Algorithmic debugging generalized

[EN] Algorithmic debugging is a semi-automatic debugging technique that abstracts the operational details of computations, allowing the programmers to debug their code from an abstract point of view. However, its use in practice is still marginal, and one of the reasons is the lack of precision of t...

Descripción completa

Detalles Bibliográficos
Autores: Insa Cabrera, David, Silva, Josep|||0000-0001-5096-0008
Tipo de recurso: artículo
Fecha de publicación:2018
País:España
Institución:Universitat Politècnica de València (UPV)
Repositorio:RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia
Idioma:inglés
OAI Identifier:oai:riunet.upv.es:10251/121763
Acceso en línea:https://riunet.upv.es/handle/10251/121763
Access Level:acceso abierto
Palabra clave:Algorithmic debugging
Transformation
Generalization
LENGUAJES Y SISTEMAS INFORMATICOS
Descripción
Sumario:[EN] Algorithmic debugging is a semi-automatic debugging technique that abstracts the operational details of computations, allowing the programmers to debug their code from an abstract point of view. However, its use in practice is still marginal, and one of the reasons is the lack of precision of this technique when reporting errors (current algorithmic debuggers do not point an expression or line as buggy, but they point a whole procedure/function/method as containing the bug). In this paper, we make a step forward to overcome this problem. We identify two specific causes of that problem in the standard formulation and implementations of algorithmic debugging, and we present a reformulation to solve both problems. We show that the novel ideas included in the reformulation proposed cannot be supported by the standard internal data structures (such as the Execution Tree) used in this technique and, hence, a generalisation of the standard definitions and algorithms is needed. The reformulation has been done in a language independent manner to make it useful and reusable in different programming languages. (C) 2018 Elsevier Inc. All rights reserved.