Thesis (M.A.)--Rowan University, 2005. / Typescript. Includes bibliographical references.

In this thesis we present a language and technique to facilitate model-based testing. The core of our approach is an xUnit-like language that allows tests to refer to model entities such as associations. This language can be used by developers to describe tests based on an existing UML model. The tests might even be written before creating a UML model, and be based on requirements. The testing language, including its parser and generators, is written entirely in Umple, an open-source textual modeling tool with semantics closely based on UML, and which generates Java, PHP and several other target languages. Tests in our language can be embedded in Umple or in standalone files. The test language compiler converts our abstract testing language into JUnit, PHPUnit and other domain-language testing environments. In addition to allowing developers to write tests manually, we have created generators that create abstract tests for any Umple model. These generators can be used to verify the Umple compiler and to give Umple users extra confidence in their models. User-defined tests can be standalone or embedded in methods; they can be generic, referring to metamodel elements. Tests can also be located in traits or mixsets to allow testing of separate concerns or product lines. To test our language and the tests written in it, we have created an extensive test suite. We have also implemented mutation testing, that enables varying of features of the models to ensure that runs of the pre-mutation tests then fail.

The aim of this master's thesis is to deal with the problems of tests and testing. Firstly, tests, testing items, problems of solvability and unbiased assessment were described. On the basis of particular language tests - level A1 according to Common European Framework of Reference for Languages, which test foreigners' Czech (a foreign language for them) - the success rate of receptive skills (reading) and productive skills (writing) is compared. Concerning writing skills, there is a problem how to reach unbiased assessment. Finally, linguodidactic recommendation intended mainly for teachers of Czech as a foreign language is formulated. Keywords test, test's task, language level A1

Testing is an important part of software development. It ensures that the developed product is of high standard and quality. Tieto frequently develops large complex systems which require comprehensive testing. Testing employs manually designed test cases. According to recent development within software testing it has been shown that design of test cases can be simplified with domain specific test languages (DSTL).     The thesis project is a research and development study in the field of software testing and has been performed at Tieto office in Karlstad. The project concerns the development of a DSTL and a suitable development environment based on the Eclipse Platform. The project evaluates the development of a domain specific test language using the tools Eclipse JDT, developed by Eclipse Foundation, and Xtext, developed by itemis AG. The experiment was evaluated by its functionality and integrability.     The project has shown promise in using a domain specific test language with a suitable development environment. The result and evaluation have shown that the subject shows promise, but needs further development if it is to be adapted within Tieto´s organization.

The issue of civic integration and requirements for citizenship has been a far-reaching debate that has taken place in the Swedish political context for several years. From the Liberals’ pioneering proposal for a language test for immigrants in 2002, the issue has 19 years later moved towards an increasing consensus that today finds majority support in the Swedish parliament. Although the proposal seems to contribute to greater integration, there are many critical approaches to analyse. The question that arises is whether these types of proposals, through their discursive design, create conditions that actually support integration or instead induce discrimination and exclusion of certain groups. The purpose of this thesis is twofold. One aim is to critically examine and analyse the political proposal on language- and civic knowledge for Swedish citizenship by looking at government documents and motions from Sweden's opposition parties. The thesis examines whether racist discourses are reproduced or maintained by analysing the proposal to review further how it is legitimised. The secondary purpose will be to problematise how such a proposal can affect the right to citizenship based on the risks of discrimination and exclusion. This is done by guidance from Fairclough's three- dimensional model, whose method allows to examine the textual structure, the discursive and social practice of the given material. The theoretical framework evolved from theories that sufficiently understand how civic requirements can be understood based on different exclusion mechanisms. The essay concludes that the proposal raises issues of human rights, that the state legitimises the proposal through discursive means and that there are tendencies that reproduce racist discourses.

Le développement de tests pour les systèmes d’avioniques met en jeu une multiplicité de langages de test propriétaires, sans aucun standard émergent.Les fournisseurs de solutions de test doivent tenir compte des habitudes des différents clients, tandis que les échanges de tests entre les avionneurs et leurs équipementiers / systémiers sont entravés. Nous proposons une approche dirigée par les modèles pour s’attaquer à ces problèmes: des modèles de test sont développés et maintenus à la place du code, avec des transformations modèle-vers-code vers des langages de test cibles. Cette thèse présente trois contributions dans ce sens. La première consiste en l’analyse de quatre langages de test propriétaires actuellement déployés. Elle nous a permis d’identifier les concepts spécifiques au domaine, les meilleures pratiques,ainsi que les pièges à éviter. La deuxième contribution est la définition d’un méta-modèle en EMF Ecore, qui intègre tous les concepts identifiés et leurs relations. Le méta-modèle est la base pour construire des éditeurs de modèles de test et des motifs de génération de code. Notre troisième contribution est un d´démonstrateur de la façon dont ces technologies sont utilisées pour l’élaboration des tests. Il comprend des éditeurs personnalisables graphiques et textuels pour des modèles de test, ainsi que des transformations basées-motifs vers un langage du test exécutable sur une plateforme de test réelle. / The development of tests for avionics systems involves a multiplicity of in-house test languages, with no standard emerging. Test solution providers have to accommodate the habits of different clients, while the exchange of tests between aircraft manufacturers and their equipment/system providers is hindered. We propose a model-driven approach to tackle these problems: test models would be developed and maintained in place of code, with model-to-code transformations towards target test languages. This thesis presents three contributions in this direction. The first one consists in the analysis of four proprietary test languages currently deployed. It allowed us to identify the domain-specific concepts, best practices, as well as pitfalls to avoid. The second contribution is the definition of a meta-model in EMF Ecore that integrates all identified concepts and their relations. The meta-model is the basis for building test model editors and code generation templates. Our third contribution is a demonstrator of how these technologies are used for test development. It includes customizable graphical and textual editors for test models, together with template-based transformations towards a test language executable on top of a real test platform.