Hi Milos.
On 17 Jul 2007, at 10:05, Milos Prodanovic wrote:
> It would be nice to make some functionality tests, and I'm not sure
> that I can make
> tests of all possible 'use cases', but I'm glad to do tests if you
> have some crucial test
> that could show usability.
Please do make tests. For every check-in of code, we verify on every
platform we support that the entire regression suite passes its
tests. Any patch is incomplete without tests that show its
functionality and give proof that it's not broken accidentally in the
future by someone else. As the author of the code, you're better
equipped to write its tests than anyone else.
We also verify that all lines of code are covered by the test suite,
so the tests should exercise every kind of branch in the tests. Your
test should intentionally create errors, and catch them. It should
show that the freaky edge cases work. It should use new
functionality in a greater context to show it works well with
others. It will receive a lot of attention, and serve as examples
and reference for the documentation team to use when describing the
functionality in the manual.
In short, the tests are at least as important as the code.
- chad
After "make", "{ cd mysql-test; ./mtr
$optional_specific_test_name; }". Write new queries & directives at
"mysql-test/t/${test_name}.test" and write the output at "mysql-test/
r/${test_name}.result". I suggest "type_net_addresses" is a good
test name. You may use "-record" to make the result, but don't
blindly do it; recording a wrong result is much worse than not having
a result.
--
Chad Miller, Software Developer chad@stripped
MySQL Inc., www.mysql.com
Orlando, Florida, USA 13-20z, UTC-0400
Office: +1 408 213 6740 sip:6740@stripped
Attachment: [application/pgp-signature] This is a digitally signed message part PGP.sig