Many years ago, I took part in the development of a taxi-hailing mobile app that is still widely used today. I don’t know what kind of code they’re running now, but in those early days, the driver assignment code –if I remember it correctly– was similar in spirit to the grossly simplified example that follows. There are five levels of nested if statements in less than 30 lines of code. It doesn’t look so bad, some might say, but it’s not difficult to imagine how complicated this code can become with just a few more checks…
EDIT: read the article turns out it’s super useful… It gives insight into decision table which is a pattern I did not know about until recently…
Is this really a recurring design pattern for y’all?
I mean, you can just use a switch. anyways I’ll read the article and see ;)
Also take a look at the Specification Pattern for something similar.
That’s something I would only use if the logic becomes very complex, but it can help break things down nicely in those cases.
Decision tables are nice. They hide the important part of the logic away out of view of another programmer trying to figure out a bug in the code.
Very helpful! You take longer to find and fix bugs, and potentially miss a few extra ones because of stuff like this. Increased tech debt. Highly recommended! 👍
I mean, you can just right click “Definition” in VSCode and see how it works… It’s not that inconvenient.
It’s easy to read, write and refactor so I don’t really see what you mean.
What’s fun is determining which function in that list of functions actually is the one where the bug happens and where. I don’t know about other langauges, but it’s quite inconvenient to debug one-linres since they are tougher to step through. Not hard, but certainly more bothersome.
I’m also not a huge fan of un-named functions so their functionality/conditions aren’t clear from the naming, it’s largely okay here since the conditional list is fairly simple and it uses only AND comparisons. They quickly become mentally troublesome when you have OR mixed in along with the changing booleans depending on which condition in the list you are looking at.
At the end of the day though, unit tests should make sure the right driver is returned for the right conditions. That way, you know it works, and the solution is resistant to refactor mishaps.
I can’t find it right now, but there is some explanation in “Clean Code” why switches shouldn’t be used all over the place.
Google for “replace conditional with polymorphism”.
Just checked and it is in “Clean Code” - Chaper 17; Section G23 “Prefer Polymorphism to if/else or switch/case”.
This is really terrible advice. Sometimes it’s better to do that, but definitely not in the example from this article.
If anyone says you should always prefer polymorphism to switches they are a bloody idiot.
Thank you!