Tester's Digest

A weekly source of software testing news

TESTER’S DIGEST

ISSUE #12 - Apr 23, 2017


Today’s theme is unit tests, a huge space, so don’t expect exhaustive coverage!

Topic: Unit Tests

Why unit tests are worth the trouble:

https://hackernoon.com/treat-yourself-e55a7c522f71#.cdbivs605

A developer’s approach to writing (mostly unit) tests:

https://blog.nelhage.com/2016/12/how-i-test/

Unit test tips from everydayunittesting.com which is a blog and a book “being written in an agile manner”:

http://www.everydayunittesting.com/2017/02/unit-testing-anti-pattern-asserting-on-not-null.html

If your unit tests are difficult to write, that is a signal of poor design:

http://www.drdobbs.com/testing/the-relationship-between-testability-and/240167101

Test smells, i.e. markers of poor quality in your test code (not just for unit tests):

http://xunitpatterns.com/Test%20Smells.html

Always test these conditions or inputs: null, zero, one, many, too many. Obvious? I should hope so. But you still gotta do it. Every time.

https://medium.com/sqa-mate/always-test-these-5-conditions-in-software-ed79d7a5cdd5#.joopscsv8

A classic on using mock objects in your tests:

https://martinfowler.com/articles/mocksArentStubs.html

Best practices specific to RSpec, Ruby’s BDD test framework:

https://github.com/thoughtbot/guides/tree/master/best-practices#testing

Sazerac is an interesting data-driven framework for Javascript unit tests:

https://hackernoon.com/sazerac-data-driven-testing-for-javascript-e3408ac29d8c#.1b1fm0ipr

Off-Topic

Google’s analysis of the degree of flakiness in their corpus of 4.2 million tests, broken down by test size/type. They find that “large” tests, i.e. end-to-end/integration tests, are the flakiest at 14% bogus failure rate over a week, compared to “small” (unit) tests at only 0.5%, and conclude that one needs to weigh whether to write a “large” test at all, and be careful if doing so.

https://testing.googleblog.com/2017/04/where-do-our-flaky-tests-come-from.html

While it’s still April, this April’s Fools post on reducing your MTTR for production incidents was the best. Achieve zero MTTR through skipping the investigation and resolution stages and simply auto-closing the tickets as soon as they are opened. Can be applied to bugs, too!

https://victorops.com/blog/mttr-zero-one-weird-trick-solves-problems/


If you received this email directly then you’re already signed up, thanks! Else if this newsletter issue was forwarded to you and you’d like to get one weekly, then you can subscribe at http://testersdigest.mehras.net

If you come across content worth sharing, please send me a link at testersdigest@mehras.net