Reporting Context Aware Partial Translation engine based on immediate and delayed Rule application - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Rapport Année : 2022

Reporting Context Aware Partial Translation engine based on immediate and delayed Rule application

Résumé

Language migration has been the driver of many efforts resulting in multiple solutions and strategies. One of the most popular approaches for dealing with it is source code translation: it proposes to translate the source code of an application to a target language. For doing so it leverages a set of translation rules based on the grammatical constructions provided by the source and target languages. However, we notice that even when most of the literature acknowledges translating implies migrating the runtime, libraries and the Software Development Kit (SDK), none of them proposes a systematic way to solve this problem. Along with this, we notice that there are many proposals to shift the paradigm from procedural to object-oriented programming based on how to propose classes automatically. But we found nothing on how to translate the expressions that use functions into expressions that use methods. In the context of migration from Microsoft Access (MS Access) to web technologies, these two lacks threaten seriously any attempt to produce even a compilable version of the code on the target technology. This article proposes a translation engine that split the translation process into two phases. A phase of language translation, and a phase of adaptation to the target environment. The first phase is in charge of producing declarations, and the second one is in charge of adapting the usage of this declaration to fit the translated version of our artefacts. We argue that enabling to adapt the code to fit the translated version of our artefacts allows the definition of simple adapting rules able to deal with a large share of both problems: (i) runtime, libraries and the Software Development Kit (SDK), and (ii) simple paradigm shift. This article presents some basic adapting rules and validates our approach by translating a battery of simple tests that feature the usage of a carefully chosen set of features.
Fichier principal
Vignette du fichier
article.pdf (308.72 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03898154 , version 1 (14-12-2022)

Identifiants

  • HAL Id : hal-03898154 , version 1

Citer

Santiago Bragagnolo, Stéphane Ducasse, Nicolas Anquetil, Abderrahmane Seriai, Mustapha Derras. Reporting Context Aware Partial Translation engine based on immediate and delayed Rule application. INRIA Lille - Nord Europe. 2022. ⟨hal-03898154⟩
71 Consultations
46 Téléchargements

Partager

Gmail Facebook X LinkedIn More