A minimalist RSpec clone with all the essentials.
- Maintain low code complexity to avoid false negatives and false positives.
- Implement the loading of specifications using simple, atomic, and thread-safe Ruby primitives.
- Avoid cluttering the interface with unnecessary alternative syntaxes.
- Provide the basics of the RSpec DSL for writing tests.
- Monkey-patching is not an available option.
- The framework does not use hacks such as the
at_exit
hook to trigger tests. - Malicious actual values cannot compromise results.
- If no
subject
is explicitly determined, it remains undefined. - If no described class is set,
described_class
is undefined rather thannil
. - Expectations cannot be added inside a
before
block. - Arbitrary helper methods are not accessible within examples.
- The
let
method defines a helper method rather than a memoized helper method. - The one-liner
is_expected
syntax is compatible with block expectations. - Definitions of
subject
,before
, andlet
must precede examples. - The
after
hook is unsupported. - The execution of the test suite halts immediately when an error is detected.
- Each
context
block isolates its tests and any potential side effects. - The
its
method is available without the need for external dependencies.
Add this line to your application's Gemfile:
gem "r_spec-clone"
And then execute:
bundle install
Or install it yourself as:
gem install r_spec-clone
RSpec clone provides a structure for writing executable examples of how your code should behave.
Inspired by RSpec, it includes a domain-specific language (DSL) that allows you to write examples in a way similar to plain english.
A basic spec looks something like this:
To utilize the RSpec
module and its DSL, include require "r_spec"
in your spec files.
Many projects organize these includes through a custom spec helper.
Concrete test cases are defined in it
blocks.
An optional (but recommended) descriptive string or module indicates the purpose of the test and a block contains the main logic of the test.
Test cases that have been defined or outlined but are not yet expected to work can be defined using pending
instead of it
.
They will not be run but show up in the spec report as pending.
An it
block contains an example that should invoke the code to be tested and define what is expected of it.
Each example can contain multiple expectations, but it should test only one specific behaviour.
The its
method can also be used to generate a nested example group with a single example that specifies the expected value (or the block expectations) of an attribute of the subject using is_expected
.
To express an expectation, wrap an object or block in expect
, call to
(or not_to
) and pass it a matcher object.
If the expectation is met, code execution continues.
Otherwise the example has failed and other code will not be executed.
In test files, specs can be structured by example groups which are defined by describe
and context
sections.
Typically a top level describe
defines the outer unit (such as a class) to be tested by the spec.
Further describe
sections can be nested within the outer unit to specify smaller units under test (such as individual methods).
For unit tests, it is recommended to follow the conventions for method names:
- outer
describe
is the name of the class, innerdescribe
targets methods; - instance methods are prefixed with
#
, class methods with.
.
To establish certain contexts — think empty array versus array with elements — the context
method may be used to communicate this to the reader.
Unlike a describe
block, all specifications executed within a context
are isolated in a subprocess.
This prevents possible side effects on the Ruby object environment from being propagated outside their context, which could alter the result of the unit test suite.
Note: if you are wondering what kind of code might be generated by the DSL, an article that shows the dynamic transcription of the main methods with simple examples is available in Chinese, in English and in Japanese.
Expectations define if the value being tested (actual) matches a certain value or specific criteria.
expect(actual).to eql(expected) # passes if expected.eql?(actual)
expect(actual).to eq(expected) # passes if expected.eql?(actual)
expect(actual).to equal(expected) # passes if expected.equal?(actual)
expect(actual).to be(expected) # passes if expected.equal?(actual)
expect(actual).to be_within(delta).of(expected) # passes if (expected - actual).abs <= delta
expect(actual).to match(expected) # passes if expected.match?(actual)
expect { actual }.to raise_exception(expected) # passes if expected exception is raised
expect(actual).to be_true # passes if true.equal?(actual)
expect(actual).to be_false # passes if false.equal?(actual)
expect(actual).to be_nil # passes if nil.equal?(actual)
expect(actual).to be_instance_of(expected) # passes if expected.equal?(actual.class)
expect(actual).to be_an_instance_of(expected) # passes if expected.equal?(actual.class)
expect(actual).to be_xxx # passes if actual.xxx?
expect(actual).to be_have_xxx(:yyy) # passes if actual.has_xxx?(:yyy)
expect([]).to be_empty
expect(foo: 1).to have_key(:foo)
expect { object.action }.to change(object, :value).to(new)
expect { object.action }.to change(object, :value).from(old).to(new)
expect { object.action }.to change(object, :value).by(delta)
expect { object.action }.to change(object, :value).by_at_least(minimum_delta)
expect { object.action }.to change(object, :value).by_at_most(maximum_delta)
expect(actual).to(satisfy { |value| value == expected })
By convention, specs live in the spec/
directory of a project. Spec files should end with _spec.rb
to be recognizable as such.
Depending of the project settings, you may run the specs of a project by running rake spec
(see Rake integration example section below).
A single file can also be executed directly with the Ruby interpreter.
Run all specs in files matching spec/**/*_spec.rb
:
bundle exec rake spec
Run a single file:
ruby spec/my/test/file_spec.rb
It is not recommended, but the RSpec's rspec
command line might also work:
rspec spec/my/test/file_spec.rb
rspec spec/my/test/file_spec.rb:42
rspec spec/my/test/
rspec
Many projects use a custom spec helper file, usually named spec/spec_helper.rb
.
This file is used to require r_spec/clone
and other includes, like the code from the project needed for every spec file.
The following Rakefile
settings should be enough:
require "bundler/gem_tasks"
require "rake/testtask"
Rake::TestTask.new do |t|
t.pattern = "spec/**/*_spec.rb"
end
task spec: :test
task default: :test
And then execute:
bundle exec rake
The benchmarks compare the performance of r_spec-clone
with the following frameworks (in alphabetical order):
Benchmark against 100 executions of a file containing 1 expectation (lower is better).
Benchmark against 1 execution of a file containing 100,000 expectations (lower is better).
RSpec clone's specifications are self-described here: spec/
- Home page: https://r-spec.dev/
- Cheatsheet: https://r-spec.dev/cheatsheet.html
- Blog post: https://cyrilllllll.medium.com/introducing-a-new-rspec-850d48c0f901
- Source code: https://github.com/cyril/r_spec-clone.rb
- API Doc: https://rubydoc.info/gems/r_spec-clone
- Twitter: https://twitter.com/cyri_
I would like to thank the whole RSpec team for all their work. It's a great framework and it's a pleasure to work with every day.
Without RSpec, this clone would not have been possible.
If you like this project, please consider making a small donation.
RSpec clone follows Semantic Versioning 2.0.
The gem is available as open source under the terms of the MIT License.
Under the hood, RSpec clone is largely animated by a collection of testing libraries designed to make programmers happy.
It's a living example of what we can do combining small libraries together that can boost the fun of programming.