Background
Contextual abstractions are a way to abstract over context. They represent a unified paradigm with a great variety of use cases, among them:
- Implementing type classes
- Establishing context
- Dependency injection
- Expressing capabilities
- Computing new types, and proving relationships between them
Other languages have been influenced by Scala in this regard. E.g., Rust’s traits or Swift’s protocol extensions. Design proposals are also on the table for Kotlin as compile time dependency resolution, for C# as Shapes and Extensions or for F# as Traits. Contextual abstractions are also a common feature of theorem provers such as Coq or Agda.
Even though these designs use different terminology, they’re all variants of the core idea of term inference: given a type, the compiler synthesizes a “canonical” term that has that type.
Scala 3 Redesign
In Scala 2, contextual abstractions are supported by marking definitions (methods and values) or parameters as implicit (see Context Parameters).
Scala 3 includes a redesign of contextual abstractions. While these concepts were gradually “discovered” in Scala 2, they’re now well known and understood, and the redesign takes advantage of that knowledge.
The design of Scala 3 focuses on intent rather than mechanism. Instead of offering one very powerful feature of implicits, Scala 3 offers several use-case oriented features:
-
Retroactively extending classes. In Scala 2, extension methods are encoded by using implicit conversions or implicit classes. In contrast, in Scala 3 extension methods are now directly built into the language, leading to better error messages and improved type inference.
-
Abstracting over contextual information. Using clauses allow programmers to abstract over information that is available in the calling context and should be passed implicitly. As an improvement over Scala 2 implicits, using clauses can be specified by type, freeing function signatures from term variable names that are never explicitly referred to.
-
Providing Type-class instances. Given instances allow programmers to define the canonical value of a certain type. This makes programming with type-classes more straightforward without leaking implementation details.
-
Viewing one type as another. Implicit conversions have been redesigned from the ground up as instances of a type-class
Conversion. -
Higher-order contextual abstractions. The all-new feature of context functions makes contextual abstractions a first-class citizen. They are an important tool for library authors and allow to express concise domain specific languages.
-
Actionable feedback from the compiler. In case an implicit parameter can not be resolved by the compiler, it now provides you import suggestions that may fix the problem.
Benefits
These changes in Scala 3 achieve a better separation of term inference from the rest of the language:
- There’s a single way to define givens
- There’s a single way to introduce implicit parameters and arguments
- There’s a separate way to import givens that does not allow them to hide in a sea of normal imports
- There’s a single way to define an implicit conversion, which is clearly marked as such, and does not require special syntax
Benefits of these changes include:
- The new design thus avoids feature interactions and makes the language more consistent
- It makes implicits easier to learn and harder to abuse
- It greatly improves the clarity of the 95% of Scala programs that use implicits
- It has the potential to enable term inference in a principled way that is also accessible and friendly
This chapter introduces many of these new features in the following sections.