simplecov 0.6.0 → 0.22.0

data/features/config_merge_timeout.feature

@@ -1,38 +0,0 @@
-@test_unit @rspec @merging @config
-Feature:
-
-  The maximum time between resultset merges can be customized
-  using SimpleCov.merge_timeout SECONDS. This can be helpful for
-  long-running test-suites that fail to merge because of the time
-  between individual suite finishes is more then the default timeout
-  of 10 minutes.
-
-  Here, for the sake of testing the opposite case is shown, choosing
-  a merge timeout so short that the first test suite's results actually
-  are out of date when the second suite finishes and thus does not end up
-  in the report.
-
-  Scenario:
-    Given SimpleCov for Test/Unit is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start do
-        merge_timeout 1
-      end
-      """
-    Given SimpleCov for RSpec is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start do
-        merge_timeout 1
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then the report should be based upon:
-      | Unit Tests |
-
-    When I open the coverage report generated with `bundle exec rspec spec`
-    Then the report should be based upon:
-      | RSpec |
-

data/features/config_nocov_token.feature

@@ -1,79 +0,0 @@
-@test_unit @nocov
-Feature:
-
-  Code wrapped in # :nocov: will be ignored by coverage reports.
-  The name of the token can be configured with SimpleCov.nocov_token or SimpleCov.skip_token
-
-  Scenario: Custom nocov token using nocov_token
-    Given SimpleCov for Test/Unit is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start 'test_frameworks' do
-        nocov_token 'skippit'
-      end
-      """
-
-    Given a file named "lib/faked_project/nocov.rb" with:
-      """
-      class SourceCodeWithNocov
-        # :skippit:
-        def some_weird_code
-          never_reached
-        rescue => err
-          but no one cares about invalid ruby here
-        end
-        # :skippit:
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-
-    Then I should see the source files:
-      | name                                    | coverage |
-      | lib/faked_project.rb                    | 100.0 %  |
-      | lib/faked_project/some_class.rb         | 80.0 %   |
-      | lib/faked_project/framework_specific.rb | 75.0 %   |
-      | lib/faked_project/meta_magic.rb         | 100.0 %  |
-      | lib/faked_project/nocov.rb              | 100.0 %  |
-
-    And there should be 5 skipped lines in the source files
-
-    And the report should be based upon:
-      | Unit Tests |
-
-  Scenario: Custom nocov token using skip_token
-    Given SimpleCov for Test/Unit is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start 'test_frameworks' do
-        skip_token 'skippit'
-      end
-      """
-
-    Given a file named "lib/faked_project/nocov.rb" with:
-      """
-      class SourceCodeWithNocov
-        # :skippit:
-        def some_weird_code
-          never_reached
-        rescue => err
-          but no one cares about invalid ruby here
-        end
-        # :skippit:
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-
-    Then I should see the source files:
-      | name                                    | coverage |
-      | lib/faked_project.rb                    | 100.0 %  |
-      | lib/faked_project/some_class.rb         | 80.0 %   |
-      | lib/faked_project/framework_specific.rb | 75.0 %   |
-      | lib/faked_project/meta_magic.rb         | 100.0 %  |
-      | lib/faked_project/nocov.rb              | 100.0 %  |
-
-    And there should be 5 skipped lines in the source files
-
-    And the report should be based upon:
-      | Unit Tests |

data/features/config_project_name.feature

@@ -1,27 +0,0 @@
-@test_unit @config
-Feature:
-
-  SimpleCov guesses the project name from the project root dir's name.
-  If this is not sufficient for you, you can specify a custom name using
-  SimpleCov.project_name('xyz')
-
-
-  Scenario: Guessed name
-    Given SimpleCov for Test/Unit is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then I should see "Code coverage for Project"
-
-  Scenario: Custom name
-    Given SimpleCov for Test/Unit is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start { project_name "Superfancy 2.0" }
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then I should see "Code coverage for Superfancy 2.0"

