Copy Page

You are not logged in. Any changes you make will be marked as anonymous. You may want to Log In if you already have an account. You can also Sign Up for a new account.

Here is an experiment for you. <img class="emoticon emoticon-smile" data-emoticon-name="smile" border="0" src="/s/en_GB/3278/15/_/images/icons/emoticons/smile.png" alt="(smile)" title="(smile)" />

<ol>
	<li>Measure the time it takes to run your tests without Ashcroft. Call the time TP (seconds).</li>
	<li>Measure the time it takes to run your tests with Ashcroft. Call the time TA (seconds).</li>
	<li>Count the number of tests that were run (they should be the same for both test executions). Call this number N</li>
</ol>

Calculate N/TP. This is the number of tests you could have run per seconds if all your tests had obeyed the <a class="confluence-link unresolved" data-content-title="Ashcroft Commandments" data-linked-resource-default-alias="Ashcroft Commandments" href="#">Ashcroft Commandments</a>. You will probably find that N/TP is somewhere around 100 test methods per second. Or 10 milliseconds per test method if you will. (TODO: verify these numbers on a couple of OSS projects)

Calculate TP/TA. This is your Testing Pain Index (TPI). It should be as close as possible to 1. Anything over 10 is extremely painful. The Test Execution Frequency (TEF) is inversely proportional to the TPI. Examples to come hopefully - in the meanwhile take it with a grain of salt. <img class="emoticon emoticon-smile" data-emoticon-name="smile" border="0" src="/s/en_GB/3278/15/_/images/icons/emoticons/smile.png" alt="(smile)" title="(smile)" />

Does this sound intriguing to you? If it does, go ahead and enable Ashcroft and follow his advice to make your tests fast. -And your business logic decoupled from infrastructure!