails
    module Runner
      class << self
        private
        def run_cmd(cmd)
          # puts "Running `#{cmd}`"
          system(cmd)
        end
      end
    end
  end
end

[RyanV]

I think your right there @doryphores. I overlooked it because the current project i'm testing it on isn't running CI.

[searls]

Closing unless someone's still bugged by this

[juandazapata]

I've hit this very same problem with a small difference. My tests in the browser are green, in the console are green as well, but the command fails with an error soon after the tests are green.

ruby 2.2.2
jasmine 0.12.4
rails 4

[juandazapata]

Ended up by removing jasmine-rails and going with vanilla jasmine gem.