data/features/config_styles.feature

@@ -1,93 +0,0 @@
-@test_unit @config
-Feature:
-
-  There's several ways to configure SimpleCov. All of those
-  config schemes below are equivalent and can be chosen by personal
-  preference or project requirements.
-
-  Background:
-    Given SimpleCov for Test/Unit is configured with:
-      """
-      require 'simplecov'
-      """
-
-  Scenario: Inside start block
-    Given a file named ".simplecov" with:
-      """
-      SimpleCov.start do
-        add_filter 'test'
-        command_name 'Config Test Runner'
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then I should see "4 files in total."
-    And I should see "using Config Test Runner" within "#footer"
-
-  Scenario: Explicitly before start block
-    Given a file named ".simplecov" with:
-      """
-      SimpleCov.add_filter 'test'
-      SimpleCov.command_name 'Config Test Runner'
-      SimpleCov.start
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then I should see "4 files in total."
-    And I should see "using Config Test Runner" within "#footer"
-
-  Scenario: Explicitly after start block
-    Given a file named ".simplecov" with:
-      """
-      SimpleCov.start
-      SimpleCov.add_filter 'test'
-      SimpleCov.command_name 'Config Test Runner'
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then I should see "4 files in total."
-    And I should see "using Config Test Runner" within "#footer"
-
-  Scenario: Using configure block after start
-    Given a file named ".simplecov" with:
-      """
-      SimpleCov.start
-      SimpleCov.configure do
-        add_filter 'test'
-        command_name 'Config Test Runner'
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then I should see "4 files in total."
-    And I should see "using Config Test Runner" within "#footer"
-
-  Scenario: Using configure block before start
-    Given a file named ".simplecov" with:
-      """
-      SimpleCov.configure do
-        add_filter 'test'
-        command_name 'Config Test Runner'
-      end
-      SimpleCov.start
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then I should see "4 files in total."
-    And I should see "using Config Test Runner" within "#footer"
-
-  Scenario: Mixing configure and start block config
-    Given a file named ".simplecov" with:
-      """
-      SimpleCov.configure do
-        command_name 'Config Test Runner'
-      end
-      SimpleCov.start do
-        add_filter 'test'
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then I should see "4 files in total."
-    And I should see "using Config Test Runner" within "#footer"
-

data/features/cucumber_basic.feature

@@ -1,29 +0,0 @@
-@cucumber
-Feature:
-
-  Simply adding the basic simplecov lines to a project should get
-  the user a coverage report after running `cucumber features`
-
-  Scenario:
-    Given SimpleCov for Cucumber is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start
-      """
-
-    When I open the coverage report generated with `bundle exec cucumber features`
-    Then I should see the groups:
-      | name      | coverage | files |
-      | All Files | 91.23%   | 6     |
-
-    And I should see the source files:
-      | name                                    | coverage |
-      | lib/faked_project.rb                    | 100.0 %  |
-      | lib/faked_project/some_class.rb         | 80.0 %  |
-      | lib/faked_project/framework_specific.rb | 75.0 %   |
-      | lib/faked_project/meta_magic.rb         | 100.0 %  |
-      | features/step_definitions/my_steps.rb   | 100.0 %  |
-      | features/support/simplecov_config.rb    | 100.0 %  |
-
-    And the report should be based upon:
-      | Cucumber Features |

data/features/merging_test_unit_and_rspec.feature

@@ -1,44 +0,0 @@
-@test_unit @rspec @merging
-Feature:
-
-  Test suites like RSpec and Test/Unit should be merged automatically
-  when both have been run recently. The coverage report will feature
-  the joined results of all test suites that are using SimpleCov.
-
-  Scenario:
-    Given SimpleCov for Test/Unit is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start do
-        add_filter 'test.rb'
-        add_filter 'spec.rb'
-      end
-      """
-    And SimpleCov for RSpec is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start do
-        add_filter 'test.rb'
-        add_filter 'spec.rb'
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-    Then the report should be based upon:
-      | Unit Tests |
-
-    When I open the coverage report generated with `bundle exec rspec spec`
-    Then the report should be based upon:
-      | RSpec      |
-      | Unit Tests |
-
-    And I should see the groups:
-      | name      | coverage | files |
-      | All Files | 90.48%    | 4     |
-
-    And I should see the source files:
-      | name                                    | coverage |
-      | lib/faked_project.rb                    | 100.0 %  |
-      | lib/faked_project/some_class.rb         |  80.0 %  |
-      | lib/faked_project/framework_specific.rb |  87.5 %  |
-      | lib/faked_project/meta_magic.rb         | 100.0 %  |

