Programming languages interoperability

Researchers : G. Ospina

Collaborations : F. Gobert (INGI, UCL, Belgium)

Framework : Centre Fédéré en Vérification funded by the FNRS

Funds : FNRS, FSR

Publication : here

Description :

Interoperability at the application level is an unavoidable issue in many practical situations such as reusing existing "legacy" applications. Moreover, it is quite often the case that the components to make interoperate are written in different programming languages. Hence, interoperability of programming languages is an important practical problem. But, at the theoretical level, no satisfactory solution to this problem has been proposed yet.

Our objective is to develop a methodology to specify interoperability mechanisms between programming languages as different as C, Java, Mercury, ML, Prolog, or Python. We propose to specify the operational semantics of all these languages into a common mathematical framework and we develop and compare various interoperability mechanisms on top of these semantics.

Another main unsolved problem is to make these programming languages interoperate at the implementation level on concrete operating systems. We want to implement interfacing tools that implement the interoperability mechanisms defined at the theoretical level in a way that is compatible with existing implementations of the programming languages. Incompatibility issues must be identified clearly and pragmatic solutions must be proposed (such as tuning an interoperabily mechanisms experimentally).

Finally, we want to develop tools to allow multilanguage programming based on different programming languages designed independently of each other. (This should be contrasted with the more common approach of a unique multiparadigm language, such as Oz, for instance.)