Employing classifying terms for testing model transformations

Abstract

Abstract-This contribution proposes a new technique for developing test cases for UML and OCL models. The technique is based on an approach that automatically constructs object models for class models enriched by OCL constraints. By guiding the construction process through so-called classifying terms, the built test cases in form of object models are classified into equivalence classes. A classifying term can be an arbitrary OCL term on the class model that calculates for an object model a characteristic value. From each equivalence class of object models with identical characteristic values one representative is chosen. The constructed test cases behave significantly different with regard to the selected classifying term. By building few diverse object models, properties of the UML and OCL model can be explored effectively. The technique is applied for automatically constructing relevant source model test cases for model transformations between a source and target metamodel.

Diversity of graph models and graph generators in mutation testing

“…The above proposed generalized neighborhood-based shaping procedure can easily incorporate such additional white-box knowledge in the form of extra predicates (similarly to the classifying terms proposed in [22,26]). The additional predicates discussed in Sect.…”

“…Model comparison techniques [54] frequently rely upon the existence of node identifiers, which can easily lead to many isomorphic models. Since checking graph isomorphism is computationally costly, practical solutions tend to use approximation techniques to achieve certain diversity by random sampling [31], incremental generation [33,51], using symmetrybreaking predicates [56], or -in case of a white-box testing -by generating models satisfying (or violating) relevant graph predicates [51] (possibly derived from OCL expressions [22,26]).…”

“…In this paper (which extends our initial work in [50]), we propose and evaluate various diversity metrics to characterize a single model and a set of models. For that purpose, we innovatively reuse neighborhood graph shapes [43] and graph predicate abstraction techniques [22,26,44] which refine the type of each graph object based on the structure of its neighborhood (e.g., incoming and outgoing edges) and the truth values of predicates (e.g., a node satisfies a predicate or not). The internal diversity of a single model will be characterized by the number of shape nodes covered by the model, which generalizes traditional metamodel coverage (used in many research papers).…”

An OCL-Based Framework for Model Transformations

“…The work [26] proposes a technique for developing test cases for UML and OCL models. By guiding the construction process through so-called classifying terms, the built test cases in form of object models are classified into equivalence classes.…”

Testing models and model transformations using classifying terms

“…More specifically, we show how this approach of guiding the construction of test cases in an orderly, systematic and efficient manner can be effectively used in the specification and testing of both directional and bidirectional model transformations. The work presented in this paper is a revised and enhanced version of our original paper [26], extended to improve the description of the testing process of model transformations using classifying terms, to cover the testing of bidirectional transformations, and by providing the results of the validation experiments we have conducted to assess the usability, effectiveness and scalability of our proposal. This paper is organized in 7 sections.…”