data/features/rspec_basic.feature

@@ -1,31 +0,0 @@
-@rspec
-Feature:
-
-  Simply adding the basic simplecov lines to a project should get
-  the user a coverage report after running `rspec`
-
-  Scenario:
-    Given SimpleCov for RSpec is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start
-      """
-
-    When I open the coverage report generated with `bundle exec rspec spec`
-    Then I should see the groups:
-      | name      | coverage | files |
-      | All Files | 90.74%   | 6     |
-
-    And I should see the source files:
-      | name                                    | coverage |
-      | lib/faked_project.rb                    | 100.0 %  |
-      | lib/faked_project/some_class.rb         | 80.0 %  |
-      | lib/faked_project/framework_specific.rb | 75.0 %   |
-      | lib/faked_project/meta_magic.rb         | 100.0 %  |
-      | spec/meta_magic_spec.rb                 | 100.0 %  |
-      | spec/some_class_spec.rb                 | 100.0 %  |
-
-      # Note: faked_spec.rb is not appearing here since that's the first unit test file
-      # loaded by Rake, and only there test_helper is required, which then loads simplecov
-      # and triggers tracking of all other loaded files! Solution for this would be to
-      # configure simplecov in this first test instead of test_helper.

data/features/rspec_fails_on_initialization.feature

@@ -1,14 +0,0 @@
-@rspec
-Feature:
-
-  Running specs with a failing rspec setup
-
-  Scenario: Fail if rspec fails before starting its tests
-    Given a file named "spec/spec_helper.rb" with:
-      """
-      require 'simplecov'
-      SimpleCov.start
-      raise "some exception in the class loading before the tests start"
-      """
-    When I run `bundle exec rspec spec`
-    Then the exit status should not be 0

data/features/rspec_groups_and_filters_basic.feature

@@ -1,29 +0,0 @@
-@rspec
-Feature:
-
-  Defining some groups and filters should give a corresponding
-  coverage report that respects those settings after running rspec
-
-  Scenario:
-    Given SimpleCov for RSpec is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start do
-        add_group 'Libs', 'lib/faked_project/'
-        add_filter '/spec/'
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rspec spec`
-    And I should see the groups:
-      | name      | coverage | files |
-      | All Files | 88.1%    | 4     |
-      | Libs      | 86.11%   | 3     |
-      | Ungrouped | 100.0%   | 1     |
-
-    And I should see the source files:
-      | name                                    | coverage |
-      | lib/faked_project.rb                    | 100.0 %  |
-      | lib/faked_project/some_class.rb         | 80.0 %  |
-      | lib/faked_project/framework_specific.rb | 75.0 %   |
-      | lib/faked_project/meta_magic.rb         | 100.0 %  |

data/features/rspec_groups_and_filters_complex.feature

@@ -1,35 +0,0 @@
-@rspec
-Feature: Sophisticated grouping and filtering on RSpec
-
-  Defining groups and filters can be done by passing blocks or strings.
-  Blocks get each SimpleCov::SourceFile instance passed an can use arbitrary
-  and potentially weird conditions to remove files from the report or add them
-  to specific groups.
-
-  Scenario:
-    Given SimpleCov for RSpec is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start do
-        add_group 'By block' do |src_file|
-          src_file.filename =~ /MaGiC/i
-        end
-        add_group 'By string', 'project/meta_magic'
-
-        add_filter 'faked_project.rb'
-        # Remove all files that include "describe" in their source
-        add_filter {|src_file| src_file.lines.any? {|line| line.src =~ /describe/ } }
-        add_filter {|src_file| src_file.covered_percent < 100 }
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rspec spec`
-    Then I should see the groups:
-      | name      | coverage | files |
-      | All Files | 100.0%   | 1     |
-      | By block  | 100.0%   | 1     |
-      | By string | 100.0%   | 1     |
-
-    And I should see the source files:
-      | name                            | coverage |
-      | lib/faked_project/meta_magic.rb | 100.0 %  |

