Hi all,
The company I now work for uses testing modules developed here several years ago. One
of my
coworkers is curious to know how Perl's standard testing modules compare to the ones
they use.
He'll happily consider going with Perl's testing modules if there is a compelling
reason to do so.
I'm not particularly advocating one way or another, but I promised him I would shoot
an email to
the Perl QA list and ask for feedback. Frankly, I'm happy just to have testing here.
This is rather long and contains a fair number of samples, so I beg your patience.
Thanks!
Here's some sample code (stripped down for clarity):
package RTK::Retail::Product::TEST::Supplier;
use base qw/RTK::Retail::Meta::TestCase/;
use RTK::Meta::Test::Asserts;
use RTK::Retail::Product::TEST::Supplier::TestUtil;
use Aliased 'RTK::Retail::Product::Supplier';
sub TEST_FIND_ALL
{
assert_lists_eq([Supplier->find_all], []);
my $supplier1 = create_test_supplier(
-supplier_no => 1,
-name => 'Foo, Inc.'
);
my $supplier2 = create_test_supplier(
-supplier_no => 2,
-name => 'Bar, Inc.'
);
my @suppliers = sort map { $_->name } Supplier->find_all;
my @actual_suppliers = ('Bar, Inc.', 'Foo, Inc.');
assert_lists_eq([EMAIL PROTECTED],[EMAIL PROTECTED]);
my $supplier3 = create_test_supplier(
-supplier_no => 3,
-name => 'Baz, Inc.'
);
@suppliers = sort map { $_->name } Supplier->find_all;
@actual_suppliers = sort ('Baz, Inc.', 'Bar, Inc.', 'Foo, Inc.');
assert_lists_eq([EMAIL PROTECTED],[EMAIL PROTECTED]);
}
sub TEST_FIND_ALL_SORTED_BY_NAME
{
[snip]
Each /^TEST__\w+/ subroutine is run and has a number of asserts in it. Each test
subroutine is
considered to be a test of a single type of functionality and if all asserts pass, the
test is
considered successful. Here's a sample of the output, running one test.
TEST_FIND_ALL
TEST_FIND_ALL --> Passed! Life is Wonderful!
Running an entire test module (again, simplified for clarity):
TEST_FIND_ALL ..
Connecting to TEST database '[EMAIL PROTECTED]'
passed (0.12s - 0.07 CPU)
TEST_FIND_ALL_SORTED_BY_NAME ..passed (0.06s - 0.02 CPU)
The "Connecting to TEST database .." line is important. There are three test
databases are
completely empty. At the beginning of a test program, the user is assigned on of the
three
databases (test fails if all are in use) and they are responsible for populating it
with test
data. Those are the 'create_test_supplier' lines in the code. At the end of each
test function,
the data is automatically removed from the database. Each function has setup and
teardown
functions that are automatically run behind the scenes.
Running the entire test suite is simply a matter of using their rtk_test program which
spiders
through the entire directory structure and runs all tests that it finds in any module
(allowing
inline tests, if desired).
And a sample failure:
TEST_FIND_ALL ..
Connecting to TEST database '[EMAIL PROTECTED]'
FAILED!!
-- Test failed: Different values at index 1: 'Baz, Inc.' vs. 'Foo, Inc.': 'Baz,
Inc.' ne 'Foo,
Inc.' on /home/cxp/work/popup_sort/perl_lib/RTK/Retail/Product/TEST/Supplier.pm, line
34
Yikes!!. Errors in:
/home/cxp/work/popup_sort/perl_lib/RTK/Retail/Product/TEST/Supplier.pm
Let me know if anything is unclear. If you can think of any pros and cons of this
approach, I
would love to hear them.
Cheers,
Ovid
=====
Silence is Evil http://users.easystreet.com/ovid/philosophy/indexdecency.htm
Ovid http://www.perlmonks.org/index.pl?node_id=17000
Web Programming with Perl http://users.easystreet.com/ovid/cgi_course/
__________________________________
Do you Yahoo!?
Yahoo! SiteBuilder - Free, easy-to-use web site design software
http://sitebuilder.yahoo.com
