http://www.bugfree.dk/blog/2009/06/19/basic-unit-testing-guidelines/ Not anti-anything, just pro-quality Thu, 16 Dec 2010 07:18:58 +0000 hourly 1 http://wordpress.org/?v=3.3 http://www.bugfree.dk/blog/2009/06/19/basic-unit-testing-guidelines/comment-page-1/#comment-22117 Bugfree.dk » Blog Archive » 2009 in retrospect Mon, 28 Dec 2009 11:28:55 +0000 http://www.bugfree.dk/blog/?p=111#comment-22117 [...] Basic unit testing guidelines [...] [...] Basic unit testing guidelines [...]

]]> http://www.bugfree.dk/blog/2009/06/19/basic-unit-testing-guidelines/comment-page-1/#comment-16558 Bugfree.dk » Blog Archive » Basic logging guidelines Wed, 22 Jul 2009 22:42:28 +0000 http://www.bugfree.dk/blog/?p=111#comment-16558 [...] Bugfree.dk My random technology musings « Basic unit testing guidelines [...] [...] Bugfree.dk My random technology musings « Basic unit testing guidelines [...]

]]> http://www.bugfree.dk/blog/2009/06/19/basic-unit-testing-guidelines/comment-page-1/#comment-16402 Shyam Thu, 16 Jul 2009 20:14:41 +0000 http://www.bugfree.dk/blog/?p=111#comment-16402 Agreed, I especially agree on the naming conventions for the test. I love it when a test fails and I know exactly what is wrong, rather than having to dig in deep and look at the line of code. And I totally believe that you should write scenarios in your test rather than looking at each line of code and setting expectations so it hits it. You get a much better sense of what your method does when the tests are real world scenarios rather than random things to get 100% coverage. Agreed, I especially agree on the naming conventions for the test. I love it when a test fails and I know exactly what is wrong, rather than having to dig in deep and look at the line of code. And I totally believe that you should write scenarios in your test rather than looking at each line of code and setting expectations so it hits it. You get a much better sense of what your method does when the tests are real world scenarios rather than random things to get 100% coverage.

]]> http://www.bugfree.dk/blog/2009/06/19/basic-unit-testing-guidelines/comment-page-1/#comment-16386 Nirav Assar Thu, 16 Jul 2009 01:25:11 +0000 http://www.bugfree.dk/blog/?p=111#comment-16386 very well explained. Even though strict unit testing definitely has its benefits, some of which you mentioned like isolation of bugs, testable code, and code confidence as the test suite grows, unit tests have its limitations. when it comes to meeting the requirements of the applications, unit testing just satifies module by module. integration tests carry more of the burden to meet the "behavior" of the system. Read my post at http://assarconsulting.blogspot.com/2009/05/clearing-up-mudd-explaining-various_22.html explaining TDD and BDD. I try to point out the differences an benefits of each. TDD works very well with unit testing, although the larger benefit is driving the design of code. very well explained. Even though strict unit testing definitely has its benefits, some of which you mentioned like isolation of bugs, testable code, and code confidence as the test suite grows, unit tests have its limitations. when it comes to meeting the requirements of the applications, unit testing just satifies module by module. integration tests carry more of the burden to meet the “behavior” of the system. Read my post at http://assarconsulting.blogspot.com/2009/05/clearing-up-mudd-explaining-various_22.html explaining TDD and BDD. I try to point out the differences an benefits of each. TDD works very well with unit testing, although the larger benefit is driving the design of code.

]]> http://www.bugfree.dk/blog/2009/06/19/basic-unit-testing-guidelines/comment-page-1/#comment-16374 Ronnie Wed, 15 Jul 2009 08:39:03 +0000 http://www.bugfree.dk/blog/?p=111#comment-16374 @Jimmy: I agree. Code coverage is an important metric. But even a 100% code coverage doesn't guarantee bug free code. Therefore I'd focus on writing tests that cause one to reflect on ones design and business requirements. @Jimmy: I agree. Code coverage is an important metric. But even a 100% code coverage doesn’t guarantee bug free code. Therefore I’d focus on writing tests that cause one to reflect on ones design and business requirements.

]]> http://www.bugfree.dk/blog/2009/06/19/basic-unit-testing-guidelines/comment-page-1/#comment-16373 Jimmy the Geek Wed, 15 Jul 2009 08:00:42 +0000 http://www.bugfree.dk/blog/?p=111#comment-16373 I would work at code coverage with the unit tests as a goal. It is the sections of code that aren't hit often that cause 95% of the bugs. I would work at code coverage with the unit tests as a goal. It is the sections of code that aren’t hit often that cause 95% of the bugs.

]]>