data/features/rspec_groups_using_filter_class.feature

@@ -1,40 +0,0 @@
-@rspec
-Feature: Grouping on RSpec using a custom filter class
-
-  Next to passing a block or a string to define a group, you can also pass
-  a filter class. The filter class inherits from SimpleCov::Filter and
-  must implement the matches? method, which is used to determine whether
-  or not a file should be added to the group.
-
-  Scenario:
-    Given SimpleCov for RSpec is configured with:
-      """
-      require 'simplecov'
-      class CoverageFilter < SimpleCov::Filter
-        def matches?(source_file)
-          source_file.covered_percent < filter_argument
-        end
-      end
-      SimpleCov.start do
-        add_group 'By filter class', CoverageFilter.new(90)
-        add_group 'By string', 'project/meta_magic'
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rspec spec`
-    Then I should see the groups:
-      | name             | coverage | files |
-      | All Files        | 90.74%   | 6     |
-      | By filter class  | 78.26%   | 2     |
-      | By string        | 100.0%   | 1     |
-      | Ungrouped        | 100.0%   | 3     |
-
-    And I should see the source files:
-      | name                                    | coverage |
-      | lib/faked_project/framework_specific.rb | 75.0 %   |
-      | lib/faked_project/some_class.rb         | 80.0 %   |
-      | lib/faked_project.rb                    | 100.0 %  |
-      | lib/faked_project/meta_magic.rb         | 100.0 %  |
-      | spec/meta_magic_spec.rb                 | 100.0 %  |
-      | spec/some_class_spec.rb                 | 100.0 %  |
-

data/features/rspec_without_simplecov.feature

@@ -1,20 +0,0 @@
-@rspec
-Feature:
-
-  Running specs without simplecov configuration
-
-  Scenario: No config at all
-    When I successfully run `bundle exec rspec spec`
-    Then no coverage report should have been generated
-
-  Scenario: Configured, but not started
-    Given SimpleCov for RSpec is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.configure do
-        add_filter 'somefilter'
-      end
-      """
-
-    When I successfully run `bundle exec rspec spec`
-    Then no coverage report should have been generated

data/features/skipping_code_blocks_manually.feature

@@ -1,70 +0,0 @@
-@test_unit @nocov
-Feature:
-
-  When code is wrapped in :nocov: comment blocks, it does not count
-  against the coverage numbers.
-
-  Background:
-    Given SimpleCov for Test/Unit is configured with:
-      """
-      require 'simplecov'
-      SimpleCov.start 'test_frameworks'
-      """
-
-  Scenario: Plain run with a nocov'd method
-    Given a file named "lib/faked_project/nocov.rb" with:
-      """
-      class SourceCodeWithNocov
-        #:nocov:
-        def some_weird_code
-          never_reached
-        rescue => err
-          but no one cares about invalid ruby here
-        end
-        #:nocov:
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-
-    Then I should see the source files:
-      | name                                    | coverage |
-      | lib/faked_project.rb                    | 100.0 %  |
-      | lib/faked_project/some_class.rb         | 80.0 %   |
-      | lib/faked_project/framework_specific.rb | 75.0 %   |
-      | lib/faked_project/meta_magic.rb         | 100.0 %  |
-      | lib/faked_project/nocov.rb              | 100.0 %  |
-
-    And there should be 5 skipped lines in the source files
-
-    And the report should be based upon:
-      | Unit Tests |
-
-  Scenario: Number of spaces should not mix up nocov results
-    Given a file named "lib/faked_project/nocov.rb" with:
-      """
-      class SourceCodeWithNocov
-           #    :nocov:
-        def some_weird_code
-          never_reached
-        rescue => err
-          but no one cares about invalid ruby here
-        end
-          #   :nocov:
-      end
-      """
-
-    When I open the coverage report generated with `bundle exec rake test`
-
-    Then I should see the source files:
-      | name                                    | coverage |
-      | lib/faked_project.rb                    | 100.0 %  |
-      | lib/faked_project/some_class.rb         | 80.0 %   |
-      | lib/faked_project/framework_specific.rb | 75.0 %   |
-      | lib/faked_project/meta_magic.rb         | 100.0 %  |
-      | lib/faked_project/nocov.rb              | 100.0 %  |
-
-    And there should be 5 skipped lines in the source files
-
-    And the report should be based upon:
-      | Unit Tests |

