Attribute grammars are a formalism for specifying programming languages. They have been applied to a great number of systems automatically producing language implementations from their specifications. The systems and their specification languages can be evaluated and classified according to their level of application support, linguistic characteristics, and degree of automation.A survey of attribute grammar-based specification languages is given. The modern advanced specification languages extend the core attribute grammar model with concepts and primitives from established programming paradigms. The main ideas behind the developed attribute grammar paradigms are discussed, and representative specification languages are presented with a common example grammar. The presentation is founded on mapping elements of attribute grammars to their counterparts in programming languages. This methodology of integrating two problem-solving disciplines together is explored with a classification of the paradigms into structured, modular, object-oriented, logic, and functional attribute grammars. The taxonomy is complemented by introducing approaches based on an implicit parallel or incremental attribute evaluation paradigm.

Attribute grammar paradigms—a high-level methodology in language implementation

The program transformation principle called partial evaluation has interesting applications in compilation and compiler generation. Self-applicable partial evaluators may be used for transforming interpreters into corresponding compilers and even for the generation of compiler generators. This is useful because interpreters are significantly easier to write than compilers, but run much slower than compiled code. A major difficulty in writing compilers (and compiler generators) is the thinking in terms of distinct binding times: run time and compile time (and compiler generation time). The paper gives an introduction to partial evaluation and describes a fully automatic though experimental partial evaluator, called mix, able to generate stand-alone compilers as well as a compiler generator. Mix partially evaluates programs written in Mixwell, essentially a first-order subset of statically scoped pure Lisp. For compiler generation purposes it is necessary that the partial evaluator be self-applicable. Even though the potential utility of a self-applicable partial evaluator has been recognized since 1971, a 1984 version of mix appears to be the first successful implementation. The overall structure of mix and the basic ideas behind its way of working are sketched. Finally, some results of using a version of mix are reported.

/pdf/mix-a-self-applicable-partial-evaluator-for-experiments-in-hqvfeg9z8u.pdf

Mix - a self-applicable partial evaluator for experiments in compiler generation

C together with its descendants represents a strong and indisputable status quo in the current software industry. It is a very popular general-purpose programming language, characterized by its economy of expression, its large set of operators and data types, and its concern for source code portability. The current reference document for C is the international standard ISO/IEC 9899:1990. The semantics of C is informally defined in the standard, using natural language. This causes a number of ambiguities and problems of interpretation about the intended semantics of the language. In this thesis, a formal denotational semantics for the ANSI C programming language is proposed, with emphasis on its accuracy and completeness with respect to the standard. It is demonstrated that a programming language as useful in practice and as inherently complicated as C can nonetheless be defined formally. The proposed semantics could be used as a precise, unambiguous, abstract and implementation-independent standard for the language. Moreover, it would be a basis for the formal reasoning about C programs and a valuable theoretical tool in the software development process. In order to improve in modularity and elegance, the proposed semantics uses several monads and monad transformers to model different aspects of computations. Interesting results have been achieved in the attempt to accurately model complex characteristics of C, such as the unspecified order of evaluation and sequence points, using monad notation. These results may be useful in specifying the semantics of programming languages supporting non-deterministic features and parallelism. An implementation of an abstract interpreter for C programs based on the proposed semantics has also been developed, using Haskell as the implementation language. The implementation has been used to evaluate the accuracy and completeness of the proposed semantics. Although this process is still under way, the results so far have been entirely satisfactory.

https://www.softlab.ntua.gr/~nickie/Papers/papaspyrou-1998-fscpl.pdf

A Formal Semantics for the C Programming Language

Action Semantics is a framework for the formal description of programming languages. Its main advantage over other frameworks is pragmatic: action-semantic descriptions (ASDs) scale up smoothly to realistic programming languages. This is due to the inherent extensibility and modifiability of ASDs, ensuring that extensions and changes to the described language require only proportionate changes in its description. (In denotational or operational semantics, adding an unforeseen construct to a language may require a reformulation of the entire description.)

Theory and Practice of Action Semantics

Formal semantic description is significant for design, reasoning and standardization of programming languages, and it plays an important part in the optimization of the compiler. However, compared to the amount of effort that has been made to the research of various semantic frameworks over more than forty years, their actual applications are definitely frustrating. This survey reviews the history of developments on semantic description frame-works for programming languages. It also illustrates features and actual applications of the main frameworks (including operational, deno-tational, axiomatic and hybrid semantics). In some practical aspects, such as comprehensibility, extensibility and applicability, the qualitative comparisons of these frameworks are given distinctly. It suggests that a more popular formal semantic description should behave more elegantly in readability, modularity, abstractness, comparability, reasonability, applicability and tool-support.

A survey of semantic description frameworks for programming languages

We describe how the denotational semantics of a programming language can be executed directly, if it is expressed in a suitable functional programming language such as ML. We also apply Mosses' idea of ‘semantic algebras’ to construct semantic descriptions that are significantly more modular and understandable than usual. The possibility of executing a language's semantic description directly supports a methodology of language design that we advocate: express the design as a formal language description, and use this to test and refine the design, before becoming committed to constructing a compiler.

David A. Watt

Papers

Executable semantic descriptions