Test::Unit::Assertion::CodeRef - A delayed evaluation assertion using a Coderef


    require Test::Unit::Assertion::CodeRef;

    my $assert_eq =
      Test::Unit::Assertion::CodeRef->new(sub {
        $_[0] eq $_[1]
          or Test::Unit::Failure->throw(-text =>
                                          "Expected '$_[0]', got '$_[1]'\n");

    $assert_eq->do_assertion('foo', 'bar');

Although this is how you'd use Test::Unit::Assertion::CodeRef directly, it is more usually used indirectly via Test::Unit::Test::assert, which instantiates a Test::Unit::Assertion::CodeRef when passed a Coderef as its first argument.


Test::Unit::Assertion::CodeRef implements the Test::Unit::Assertion interface, which means it can be plugged into the Test::Unit::TestCase and friends' assert method with no ill effects.


This class is used by the framework to allow us to do assertions in a 'functional' manner. It is typically used generated automagically in code like:

    $self->assert(sub {
                    $_[0] == $_[1]
                      or $self->fail("Expected $_[0], got $_[1]");
                  }, 1, 2); 

(Note that if Damian Conway's Perl6 RFC for currying ever comes to pass then we'll be able to do this as:

    $self->assert(^1 == ^2 || $self->fail("Expected ^1, got ^2"), 1, 2)

which will be nice...)

If you have a working B::Deparse installed with your perl installation then, if an assertion fails, you'll see a listing of the decompiled coderef (which will be sadly devoid of comments, but should still be useful)


Copyright (c) 2001 Piers Cawley <>.

All rights reserved. This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself.