https://github.com/schwern/test-random

Make testing random things a bit more deterministic
https://github.com/schwern/test-random

Last synced: about 1 month ago
JSON representation

Make testing random things a bit more deterministic

• Host: GitHub
• URL: https://github.com/schwern/test-random
• Owner: schwern
• License: other
• Created: 2010-01-20T01:12:31.000Z (about 15 years ago)
• Default Branch: master
• Last Pushed: 2013-04-28T02:53:23.000Z (over 11 years ago)
• Last Synced: 2024-10-29T20:08:28.345Z (3 months ago)
• Language: Perl
• Homepage: http://search.cpan.org/dist/Test-Random
• Size: 121 KB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README
  • Changelog: Changes
  • License: LICENSE

Awesome Lists containing this project

README

        
NAME

    Test::Random - Make testing random functions deterministic

SYNOPSIS

        use Test::Random;

        ... test as normal ...

DESCRIPTION

    This is a testing module to make testing random things a bit more

    deterministic.

  Controlling randomness

    Its main function is to allow you to repeat a failing test in the same

    way it ran before, even if it contained random elements. Test::Random

    will output the seed used by the random number generator. You can then

    use this seed to repeat the last test with exactly the same random

    elements.

    You can control the random seed used by Test::Random by setting the

    "TEST_RANDOM_SEED" environment variable. This is handy to make test runs

    repeatable.

        TEST_RANDOM_SEED=12345 perl -Ilib t/some_test.t

    Test::Random will output the seed used at the end of each test run. If

    the test failed it will be visible to the user (ie. on STDERR) otherwise

    it will be a TAP comment and only visible if the test is run verbosely.

    If having new data every run is too chaotic for you, you can set

    TEST_RANDOM_SEED to something which will remain fixed during a

    development session. Perhaps the PID of your shell or your uid or the

    date (20090704, for example).

EXAMPLE

    When you run a test with Test::Random you will see something like this:

        perl some_test.t

        1..3

        ok 1

        ok 2

        ok 3

        # TEST_RANDOM_SEED=20891494266

    If you wish to repeat the circumstances of that test, with the same

    randomly generated data, you can run it again with the

    " environment variable set to the given seed.

        TEST_RANDOM_SEED=20891494266 perl some_test.t

        1..3

        ok 1

        ok 2

        ok 3

        # TEST_RANDOM_SEED=20891494266

    See your shell and operating system's documentation for details on how

    to set environment variables.

CAVEATS

    If something in your code calls srand() all bets are off.

SEE ALSO

    Test::RandomResults, Test::Sims, Data::Random, Data::Generate