A language for generic programming in the large

Abstract

a b s t r a c tGeneric programming is an effective methodology for developing reusable software libraries. Many programming languages provide generics and have features for describing interfaces, but none completely support the idioms used in generic programming. To address this need we developed the language G. The central feature of G is the concept, a mechanism for organizing constraints on generics that is inspired by the needs of modern C++ libraries. G provides modular type checking and separate compilation (even of generics). These characteristics support modular software development, especially the smooth integration of independently developed components. In this article we present the rationale for the design of G and demonstrate the expressiveness of G with two case studies: porting the Standard Template Library and the Boost Graph Library from C++ to G. The design of G shares much in common with the concept extension proposed for the next C++ Standard (the authors participated in its design) but there are important differences described in this article.

Type classes as objects and implicits

“…Also, an existing comparison between different languages in terms of their support for generic programming in the large (Siek and Lumsdaine 2008) is revised to include Scala and JavaGI. This comparison shows that Scala is very suitable for generic programming in the large.…”

“…Furthermore, prioritized overlapping implicits allow the definition of type-level functions with a default, catch-all case, that can be overridden by other cases in subclasses. This functionality provides a limited form of concept-based overloading (Siek and Lumsdaine 2008). In Haskell such overlapping definitions are forbidden for associated types and type-families (Chakravarty et al 2005b;Schrijvers et al 2008); and have limited applicability with functional dependencies.…”

“…In particular type classes support retroactive extension: the ability to extend existing software modules with new functionality without needing to touch or re-compile the original source. Type classes are also recognized (Bernardy et al 2008;Garcia et al 2007;Siek and Lumsdaine 2008) as a good mechanism for concept-based C++ style generic programming (Musser and Stepanov 1988), and have more recently evolved into a mechanism for type-level computation (Chakravarty et al 2005b;Jones 2000;Schrijvers et al 2008).…”

Generic programming and high-performance libraries

“…Concept checking [16,17] verifies syntactic concept conformance of arguments to generic functions, reducing the amount of irrelevant and misleading information in error messages, while postprocessing of compiler error messages [15] eliminates redundant information and improves clarity. Concept archetypes [16,17], on the other hand, are minimal syntactic models of concepts that can be passed to generic functions to verify that the generic functions do not require syntax not captured in a concept. Concept-based overloading has enjoyed more success in C++, both via the widely-used method of tag dispatching [12] and via arbitrary overloading [18].…”

“…Presence or absence of a functor with a suitable signature can be detected in languages such as ML or Haskell through the use of signatures. This is possible even in C++, currently through the use of the Boost Concept-Checking Library [16,32] but possibly in the future with concept constraints expressed within the language as proposed by Stroustrup and Dos Reis [33][34][35] to the C++ standards committee. In none of these cases, however, is there provision to check for satisfaction of the axioms.…”