Author Archives:

Fail Loudly

I ended my last post, Generating Test Cases at Runtime, with a pop quiz: “Can you spot the bug? It’s possible to write test methods that never get invoked.” Today, let’s cover the bug as well as its fix. Reproducing … Continue reading 

Posted in Uncategorized | 3 Comments

Generating Test Cases at Runtime

Last time, we saw how Fixie can integrate with AutoFixture. That was a situation in which parameterized tests were meant to be called once. They were parameterized because the producer of the inputs was interesting while the count was uninteresting. … Continue reading 

Posted in Uncategorized | Leave a comment

AutoFixie

Last week, we saw how the Fixie test framework deals with test method parameters. Your custom convention calls a Parameters(…) hook, supplying an explanation of where parameters should come from and how many times the test method should be called. … Continue reading 

Posted in Uncategorized | 3 Comments

What’s in a Name?

In my last post, we saw how Fixie lets you define conventions for parameterized tests. My goal was to provide a way for convention authors to tell Fixie what inputs should be passed into any given test method, allowing for … Continue reading 

Posted in Uncategorized | 1 Comment

A Swiss Army Katana

Before now, test methods for the Fixie test framework had to have zero parameters. If your test method had a parameter, it would fail without being called. Fixie would have no idea what values to pass in. As of Fixie … Continue reading 

Posted in Uncategorized | 1 Comment

Test Execution Order

In a perfect world, the order of execution of your tests shouldn’t matter. If each test is isolated and makes no unfair assumptions about the state under test, they’ll behave the same no matter the order. In reality, we’re all … Continue reading 

Posted in Uncategorized | 3 Comments

Shufflin’ Ain’t Sortin’

I recently added a test-shuffling feature to the Fixie test framework. The basic idea is that you can opt into randomizing the order of execution for tests within a test class. I’ll cover the details of the test shuffling feature … Continue reading 

Posted in Uncategorized | 5 Comments

Austin Code Camp 2013

At today’s Austin Code Camp, I gave a talk titled, “Fixie: a Test Framework for the Discerning Control Freak”. It covered the motivation behind Fixie as well as a series of customization examples. All of the examples are based on … Continue reading 

Posted in Uncategorized | Leave a comment

Tiny Nudges in the General Direction of Success

Yesterday, I described a leaky abstraction in the Fixie test framework. I introduced a complex and viral solution to a simple and localized problem. Eventually, it caused a bug: consumers who customized Fixie in just the right way could experience … Continue reading 

Posted in Uncategorized | Leave a comment

The Leakiest Abstraction

Centuries from now, when archaeologists are digging through the rubble patch located over the former home of the GitHub servers, they will recover a copy of Fixie’s repository. Inspecting its history, an unfortunate soul will discover a particularly horrible abstraction. … Continue reading 

Posted in Uncategorized | 8 Comments