Pull request Compare This branch is 319 commits behind sunaku:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Failed to load latest commit information.
bin
lib/tork
.gitignore
Gemfile
HISTORY.markdown
LICENSE
README.markdown
Rakefile
tork.gemspec

README.markdown

_______      _______
 ___  /___________ /__
  _  __/ __ \  __/ /_/
  / /_/ /_/ / / / ,\
  \__/\____/_/ /_/|_\
             >>>------>

Test with fork

Tork is a continuous testing tool for Ruby that automatically detects and tests changes in your Ruby application or test suite in an efficient manner:

  1. Absorbs your test execution overhead into a master process.

  2. Forks to run your test files in parallel; overhead inherited.

  3. Avoids running unchanged tests inside changed test files.


Features

  • No configuration needed: run tork for Ruby, tork rails for Rails.

  • Runs test files in parallel using fork for multi-core/CPU utilization.

  • Tests changes your Ruby application for rapid TDD: avoids running (1) unchanged test files and (2) unchanged tests inside changed test files.

  • Supports MiniTest, Test::Unit, RSpec, and any testing framework that (1) exits with a nonzero status to indicate test failures (2) is loaded by your application's test/test_helper.rb or spec/spec_helper.rb file.

  • Logs the output from your tests into separate files: one log per test. The path of a log file is simply the path of its test file plus ".log".

  • Configurable through a Ruby script in your current working directory.

  • You can override the modular tork* programs with your own in $PATH.

  • Its core is written in less than 360 lines (SLOC) of pure Ruby code! :-)

Architecture

Following UNIX philosophy, Tork is composed of simple text-based programs: so you can build your own custom Tork user interface by wrapping tork-driver!

  • tork is an interactive command-line user interface (CLI) for driver
  • tork-herald monitors current directory tree and reports changed files
  • tork-driver tells master to run tests and keeps track of test results
  • tork-master absorbs test execution overhead and forks to run your tests

When the herald observes that files in or beneath the current directory have been written to, it tells the driver, which then commands the master to fork a worker process to run the tests affected by those changed files. This is all performed automatically. But what if you want to manually run a test file?

You can (re)run any test file by simply saving it! When you do, Tork tries to figure out which tests inside your newly saved test file have changed (using diff and regexps) and then attempts to run just those. To make it run all tests in your saved file, simply save the file again without changing it.


Installation

gem install tork

Prerequisites

  • Ruby 1.8.7 or 1.9.2 or newer.

  • Operating system that supports POSIX signals and the fork() system call. To check if your system qualifies, launch irb and enter the following:

    Process.respond_to? :fork  # must be true
    Signal.list.key? 'CHLD'    # must be true
    Signal.list.key? 'TERM'    # must be true
    
  • To make the tork-herald program's filesystem monitoring more efficient:

    gem install rb-inotify  # linux
    gem install rb-fsevent  # macosx
    

Development

git clone git://github.com/sunaku/tork
cd tork
bundle install --binstubs=bundle_bin
bundle_bin/tork --help  # run it directly
bundle exec rake -T     # packaging tasks

Usage

At the command line

tork --help

You can monitor your test processes from another terminal:

watch 'ps xuw | sed -n "1p; /tor[k]/p" | fgrep -v sed'

With Ruby on Rails

For Rails 3 or newer, use the rails configuration helper (see below). Otherwise, ensure that your config/environments/test.rb file contains:

config.cache_classes = false

To use SQLite3 as your test database, install its in-memory database adapter. Otherwise, you might face these errors:

SQLite3::BusyException: database is locked

cannot start a transaction within a transaction

With factory_girl

Do not load your factories into the master process as part of your test execution overhead in your test/spec helper because that would necessitate overhead reabsorption whenever you change or create factory definitions.

Instead, use at_exit() to wait until (1) after the master process has forked a worker process and (2) just before that worker process runs its test suite (whose execution is started by your test framework's own at_exit() handler):

# in your test/spec helper
require 'factory_girl'
at_exit { FactoryGirl.find_definitions unless $! }

This way, worker processes will pick up changes in your factories "for free" whenever they (re)run your test files. Skip if Ruby is exiting because of a raised exception (denoted by the $! global variable in the snippet above).

