CppCon 2018

C++ is often described as a type-safe language — the compiler will not allow the programmer to mix types freely. But do we really use C++ in a type-safe manner? When a programmer uses the same type to represent different quantities, one could argue that the compiler's type-checks are bypassed and errors due to mix-ups not only become possible, but frequent! In this session a solution to this problem will be presented — strong types.

Strong types are a type-safe alternative to typedefs — but where typedefs only provide documentation and ease of writing, their strong counterparts are actually enforced by the compiler. We will explore how they work, and how using them can improve code quality. We will look at how one can implement a strong typedef by oneself, but also look at some of the readily available implementations which are available.

Finally we will look at how we can handle some special cases, e.g. where we need to allow two different strong types to interact with each other in well defined ways. These could be types representing two different linked quantities like positions and offsets; or scaled versions of the same quantity like different units of length.