Imperative Random Data Generator and Quickcheck

You can use Rantly to generate random test data, and use its Test::Unit extension for property-based testing.

Rantly is basically a recursive descent interpreter, each of its method returns a random value of some type (string, integer, float, etc.).

Its implementation has no alien mathematics inside. Completely side-effect-free-free.

Install

Rantly requires Ruby 3.2 or higher. To install Rantly add it to your Gemfile or run:

$ gem install rantly

You can try it in the console by running:

$ irb -rrantly
> Rantly { [integer,float] } # same as Rantly.value { integer }
=> [20991307, 0.025756845811823]
> Rantly { [integer,float]}
=> [-376856492, 0.452245765751706]
> Rantly(5) { integer } # same as Rantly.map(5) { integer }
=> [-1843396915550491870, -1683855015308353854, -2291347782549033959, -951461511269053584, 483265231542292652]

Data Generation

Getting Random Data Values

Rantly#map(n,limit=10,&block)
  call the generator n times, and collect values
Rantly#each(n,limit=10,&block)
  call a random block n times
Rantly#value(limit=10,&block)
  call a random block once, and get its value.

To collect an array of random data,

# we want 5 random integers
> Rantly(5) { integer }
=> [-380638946, -29645239, 344840868, 308052180, -154360970]

To iterate over random data,

> Rantly.each(5) { puts integer }
296971291
504994512
-402790444
113152364
502842783
=> nil

To get one value of random data,

> Rantly { integer }
=> 278101042

The optional argument limit is used with generator guard. By default, if you want to generate n items, the generator tries at most n * 10 times.

This almost always succeeds,

> Rantly(5) { i = integer; guard i > 0; i }
=> [511765059, 250554234, 305947804, 127809156, 285960387]

This always fails,

> Rantly(10) { guard integer.is_a?(Float) }
Rantly::TooManyTries: Exceed gen limit 100: 101 failed guards)

Random Generating Methods

The API is similiar to QuickCheck, but not exactly the same. In particular choose picks a random element from an array, and range picks a integer from an interval.

Simple Randomness

Rantly#integer(n=nil)
  random positive or negative integer. Fixnum only.
Rantly#range(lo,hi)
  random integer between lo and hi.
Rantly#float
  random float
Rantly#boolean
  true or false
Rantly#literal(value)
  No-op. returns value.
Rantly#choose(*vals)
  Pick one value from among vals.

Meta Randomness

A rant generator is just a mini interpreter. It's often useful to go meta,

Rantly#call(gen)
  If gen is a Symbol, just do a method call with send.
  If gen is an Array, the first element of the array is the method name, the rest are args.
  If gen is a Proc, instance_eval it with the generator.

> Rantly { call(:integer) }
=> -240998958

> Rantly { call([:range,0,10]) }
=> 2

> Rantly { call(Proc.new { [integer] })}
=> [522807620]

The call method is useful to implement other abstractions (See next subsection).

Rantly#branch(*args)
  Pick a random arg among args, and Rantly#call it.

50-50 chance getting an integer or float,

> Rantly { branch :integer, :float }
=> 0.0489446702931332
> Rantly { branch :integer, :float }
=> 494934533

Frequencies

Rantly#freq(*pairs)
  Takes a list of 2-tuples, the first of which is the weight, and the second a Rantly#callable value, and returns a random value picked from the pairs. Follows the distribution pattern specified by the weights.

Twice as likely to get a float than integer. Never gets a ranged integer.

> Rantly { freq [1,:integer], [2,:float], [0,:range,0,10] }

If the "pair" is not an array, but just a symbol, freq assumes that the weight is 1.

# 50-50 between integer and float
> Rantly { freq :integer, :float }

If a "pair" is an Array, but the first element is not an Integer, freq assumes that it's a Rantly method-call with arguments, and the weight is one.

# 50-50 chance generating integer limited by 10, or by 20.
> Rantly { freq [:integer,10], [:integer 20] }

Sized Structure

A Rantly generator keeps track of how large a datastructure it should generate with its size attribute.

Rantly#size
 returns the current size
Rantly#sized(n,&block)
 sets the size for the duration of recursive call of block. Block is instance_eval with the generator.

Rantly provides two methods that depends on the size

Rantly#array(size=default_size,&block)
  returns a sized array consisted of elements by Rantly#calling random branches.
Rantly#string(char_class=:print)
  returns a sized random string, consisted of only chars from a char_class.
Rantly#dict(size=default_size,&block)
  returns a sized random hash. The generator block should generate tuples of keys and values (arrays that have two elements, the first one is used as key, and the second as value).

