When should you use a computed property or a stored property?
Updated for Xcode 14.2
Properties let us attach information to structs, and Swift gives us two variations: stored properties, where a value is stashed away in some memory to be used later, and computed properties, where a value is recomputed every time it’s called. Behind the scenes, a computed property is effectively just a function call that happens to belong to your struct.
Deciding which to use depends partly on whether your property’s value depends on other data, and partly also on performance. The performance part is easy: if you regularly read the property when its value hasn’t changed, then using a stored property will be much faster than using a computed property. On the other hand, if your property is read very rarely and perhaps not at all, then using a computed property saves you from having to calculate its value and store it somewhere.
When it comes to dependencies – whether your property’s value relies on the values of your other properties – then the tables are turned: this is a place where computed properties are useful, because you can be sure the value they return always takes into account the latest program state.
Lazy properties help mitigate the performance issues of rarely read stored properties, and property observers mitigate the dependency problems of stored properties – we’ll be looking at them soon.
SPONSORED From March 20th to 26th, you can join a FREE crash course for mid/senior iOS devs who want to achieve an expert level of technical and practical skills – it’s the fast track to being a complete senior developer!
Sponsor Hacking with Swift and reach the world's largest Swift community!
