Thursday, 9 April 2015

DRY vs DAMP in Tests

In the previous post I mentioned that duplication in tests is not always bad. Sometimes duplication becomes a problem. Tests can become large or virtually identically excluding a few lines. Changes to these tests can take a while and increase the maintenance overhead. At this point, DRY violations need to be resolved.


Test Helpers

A common solution is to extract common functionality into setup methods or other helper utilities. While this will remove and reduce duplication this can make tests a bit harder to read as the test is now split amongst unrelated components. There is a limit to how useful such extractions can help as each test may need to do something slightly differently.

DAMP - Descriptive and Meaningful Phrases

Descriptive and Meaningful Phrases is the alter ego of DRY. DAMP tests often use the builder pattern to construct the System Under Test. This allows calls to be chained in a fluent API style, similar to the Page Object Pattern. Internally the implementation will still use literals or value objects, but each test can provide just the differences it needs in order to execute. The key point regardless of how DAMP tests are implemented is to favor readability over anything else, while still eliminating duplication where possible.

The example shows a typical arrange aspect of a test written in the DAMP style. The end result of this builder is we will have the ability to now act and assert against the result - a controller instance. If further tests were required we could use the same setup but simply provide different order dates for example. Additionally we could add or remove further chained calls. Behind the scenes the implementation of these builders is straightforward.

I tend to introduce this pattern after the third time of seeing duplication between tests. There is a bit of an overhead otherwise, the builder itself requires implementation and careful construction. Once you go past three tests the overhead pays itself off by allowing you to rapidly add new tests and make large, structural changes.

Beware the builders becoming too big or complex. If this starts to happen you may wish to refactor as there may be missing abstractions in your design. DAMP tests have numerous advantages, but they should be applied where required rather than for every scenario. Tests for objects that are lower in the dependency graph tend to fit into the more traditional testing patterns, while higher up your stack DAMP tests can prove useful.

No comments:

Post a Comment