The avaiable char classes for strings are:

:alnum
:alpha
:blank
:cntrl
:digit
:graph
:lower
:print
:punct
:space
:upper
:xdigit
:ascii

# sized 10 array of integers
> Rantly { array(10) { integer }}
=> [417733046, -375385433, 0.967812380000118, 26478621, 0.888588160450082, 250944144, 305584916, -151858342, 0.308123867823313, 0.316824642414253]

If you set the size once, it applies to all subsequent recursive structures. Here's a sized 10 array of sized 10 strings,

> Rantly { sized(10) { array {string}} }
=> ["1c}C/,9I#}", "hpA/UWPJ\\j", "H'~ERtI`|]", "%OUaW\\%uQZ", "Z2QdY=G~G!", "H<o|<FARGQ", "g>ojnxGDT3", "]a:L[B>bhb", "_Kl=&{tH^<", "ly]Yfb?`6c"]

Or a sized 10 array of sized 5 strings,

> Rantly {array(10){sized(5) {string}}}
=> ["S\"jf ", "d\\F-$", "-_8pa", "IN0iF", "SxRV$", ".{kQ7", "6>;fo", "}.D8)", "P(tS'", "y0v/v"]

Generate a hash that has 5 elements,

> Rantly { dict { [string,integer] }}
{"bR\\qHn"=>247003509502595457,
 "-Mp '."=>653206579583741142,
 "gY%<SV"=>-888111605212388599,
 "+SMn:r"=>-1159506450084197716,
 "^3gYfQ"=>-2154064981943219558,
 "= :/\\,"=>433790301059833691}

The dict generator retries if a key is duplicated. If it fails to generate a unique key after too many tries, it gives up by raising an error:

> Rantly { dict { ["a",integer] }}
Rantly::TooManyTries: Exceed gen limit 60: 60 failed guards)

Property Testing

Rantly extends Test::Unit and MiniTest::Test (5.0)/MiniTest::Unit::TestCase (< 5.0) for property testing. The extensions are in their own modules. So you need to require them explicitly:

require 'rantly/testunit_extensions' # for 'test/unit'
require 'rantly/minitest_extensions' # for 'minitest'
require 'rantly/rspec_extensions'    # for RSpec

They define:

Test::Unit::Assertions#property_of(&block)
  The block is used to generate random data with a generator. The method returns a Rantly::Property instance, that has the method 'check'.

Property assertions within Test::Unit could be done like this,

# checks that integer only generates fixnum.
property_of {
  integer
}.check { |i|
  assert(i.is_a?(Integer), "integer property did not return Integer type")
}

Property assertions within Minitest could be done like this,

# checks that integer only generates fixnum.
property_of {
  integer
}.check { |i|
  assert_kind_of Integer, i, "integer property did not return Integer type"
}

Property assertions within RSpec could be done like this,

# checks that integer only generates fixnum.
it "integer property only returns Integer type" do
   property_of {
     integer
   }.check { |i|
     expect(i).to be_a(Integer)
   }
end

The check block takes the generated data as its argument. One idiom I find useful is to include a parameter of the random data for the check argument. For example, if I want to check that Rantly#array generates the right sized array, I could say,

property_of {
  len = integer
  [len,array(len){integer}]
}.check { |(len,arr)|
  assert_equal len, arr.length
}

To control the number of property tests to generate, you have three options. In order of precedence:

1. Pass an integer argument to check

property_of {
  integer
}.check(9000) { |i|
  assert_kind_of Integer, i
}

2. Set the RANTLY_COUNT environment variable

RANTLY_COUNT=9000 ruby my_property_test.rb

3. If neither of the above are set, the default will be to run the check block 100 times.

If you wish to have quiet output from Rantly, set environmental variable:

RANTLY_VERBOSE=0 # silent
RANTLY_VERBOSE=1 # verbose and default if env is not set

This will silence the puts, print, and pretty_print statements in property.rb.

Shrinking

Shrinking reduces the value of common types to some terminal lower bound. These functions are added to the Ruby types Integer, String, Array, and Hash.

For example a String is shrinkable until it is empty (e.g. ""),

"foo".shrinkable?     # => true
"foo".shrink          # => "fo"
"fo".shrink           # => "f"
"f".shrink            # => ""
"".shrinkable?        # => false

Shrinking allows Property#check to find a reduced value that still fails the condition. The value is not truely minimal because:

• we do not perform a complete in-depth traversal of the failure tree
• we limit the search to a maximum 1024 shrinking operations

but is usually reduced enough to start debugging.

Enable shrinking with

require 'rantly/shrinks'

Use Tuple class if you want an array whose elements are individually shrinked, but are not removed. Example:

property_of {
  len = range(0, 10)
  Tuple.new( array(len) { integer } )
}.check {
  # .. property check here ..
}

Use Deflating class if you want an array whose elements are individully shrinked whenever possible, and removed otherwise. Example:

property_of {
  len = range(0, 10)
  Deflating.new( array(len) { integer } )
}.check {
  # .. property check here ..
}

Normal arrays or hashes are not shrinked.

Contributors

Thanks to all contributors. 💘 New contributors are welcome! 😉

Logotype designed by: @Richardbmx

License

Code published under MIT License, Copyright (c) 2009 Howard Yeh. See LICENSE.