builtbybuffalo / spec2.cr

Enhanced `spec` testing library for [Crystal](http://crystal-lang.org/).

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

spec2 Build Status

Enhanced spec testing library for Crystal.

Example

Spec2.describe Greeting do
  subject { Greeting.new }

  describe "#greet" do
    context "when name is world" do
      let(name) { "world" }

      it "greets the world" do
        expect(subject.greet(name)).to eq("Hello, world")
      end
    end
  end
end

Installation

Add it to shard.yml

dependencies:
  spec2:
    github: waterlink/spec2.cr
    version: ~> 0.9

Goals

  • No global scope pollution
  • No Object pollution
  • Ability to run examples in random order
  • Ability to specify before and after blocks for example group
  • Ability to define let, let!, subject and subject! for example group

Roadmap

1.0

  • Configuration through CLI interface.
  • Filters.
  • Shared examples and example groups.

Usage

require "spec2"

Top-level describe

Spec2.describe MySuperLibrary do
  describe Greeting do
    # .. example groups and examples here ..
  end
end

If you have test suite written for Spec and you don't want to prefix each top-level describe with Spec2., you can just include Spec::GlobalDSL globally:

include Spec2::GlobalDSL

# and then:
describe Greeting do
  # ...
end

Writing examples

Spec2.describe "some tests" do
  it "is a test name here" do
    # .. this is the example here ..
  end

  pending "is a pending test here" do
    # .. this example will not be executed ..
  end
end

Expect syntax

expect(greeting.for("john")).to eq("hello, john")

If you have big codebase that runs on Spec, you can use this to enable #should and #should_not on Object:

Spec2.enable_should_on_object

List of builtin matchers

  • eq("hello, world") - asserts actual is equal to expected
  • raise_error(ErrorClass [, message_matcher]) - checks if block raises expected error
  • be(42) - asserts actual is the same as expected
  • match(/hello .+/) - asserts actual is matching provided regexp
  • be_true - asserts actual is equal true
  • be_false - asserts actual is equal false
  • be_truthy - asserts actual is not nil or false
  • be_falsey - asserts actual is nil or false
  • be_nil - asserts actual is equal nil
  • be_close(42, 0.01) - asserts actual is in delta-proximity of expected
  • expect(42).to_be < 45 - asserts arbitrary method call on actual to be truthy
  • be_a(String) - asserts actual to be of expected type (uses is_a?)

Random order

Spec2.random_order

# this is what happens under the hood
Spec2.configure_order(Spec2::Orders::Random)

To configure your own custom order you can use:

Spec2.configure_order(MyOrder)

Class MyOrder should implement Order protocol and Order::Factory class protocol (see it here). See also a random order implementation.

No color mode

Spec2.nocolor

# this is what happens under the hood
Spec2.configure_output(Spec2::Outputs::Nocolor)

To configure your own custom output you can use:

Spec2.configure_output(MyOutput)

Class MyOutput should implement Output protocol and Output::Factory class protocol (see it here). See also a default colorful output implementation.

Documentation reporter

Spec2.doc

# this is what happens under the hood
Spec2.configure_reporter(Spec2::Reporters::Doc)

To configure your own custom reporter you can use:

Spec2.configure_reporter(MyReporter)

Class MyReporter should implement Reporter protocol and Reporter::Factory class protocol (see it here). See also a default reporter implementation.

If you are creating a custom reporter, you might want to use ElapsedTime class to report elapsed time for the test suite. Example usage:

output.puts "Finished in #{::Spec2::ElapsedTime.new.to_s}"

Configuring custom Runner

Spec2.configure_runner(MyRunner)

Class MyRunner should implement Runner protocol and Runner::Factory class protocol (see it here). See also a default runner implementation.

before

before - register a hook that is run before any example in current and all nested contexts.

before { .. do some stuff .. }

after

after - register a hook that is run after any successful example in current and all nested contexts.

after { .. do some stuff .. }

let

let(name) { value } - register a binding of certain value to name. Lazy: provided block will only be evaluated when needed in example and only once per example.

let(answer) { 42 }

it "is correct answer" do
  expect(answer).to eq(42)
end

let!

let(name) { value } - register a binding of certain value to name. It is not lazy: provided block will be evaluated before each example exactly once.

let!(answer) { 42 }

it "is correct answer" do
  expect(answer).to eq(42)
end

described_class

For describe ... blocks, that describe a class, there is a shortcut to reference that class:

describe Example do
  it "can be created" do
    expect(described_class.new.greet).to eq("hello world")
    # instead of `Example.new.greet`.
  end
end

subject

subject { value } - register a subject of your test with provided value. Lazy.

subject { Stuff.new }

it "works" do
  expect(subject.answer).to eq(42)
end

subject(name) { value } - registers a named subject of your test with provided value with provided name. Lazy.

subject(stuff) { Stuff.new }

it "works" do
  expect(stuff.answer).to eq(42)
end

subject!

subject! { value } - register a subject of your test with provided value. It is not lazy.

subject! { Stuff.new }

it "works" do
  expect(subject.answer).to eq(42)
end

subject!(name) { value } - registers a named subject of your test with provided value with provided name. It is not lazy.

subject!(stuff) { Stuff.new }

it "works" do
  expect(stuff.answer).to eq(42)
end

delayed

Use delayed { ... } to verify expectations after test example and its after hooks finish. Example:

it "does something interesting eventually" do
  delayed { expect(value).to eq(42) }
  # .. do something else, that should eventually lead to value == 42 ..
end

Custom matchers

First, define your matcher implementing this protocol:

class MyMatcher(T, E)
  include Spec2::Matcher

  @actual_inspect : String?

  def initialize(@expected : T, @stuff : E)
  end

  def match(actual)
    @actual_inspect = actual.inspect

    # return true or false here
  end

  def failure_message
    "Expected to be valid #{@stuff.inspect}.
    Expected: #{@expected.inspect}.
    Actual:   #{@actual_inspect}."
  end

  def failure_message_when_negated
    "Expected to be invalid #{@stuff.inspect}.
    Expected: #{@expected.inspect}.
    Actual:   #{@actual_inspect}."
  end

  def description
    "(stuff in #{@expected} #{stuff})"
  end
end

And then, register shortcut helper method to use your matcher.

Spec2.register_matcher(stuff) do |stuff, expected|
  MyMatcher.new(expected, stuff)
end

And use it:

describe "stuff" do
  it "is valid stuff" do
    expect(something).to stuff(some_stuff, "expected stuff")
  end
end

Development

After you forked the repo:

  • run crystal deps to install dependencies
  • run crystal spec and crystal unit to see if tests are green (or just run scripts/test to run them both)
  • apply TDD to implement your feature/fix/etc

Contributing

  1. Fork it ( https://github.com/waterlink/spec2.cr/fork )
  2. Create your feature branch (git checkout -b my-new-feature)
  3. Commit your changes (git commit -am 'Add some feature')
  4. Push to the branch (git push origin my-new-feature)
  5. Create a new Pull Request

Contributors

  • waterlink Oleksii Fedorov - creator, maintainer

About

Enhanced `spec` testing library for [Crystal](http://crystal-lang.org/).

License:MIT License


Languages

Language:Crystal 99.8%Language:Ruby 0.1%Language:Shell 0.1%