data/features/step_definitions/html_steps.rb

@@ -1,45 +0,0 @@
-module GroupHelpers
-  def available_groups
-    all('#content .file_list_container')
-  end
-
-  def available_source_files
-    all('.source_files .source_table')
-  end
-end
-World(GroupHelpers)
-
-
-Then /^I should see the groups:$/ do |table|
-  expected_groups = table.hashes
-  # Given group names should be the same number than those rendered in report
-  expected_groups.count.should == available_groups.count
-
-  # Verify each of the expected groups has a file list container and corresponding title and coverage number
-  # as well as the correct number of links to files.
-  expected_groups.each do |group|
-    with_scope "#content ##{group["name"].gsub(/[^a-z]/i, '')}.file_list_container" do
-      file_count_in_group = page.all('a.src_link').count
-      file_count_in_group.should == group["files"].to_i
-
-      with_scope "h2" do
-        page.should have_content(group["name"])
-        page.should have_content(group["coverage"])
-      end
-    end
-  end
-end
-
-Then /^I should see the source files:$/ do |table|
-  expected_files = table.hashes
-  expected_files.length.should == available_source_files.count
-
-  # Find all filenames and their coverage present in coverage report
-  files = available_source_files.map {|f| {"name" => f.find('h3').text, "coverage" => f.find('.header span').text} }
-
-  files.sort_by {|hsh| hsh["name"] }.should == expected_files.sort_by {|hsh| hsh["name"] }
-end
-
-Then /^there should be (\d+) skipped lines in the source files$/ do |expected_count|
-  all(".source_table ol li.skipped").count.should == expected_count.to_i
-end

data/features/step_definitions/simplecov_steps.rb

@@ -1,61 +0,0 @@
-# Just a shortcut to make framework setup more readable
-# The test project is using separate config files to avoid specifying all of
-# test/spec_helper in the features every time.
-Given /^SimpleCov for (.*) is configured with:$/ do |framework, config_body|
-  framework_dir = case framework
-    when /RSpec/i
-      "spec"
-    when /Test\/Unit/i
-      "test"
-    when /Cucumber/i
-      "features/support"
-    else
-      raise ArgumentError, "Could not identify test framework #{framework}!"
-  end
-
-  steps %Q{
-    Given a file named "#{framework_dir}/simplecov_config.rb" with:
-      """
-      #{config_body}
-      """
-  }
-end
-
-When /^I open the coverage report generated with `([^`]+)`$/ do |command|
-  steps %Q{
-    When I successfully run `#{command}`
-    Then a coverage report should have been generated
-    When I open the coverage report
-  }
-end
-
-Then /^a coverage report should have been generated(?: in "([^"]*)")?$/ do |coverage_dir|
-  coverage_dir ||= 'coverage'
-  steps %Q{
-    Then the output should contain "Coverage report generated"
-    And a directory named "#{coverage_dir}" should exist
-    And the following files should exist:
-      | #{coverage_dir}/index.html      |
-      | #{coverage_dir}/.resultset.json |
-  }
-end
-
-Then /^no coverage report should have been generated(?: in "([^"]*)")?$/ do |coverage_dir|
-  coverage_dir ||= 'coverage'
-  steps %Q{
-    Then the output should not contain "Coverage report generated"
-    And a directory named "#{coverage_dir}" should not exist
-    And the following files should not exist:
-      | #{coverage_dir}/index.html      |
-      | #{coverage_dir}/.resultset.json |
-  }
-end
-
-Then /^the report should be based upon:$/ do |table|
-  frameworks = table.raw.flatten
-  steps %Q{
-    Then the output should contain "Coverage report generated for #{frameworks.join(", ")}"
-    And I should see "using #{frameworks.join(", ")}" within "#footer"
-  }
-end
-

data/features/step_definitions/transformers.rb

@@ -1,13 +0,0 @@
-#
-# Enforce the alphabetical execution of specs because rspec 2+ executes them
-# randomly with `rspec spec` while we need them in an accurate order for coverage
-# reports that include the spec files.
-#
-# This is due to the fact that coverage will not include the first loaded spec/test file.
-# To get predictable coverage results, we need to know which one that is...
-#
-Transform "bundle exec rspec spec" do |t|
-  files = nil # Avoid shadowing
-  in_current_dir { files = Dir['spec/**/*_spec.rb'] }
-  "bundle exec rspec #{files.sort.join(' ')}"
-end