Object oriented languages are an important tool to achieve software reusability in any kind of application and can increase dramatically software productivity in some special fields; they are also considered a logical step in the evolution of object oriented languages. But these languages lack a formal foundation, which is needed both to develop tools and as a basis for the future evolution of these languages; they lack also a strong type system, which would be essential to assure that level of reliability which is required to large evolving systems. Recently some researches have tried to insulate the basic mechanisms of object oriented languages and to embed them in strongly typed functional languages, giving to these mechanism a mathematical semantics and a set of strong type rules. This is a very promising approach which also allows a converging evolution of both the typed functional and the object oriented programming paradigms, making it possible a technology transfer in both directions. Most works in this field are very technical and deal just with one aspect of object oriented programming; many of them use a very similar framework. In this work we describe and exemplify that common framework and we survey some of the more recent and promising works on more specific features, using the framework introduced. We describe the results achieved and point out some problems which are still open, especially those arising from the interactions between different mechanisms.
Modelling features of object-oriented languages in second order functional languages with subtypes
GHELLI, GIORGIO
1991-01-01
Abstract
Object oriented languages are an important tool to achieve software reusability in any kind of application and can increase dramatically software productivity in some special fields; they are also considered a logical step in the evolution of object oriented languages. But these languages lack a formal foundation, which is needed both to develop tools and as a basis for the future evolution of these languages; they lack also a strong type system, which would be essential to assure that level of reliability which is required to large evolving systems. Recently some researches have tried to insulate the basic mechanisms of object oriented languages and to embed them in strongly typed functional languages, giving to these mechanism a mathematical semantics and a set of strong type rules. This is a very promising approach which also allows a converging evolution of both the typed functional and the object oriented programming paradigms, making it possible a technology transfer in both directions. Most works in this field are very technical and deal just with one aspect of object oriented programming; many of them use a very similar framework. In this work we describe and exemplify that common framework and we survey some of the more recent and promising works on more specific features, using the framework introduced. We describe the results achieved and point out some problems which are still open, especially those arising from the interactions between different mechanisms.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
