Category: integration test

Simplest solutions are usually the best solutions.

We as software developers work with hard problems and solve a lot of small problems every day. Solving a hard problem itself is a hard job. Though in my opinion it's not enough to solve a hard problem in any possible way but a hard problem should be solved with a simple solution. When a developer comes up with a simple solution to a hard problem then they can declare the problem solved.

First a disclaimer. Coming up with a simple solution to a hard problems is itself a very hard problem and takes a lot of time, effort and practice.

I've seen my share of "clever" solutions for hard problems and the problem with those is that usually the solution itself is so hard to understand that depending on the size of the problem it may take a developer from hours to days or even weeks to understand how that "clever" solution works. It's a rare occasion when a developer has come up with a simple solution to a hard problem. So simple that it needs to be read only once and it makes sense. I dare to say I've made a few of these solutions and I've also seen these from other developers but way less often than hard to understand solutions for hard problems.

So how does one come up with simple solutions to hard problems?
First step is to split the problem to smaller problems. How this works for me is that I try to identify smaller problem areas within the original problem. Once I have identified smaller problems I can usually find yet smaller problems to solve within those and I continue to do it until I have a plan, a set of tasks or small problems to solve. Usually I identify even more problems to solve once I start to work on them and I just add them to the list of small problems to solve. With this iterative process I can solve the problem one small piece at a time.

Splitting the problem to multiple smaller problems gives other benefits too. Each small problem can be tested individually and the big problem's tests work as acceptance tests for the whole set of small problems. Smaller problems are easier to solve and therefore the code is easier to write and when the code is easier to write it's easier to write readable code.

In a ideal situation the big problem would be solved by sequentially calling functions that solve a smaller problem within the problem space. In a way it could be thought like the solution to a problem can be solved by a function introduced in a interface. That interface is tested with a acceptance tests and it can be tested with mocks or spies to verify it calls the correct functions in correct order with correct parameters. The implementation of the interface is actually a series of function calls that each solve a portion of the problem. Each of those functions is tested with unit tests. Each of functions can be written as easily readable by keeping them small and naming the functions and things within those functions meaningfully. When a function is small and solves a single problem the solution is easy to define with immutable data structures or by avoiding mutating the variables. Also when each function works as it's own unit it's easier to isolate integration tests to those functions.

This is how simple solutions are crafted and this is how all the topics I have covered earlier are tied together to create simple, readable, verified and long lasting solutions to problems.

With all this tied up what started with a working title "My version of clean code" could also be simplified and after thinking about it more while writing these posts I decided to name the approach and conventions as "Simple code".