Class: Minitest::BenchSpec

Relationships & Source Files
Super Chains via Extension / Inclusion / Inheritance
Class Chain: self, Spec::DSL, Benchmark, Test, Guard, Runnable
Instance Chain: self, Benchmark, Test, Guard, Test::LifecycleHooks, Assertions, Runnable
Inherits:	Minitest::Benchmark ::Object Minitest::Runnable Minitest::Test Minitest::Benchmark Minitest::BenchSpec
Defined in:	lib/minitest/benchmark.rb

Overview

The spec version of Benchmark.

Constant Summary

Class Attribute Summary

Spec::DSL - Extended

desc

Test - Inherited

.io_lock

Class Method Summary

.bench(name, &block)

This is used to define a new benchmark method.
.bench_performance_constant(name, threshold = 0.99, &work)

Create a benchmark that verifies that the performance is constant.
.bench_performance_exponential(name, threshold = 0.99, &work)

Create a benchmark that verifies that the performance is exponential.
.bench_performance_linear(name, threshold = 0.99, &work)

Create a benchmark that verifies that the performance is linear.
.bench_range(&block)

Specifies the ranges used for benchmarking for that class.

Spec::DSL - Extended

after	Define an 'after' action.
before	Define a 'before' action.
it	Define an expectation with name `desc`.
let	Essentially, define an accessor for `name` with `block`.
register_spec_type	Register a new type of spec that matches the spec's description.
spec_type	Figure out the spec class to use based on a spec's description.
specify	Alias for Spec::DSL#it.
subject	Another lazy man's accessor generator.
children, create, describe_stack, name, nuke_test_methods!, to_s

Benchmark - Inherited

.bench_exp	Returns a set of ranges stepped exponentially from `min` to `max` by powers of `base`.
.bench_linear	Returns a set of ranges stepped linearly from `min` to `max` by `step`.
.bench_range	Specifies the ranges used for benchmarking for that class.
.io, .run, .runnable_methods

Test - Inherited

.i_suck_and_my_tests_are_order_dependent!	Call this at the top of your tests when you absolutely positively need to have ordered tests.
.make_my_diffs_pretty!	Make diffs for this Test use `#pretty_inspect` so that diff in assert_equal can have more details.
.parallelize_me!	Call this at the top of your tests when you want to run your tests in parallel.
.runnable_methods	Returns all instance methods starting with “test_”.
.test_order	Defines the order to run tests (:random by default).

Guard - Extended

jruby?	Is this running on jruby?
maglev?	Is this running on maglev?
mri?	Is this running on mri?
rubinius?	Is this running on rubinius?
windows?	Is this running on windows?

Runnable - Inherited

.methods_matching	Returns all instance methods matching the pattern `re`.
.run	Responsible for running all runnable methods in a given class, each in its own instance.
.run_one_method	Runs a single method and has the reporter record the result.
.runnable_methods	Each subclass of Runnable is responsible for overriding this method to return all runnable methods.
.runnables	Returns all subclasses of Runnable.
.inherited, .new, .on_signal, .reset, .with_info_handler

Instance Attribute Summary

Test - Inherited

#error?	Did this run error?
#passed?	Did this run pass?
#skipped?	Was this run skipped?
#time	The time it took to run this test.

Assertions - Included

#skipped?

Was this testcase skipped? Meant for #teardown.

Runnable - Inherited

#assertions	Number of assertions executed in this run.
#failures	An assertion raised during the run, if any.
#name	Name of the run.
#name=	Set the name of the run.
#passed?	Did this run pass?
#skipped?	Was this run skipped? See `#passed?` for more information.

Instance Method Summary

Benchmark - Inherited

#assert_performance	Runs the given `work`, gathering the times of each run.
#assert_performance_constant	Runs the given `work` and asserts that the times gathered fit to match a constant rate (eg, linear slope == 0) within a given `threshold`.
#assert_performance_exponential	Runs the given `work` and asserts that the times gathered fit to match a exponential curve within a given error `threshold`.
#assert_performance_linear	Runs the given `work` and asserts that the times gathered fit to match a straight line within a given error `threshold`.
#assert_performance_logarithmic	Runs the given `work` and asserts that the times gathered fit to match a logarithmic curve within a given error `threshold`.
#assert_performance_power	Runs the given `work` and asserts that the times gathered curve fit to match a power curve within a given error `threshold`.
#fit_error	Takes an array of x/y pairs and calculates the general R^2 value.
#fit_exponential	To fit a functional form: y = ae^(bx).
#fit_linear	Fits the functional form: a + bx.
#fit_logarithmic	To fit a functional form: y = a + b*ln(x).
#fit_power	To fit a functional form: y = ax^b.
#sigma	Enumerates over `enum` mapping `block` if given, returning the sum of the result.
#validation_for_fit	Returns a proc that calls the specified fit method and asserts that the error is within a tolerable threshold.
#io

Test - Inherited

#location	The location identifier of this test.
#result_code	Returns “.”, “F”, or “E” based on the result of the run.
#run	Runs a single test with setup/teardown hooks.
#capture_exceptions	LifecycleHooks.
#marshal_dump, #marshal_load, #time_it, #to_s, #with_info_handler

Guard - Included

#jruby?	Is this running on jruby?
#maglev?	Is this running on maglev?
#mri?	Is this running on mri?
#rubinius?	Is this running on rubinius?
#windows?	Is this running on windows?