As a bonus, this arrangement also works when tests are run outside of Tork!


Configuration

Tork looks for a configuration file named .tork.rb in its current working directory. The configuration file is a normal Ruby script, inside which you can query and modify the Tork::Config object: an instance of OpenStruct.


Configuration helpers

Ruby on Rails

At the command line:

tork rails

Or in your configuration file:

require 'tork/config/rails'

Cucumber

At the command line:

tork cucumber

Or in your configuration file:

require 'tork/config/cucumber'

parallel_tests

At the command line:

tork parallel_tests

Or in your configuration file:

require 'tork/config/parallel_tests'

Configuration options

Tork::Config.max_forked_workers

Maximum number of worker processes at any given time. The default value is the number of processors detected on your system, or 1 if detection fails.

Tork::Config.overhead_load_paths

Array of paths that are prepended to Ruby's $LOAD_PATH before the test execution overhead is loaded into tork-master.

Tork::Config.overhead_file_globs

Array of file globbing patterns that describe a set of Ruby scripts that are loaded into tork-master as test execution overhead.

Tork::Config.reabsorb_file_greps

Array of regular expressions that describe a set of file paths that cause the test execution overhead to be reabsorbed in tork-master when they change.

Tork::Config.all_test_file_globs

Array of file globbing patterns that describe the set of all test files in your Ruby application.

Tork::Config.test_file_globbers

Hash that maps (1) a regular expression describing a set of file paths to (2) a lambda function that accepts the path to a changed file and a MatchData object containing the results of the regular expression matching, and yields a file globbing pattern that describes a set of test files that need to be run.

In other words, whenever the source files (the regular expression) change, their associated test files (result of calling the lambda function) are run.

For example, if test files had the same names as their source files followed by an underscore and the file name in reverse like this:

  • lib/hello.rb => test/hello_olleh.rb
  • app/world.rb => spec/world_ldrow.rb

Then you would add the following to your configuration file:

Tork::Config.test_file_globbers[%r<^(lib|app)/.+\.rb$>] = lambda do |path, matches|
  name = File.basename(path, '.rb')
  "{test,spec}/**/#{name}_#{name.reverse}.rb"
end

In addition, these lambda functions can return nil if they do not wish for a particular source file to be tested. For example, to ignore tests for all source files except those within a models/ directory, you would write:

Tork::Config.test_file_globbers[%r<^(lib|app)/.+\.rb$>] = lambda do |path, matches|
  if path.include? '/models/'
    "{test,spec}/**/#{File.basename(path)}"
  end
end

Tork::Config.before_fork_hooks

Array of lambda functions that are executed inside tork-master before a worker process is forked to run a test file. These functions are given:

  1. The sequence number of the worker process that will be forked shortly.

  2. The path of the log file containing the live output of the worker process.

  3. The path of the test file that will be run by the worker process.

  4. An array of line numbers in the test file to run. If this array is empty, then the entire test file will be run.

For example, to see some real values:

Tork::Config.before_fork_hooks << lambda do |worker_number, log_file, test_file, line_numbers|
  p :before_fork_hooks => {
    :worker_number => worker_number,
    :log_file      => log_file,
    :test_file     => test_file,
    :line_numbers  => line_numbers,
  }
end

Tork::Config.after_fork_hooks

Array of lambda functions that are executed inside a worker process forked by tork-master. These functions are given:

  1. The sequence number of the worker process.

  2. The path of the log file containing the live output of the worker process.

  3. The path of the test file that will be run by the worker process.

  4. An array of line numbers in the test file to run. If this array is empty, then the entire test file will be run.

For example, to see some real values, including the worker process' PID:

Tork::Config.after_fork_hooks << lambda do |worker_number, log_file, test_file, line_numbers|
  p :after_fork_hooks => {
    :worker_pid    => $$,
    :worker_number => worker_number,
    :log_file      => log_file,
    :test_file     => test_file,
    :line_numbers  => line_numbers,
  }
end

The first function in this array instructs Test::Unit and RSpec to only run those tests that are defined on the given line numbers. This accelerates your test-driven development cycle by only running tests you are currently editing.


License

Released under the ISC license. See the LICENSE file for details.