The Perl Toolchain Summit needs more sponsors. If your company depends on Perl, please support this very important event.

NAME

TB2::Tester - Testing a Test:: module

SYNOPSIS

    use Test::More;
    use Your::Test::Module qw(this_ok that_ok);
    use TB2::Tester;

    my $capture = capture {
        this_ok $this, "some name";
        that_ok $that;
    };

    my $results = $capture->results;

    # The first one passed, and it has a name
    result_like shift @$capture, {
        is_pass => 1,
        name => "some name",
    };

    # The second one failed, and it has no name
    result_like shift @$capture, {
        is_pass => 0,
        name => ''
    };

DESCRIPTION

This is a module for testing Test modules.

Exports

These are exported by default

capture

    my $capture = capture { ...test code ... };

Captures all the events and results which happens inside the block.

Returns a TB2::History that you can reference later. This is disassociated from any other tests, so you do what you like to it without altering any other tests.

event_like

  event_like( $event, $want );
  event_like( $event, $want, $name );

Tests that a $result looks like what you $want.

$want is a hash ref of keys and values. Each of these will be checked against the $result's attributes. For example...

    result_like( $result, { name => "foo" } );

will check that $result->name eq "foo".

Values can also be regular expressions.

    result_like( $result, { name => qr/foo/ } );

will check that $result->name =~ /foo/.

result_like

  result_like( $result, $want );
  result_like( $result, $want, $name );

Works just as event_like but it also checks the $result is a result.

EXAMPLES

Avoid hardcoding the test sequence

    my $results = $history->results;
    result_like $results->[0], {
        is_pass => 0,
        name    => "this is that",
        file    => $0,
    };
    result_like $results->[1], { ... }
    result_like $results->[2], { ... }

The drawback with using array indices to access individual results is that once you decide to add or remove a test from any place but the very end of the test list, your array indices will be wrong and you'll have to update them all.

To preclude this issue from arising, simply shift the individual results off the result list, like this:

    result_like shift @$results, { ... }; # check first result
    result_like shift @$results, { ... }; # next one, and so on

Now you only have to add checks symmetrically with your new tests - existing lines won't have to be edited. It's safe to modify $results because it is only the results for your captured tests.