Test::LifecycleHooks - Included

#after_setup	Runs before every test, after setup.
#after_teardown	Runs after every test, after teardown.
#before_setup	Runs before every test, before setup.
#before_teardown	Runs after every test, before teardown.
#setup	Runs before every test.
#teardown	Runs after every test.

Assertions - Included

#assert	Fails unless `test` is truthy.
#assert_empty	Fails unless `obj` is empty.
#assert_equal	Fails unless `exp == act` printing the difference between the two, if possible.
#assert_in_delta	For comparing Floats.
#assert_in_epsilon	For comparing Floats.
#assert_includes	Fails unless `collection` includes `obj`.
#assert_instance_of	Fails unless `obj` is an instance of `cls`.
#assert_kind_of	Fails unless `obj` is a kind of `cls`.
#assert_match	Fails unless `matcher` `=~` `obj`.
#assert_mock	Assert that the mock verifies correctly.
#assert_nil	Fails unless `obj` is nil.
#assert_operator	For testing with binary operators.
#assert_output	Fails if stdout or stderr do not output the expected results.
#assert_predicate	For testing with predicates.
#assert_raises	Fails unless the block raises one of `exp`.
#assert_respond_to	Fails unless `obj` responds to `meth`.
#assert_same	Fails unless `exp` and `act` are `#equal?`
#assert_send	`send_ary` is a receiver, message and arguments.
#assert_silent	Fails if the block outputs anything to stderr or stdout.
#assert_throws	Fails unless the block throws `sym`
#capture_io	Captures $stdout and $stderr into strings:
#capture_subprocess_io	Captures $stdout and $stderr into strings, using Tempfile to ensure that subprocess IO is captured as well.
#diff	Returns a diff between `exp` and `act`.
#exception_details	Returns details for exception `e`
#flunk	Fails with `msg`
#message	Returns a proc that will output `msg` along with the default message.
#mu_pp	This returns a human-readable version of `obj`.
#mu_pp_for_diff	This returns a diff-able human-readable version of `obj`.
#pass	used for counting assertions.
#refute	Fails if `test` is truthy.
#refute_empty	Fails if `obj` is empty.
#refute_equal	Fails if `exp == act`.
#refute_in_delta	For comparing Floats.
#refute_in_epsilon	For comparing Floats.
#refute_includes	Fails if `collection` includes `obj`.
#refute_instance_of	Fails if `obj` is an instance of `cls`.
#refute_kind_of	Fails if `obj` is a kind of `cls`.
#refute_match	Fails if `matcher` `=~` `obj`.
#refute_nil	Fails if `obj` is nil.
#refute_operator	Fails if `o1` is not `op` `o2`.
#refute_predicate	For testing with predicates.
#refute_respond_to	Fails if `obj` responds to the message `meth`.
#refute_same	Fails if `exp` is the same (by object identity) as `act`.
#skip	Skips the current run.
#_synchronize

Runnable - Inherited

#result_code	Returns a single character string to print based on the result of the run.
#run	Runs a single method.
#failure, #marshal_dump, #marshal_load

Constructor Details

This class inherits a constructor from Minitest::Runnable

Class Method Details

.bench(name, &block)

This is used to define a new benchmark method. You usually don't use this directly and is intended for those needing to write new performance curve fits (eg: you need a specific polynomial fit).

See .bench_performance_linear for an example of how to use this.

[ GitHub ]

# File 'lib/minitest/benchmark.rb', line 357


def self.bench name, &block
  define_method "bench_#{name.gsub(/\W+/, "_")}", &block
end

.bench_performance_constant(name, threshold = 0.99, &work)

Create a benchmark that verifies that the performance is constant.

describe "my class Bench" do
  bench_performance_constant "zoom_algorithm!" do |n|
    @obj.zoom_algorithm!(n)
  end
end

[ GitHub ]

# File 'lib/minitest/benchmark.rb', line 401


def self.bench_performance_constant name, threshold = 0.99, &work
  bench name do
    assert_performance_constant threshold, &work
  end
end

.bench_performance_exponential(name, threshold = 0.99, &work)

Create a benchmark that verifies that the performance is exponential.

describe "my class Bench" do
  bench_performance_exponential "algorithm" do |n|
    @obj.algorithm(n)
  end
end

[ GitHub ]

# File 'lib/minitest/benchmark.rb', line 416


def self.bench_performance_exponential name, threshold = 0.99, &work
  bench name do
    assert_performance_exponential threshold, &work
  end
end

.bench_performance_linear(name, threshold = 0.99, &work)

Create a benchmark that verifies that the performance is linear.

describe "my class Bench" do
  bench_performance_linear "fast_algorithm", 0.9999 do |n|
    @obj.fast_algorithm(n)
  end
end

[ GitHub ]

# File 'lib/minitest/benchmark.rb', line 386


def self.bench_performance_linear name, threshold = 0.99, &work
  bench name do
    assert_performance_linear threshold, &work
  end
end

.bench_range(&block)

Specifies the ranges used for benchmarking for that class.

bench_range do
  bench_exp(2, 16, 2)
end

See Minitest::Benchmark#bench_range for more details.

[ GitHub ]

# File 'lib/minitest/benchmark.rb', line 370


def self.bench_range &block
  return super unless block

  meta = (class << self; self; end)
  meta.send :define_method, "bench_range", &block
end