Luis Pedro

Composing Visual Syntax for Domain Specific Languages

With the increasing interest in metamodeling techniques for Domain Specific Modeling Languages (DSML) definition, there is a strong need to improve the language modeling process. One of the problems to solve is language evolution. Possible solutions include maximizing the reuse of metamodel patterns, composing them to form new, more expressive DSMLs. In this paper we improve the process of rapid prototyping of DSML graphical editors in meta-modeling tools, by defining composition rules for the graphical syntax layer. The goal is to provide formally defined operators to specify what happens to graphical mappings when their respective metamodels are composed. This improves reuse of Domain Specific Modeling Languages definitions and reduces development time.

Principles for System Prototype and Verification Using Metamodel Based Transformations

Using domain specific modeling (DSM) allows solutions to be expressed in the idiom and at the level of abstraction of the problem domain. However, this does not imply that prototypes can be easily and rapidly generated. In reality, domain specific languages (DSLs) are difficult to design, implement and maintain, and usually there is a potential loss of efficiency when compared with hand-coded software. In this paper we explain the principles based on which we expect to solve some of these problems by means of transformation from a DSL to a formalism with a well define semantics named concurrent object oriented Petri-nets (CO-OPN). The proposed methodology uses the metamodel of the DSL as the principle for the transformation. This transformation represents the semantic mapping between the DSL and CO-OPN. The achievement is both to provide a formally defined semantics to the DSL and, since CO-OPN is integrated in a framework, to provide the functionalities that allow model verification and fast prototype generation for the DSL

A test selection language for CO-OPN specifications

In this paper we propose a test language that allows expressing test intentions for CO-OPN (concurrent object-oriented Petri nets) specifications - a formal specification language designed to handle large complex concurrent systems. Our test language is based on temporal logic formulas for expressing graphs of input/output pairs - the inputs correspond to operations performed on the system and the outputs to the observable results of those operations. We encapsulate the temporal logic using a language of constraints, which purpose is to shape the tests that are to be produced. In this paper we discuss the syntax and provide the semantics of this test language. One of our main worries while designing the test language were to keep it modular in order to promote reusability. Another worry was to be able to cope with non-determinism coming from the system under test. We illustrate managing non-determinism as well as other features of our language by showing how we can generate tests for the login part of an e-banking system. A framework for editing CO-OPN specifications exists and one of its features is the possibility of automatically generating high level Java prototypes that can be completed/extended by human developers. We discuss the applicability and the usefulness of our test language while verifying systems built using this methodology.

Formal test generation from UML models

In this paper we will explain our approach for generating test cases for a UML system model. Despite the fact that UML authors claim that UML semantics are precise enough to define non-ambiguous models, we find that the overlap of the different views makes it difficult to explore and make deductions on the state space of the modeled system in order to generate test cases. Our approach is thus based on a subset of UML (inspired from the Fondue approach) for which we have defined clear transformation semantics. We provide these semantics by delineating transformation rules using the MDA (Model Driven Architecture) architecture. We transform UML models into CO-OPN (Concurrent Object Oriented Petri Nets) ones, CO-OPN being a formal specification language defined in our Laboratory. We have also defined a language for expressing test intentions for CO-OPN models. This language allows selecting interesting executions (tests cases) of a model by providing constraints over all possible traces of that model. By exploring the models semantics with the tools we have built for our CO-OPN language we are able to generate test cases based on those test intentions. We are also able to partially eliminate redundancy in the produced test cases by finding equivalence classes in the model operations inputs.

Prototyping domain specific languages with COOPN

The work described in this article presents how we use COOPN in the context of the MDA (Model Driven Architecture) philosophy for prototyping Domain Specific Languages. With this principle we increase the abstraction of COOPN language representation enabling standard data interchange with other applications that use the same approach. In particular we will present the architecture of the transformation from Domain Specific Languages; its advantages concerning the ability to have COOPN models as a standard format for representing the semantics of Domain Specific Languages and to reuse software prototyping and testing techniques developped for this formalism. As example we will show how our work is proceeding towards transformation from UML to COOPN. We also argue how our approach can be easily used in order to produce rapid system prototyping and verification for Domain Specific Languages (DSLs).

DSL Composition for model-based test generation

Domain specific languages (DSL) which describe reactive systems generally have a need for systematic generation of tests for their models. During the design of a DSL there is a lack of support for its integration with existing model based test generation tools. In this paper, we show how this integration can be conceptualized and systematized. We introduce a framework for composing DSLs for reactive systems, with a particular DSL for Model Based Testing called SATEL (Semi-Automatic Testing Language). This DSL composition is achieved by composing both the syntaxes of the two DSLs and their semantics. The result of this composition is also a language where it is possible to express models in the target DSL and test specifications for those models. The semantics of the composed language corresponds to the generation of test cases for models expressed in the target DSL. We finish the paper by analyzing the compositional framework we obtained in terms of its applicability to other target DSLs.