Declarative Debugging of Membership Equational Logic Specifications

Recent Trends in Algebraic Development Techniques

Caballero

et al. 2009

Self Cite

Abstract. Declarative debugging is a semi-automatic technique that starts from an incorrect computation and locates a program fragment responsible for the error by building a tree representing this computation and guiding the user through it to find the wrong statement. This paper presents the fundamentals for the declarative debugging of rewriting logic specifications, realized in the Maude language, where a wrong computation can be a reduction, a type inference, or a rewrite. We define appropriate debugging trees obtained as the result of collapsing in proof trees all those nodes whose correctness does not need any justification. Since these trees are obtained from a suitable semantic calculus, the correctness and completeness of the debugging technique can be formally proved. We illustrate how to use the debugger by means of an example and succinctly describe its implementation in Maude itself thanks to its reflective and metalanguage features.

“…The work encompasses and extends previous presentations [5,6] on the declarative debugging of Maude functional modules, which constitute now a particular case of a more general setting.…”

Section: Discussionmentioning

confidence: 67%

“…The application of declarative debugging to Maude functional modules was already studied in our previous papers [5,6]. The executability requirements of Maude functional modules mean that they are assumed to be confluent, terminating, and sort-decreasing 1 [10].…”

Section: Introductionmentioning

confidence: 99%

Declarative Debugging of Rewriting Logic Specifications

Recent Trends in Algebraic Development Techniques

Caballero

et al. 2009

Self Cite

“…The source code of the debugger, examples, and much more information is available at http://maude.sip.ucm.es/debugging/. Detailed proofs of the results shown in this paper and extended information about the transformations can be found in [2].…”

Section: Introductionmentioning

confidence: 99%

Simplifying Questions in Maude Declarative Debugger by Transforming Proof Trees

Caballero

Logic-Based Program Synthesis and Transformation

et al. 2012

Declarative debugging is a debugging technique that abstracts the execution details that in general may be difficult to follow in declarative languages to focus on results. It relies on a data structure representing the wrong computation, the debugging tree, which is traversed by asking questions to the user about the correctness of the computation steps related to each node. Thus, the complexity of the questions is an important factor regarding the applicability of the technique. In this paper we present a transformation for debugging trees for Maude specifications that ensures that any subterm occurring in a question has been previously replaced by the most reduced form that it has taken during the computation, thus ensuring that questions become as simple as possible.

“…The evaluation semantics takes a pair consisting of a command and a store and returns a store. 2 However, we have committed an error in the while loop:…”

Section: System Module Example: Operational Semanticsmentioning

confidence: 99%

“…Functional modules define data types and operations on them by means of membership equational logic theories that support multiple sorts, subsort relations, equations, and assertions of membership in a sort. Declarative debugging of functional modules has been presented in [2,1]. System modules specify rewrite theories that also support rules, defining local concurrent transitions that can take place in a system.…”

Section: Introductionmentioning

confidence: 99%

A Declarative Debugger for Maude

Algebraic Methodology and Software Technology

Martí-Oliet

et al.

Self Cite

Abstract. Declarative debugging has been applied to many declarative programming paradigms; in this paper, a declarative debugger for rewriting logic specifications, embodied in the Maude language, is presented. Starting from an incorrect computation (a reduction, a type inference, or a rewrite), the debugger builds a tree representing this computation and guides the user through it to find a wrong statement. We present the debugger's main features, such as support for functional and system modules, two possible constructions of the debugging tree, two different strategies to traverse it, use of a correct module to reduce the number of questions asked to the user, selection of trusted vs. suspicious statements, and trusting of statements "on the fly."