I dagens digitala samhälle är tester på applikationer och kodbaser ett vanligt förekommande fenomen. Tester utförs på flera företag både manuellt och automatiskt och vill gärna rekrytera personer med kunskap inom testning. Mittuniversitetet i Sundsvall har ännu inga kurser för testning men det skulle vara passande för studenter att bli introducerade till detta redan tidigt när de läser ett program. Syftet med denna studie har varit att identifiera ett passande ramverk för kurser om testning på universitet. För att kunna genomföra studien utfördes en kvalitativ litteraturstudie om testning. Sedan valdes två stycken ramverk anpassade för enhetstestning utifrån en kravspecifikation. De utvalda ramverken var JUnit 5 och TestNG. Ramverken testades utifrån dess användbarhet, funktionalitet och effektivitet genom att köra tester mot en befintlig chattapplikation. Vidare utfördes användartest för att undersöka ramverkens lämplighet för användning under en universitetskurs där målgruppen hade olika erfarenheter och datorvana. Resultatet visade på att JUnit 5 var bättre när det kom till användning, dokumentation och lämplighet för en kurs på universitet. Rekommendationen är att JUnit 5 passar till att använda i en kurs på universitet. Det rekommenderas också att universitet bör introducera kurser med testning så att studenten naturligt implementerar ett iterativt förhållningssätt till programmering samt blir mer förberedd inför arbetslivet. / In today’s digital society, testing of applications and code bases is a common phenomenon. Testing are performed on several companies, both manually and automatically, and would like to recruit people with knowledge about testing. Mid Sweden University in Sundsvall does not yet have courses about testing but it would be appropriate for students to be introduced to this early on while they study a program. The purpose with this study has been to identify a suitable framework for university courses about testing. In order to be able to carry out the study, a qualitative literature study on testing was performed to gather information. Furthermore two framework were selected that was adapted for unit testing and based on a requirements specification. The chosen frameworks were JUnit 5 and TestNG. The frameworks were tested based on their usability, functionality and efficiency by running tests against an existing chat application. Furthermore, user tests were performed to examine the suitability of the frameworks for use during a university course where the target group have different experiences and computer skills. The result showed that JUnit 5 was better when it came to usability, documentation and suitability for a course at a university. The recommendation is that JUnit 5 is suitable for a course at a university. It’s also recommended that the university should introduce courses with testing so that the student naturally implements an iterative approach to programming and become more prepared for working with programming and development.

Testing is an important part in the software development process. Unit testsaim to test individual units in isolation. These units may have dependenciesto their surroundings that make the units hard to test in isolation withoutalso testing the surrounding units. A technique to help isolate these units isto replace the surrounding units with mock objects. This work investigateshow a C++ mock framework can be integrated into a modeling environ-ment’s unit test framework. Several mock frameworks are evaluated, and aproof of concept is created to show that integration is possible. Additionally,ideas for how to use mocks in a model environment are presented.

As the use of renewable energy and the implementation of smart grids becomemore prevalent in Europe, there will be a need to ensure that the quality ofpower supply is not compromised during the integration of distributedgeneration to the main grid. Europe's electricity networks should be flexible,accessible, reliable and economic. In the UK, National Grid has standardisedits substation protection and control equipment commissioning andreplacement policies, yet issues affecting system long life availability remain,one reason being long outage periods during substation secondary equipmentinstallation, commissioning and maintenance. The present use of directhardwired point to point connections between the primary power system plantequipment and substation secondary system protection and control devicesdoes not allow for easy upgrading or replacement of these substationsecondary devices without an outage of the primary plant or substation. Outage and the consequent availability problems associated with secondaryequipment can be addressed by the open utility communication architecturestandard IEC 61850. A well-designed simple, highly reliable, secure, flexibleand long-life communication IEC 61850-based architecture can help mitigatethe impact of using protection and control IEDs (Intelligent Electronic Devices).Faulty IEDs can be replaced with little or no interruption to the overall operationof the substation. Interoperability is a key feature of the adoption of IEC 61850in substations. IEC 61850-compliant protection and control devices cancommunicate with one another, even if they made from different manufacturers. This thesis has proposed a simple, long life IEC 61850 based communicationarchitecture which is expected to be flexible and robust enough to cope withboth growth and outages. Reliability analyses have been carried out on varioushypothetical applications of the proposed process bus architecture to NationalGrid substation bays. A detailed description of how to determine the optimalprocess bus architecture using the life cycle cost evaluation technique has beenprovided. The design and implementation of a test bed used for evaluating theperformance characteristics of merging units has been presented. The resultsof the tests have been fed back to National Grid and the manufacturers, whomay then use the data to assist with the drafting of a Merging Unit Test BedSpecification, and also to help the manufacturers to make refinements to themerging units in order to make interoperability more readily achievable.

The thesis is focused on the JUnit framework, which currently represents the most popular framework for unit testing across the Java platform. The paper begins by giving a definition of necessary theory, with focus on unit testing. The goal of this section is to introduce the reader to a test theory on a general level (test classi-fication, cost of bugs, etc.). The last part of the introduction is related to, and focused on, unit testing. The next part of this paper (after the introduction at the beginning) is focused on the JUnit framework itself. The aim is to provide general information about the framework, followed with the description of architecture and detailed specification of every feature for the JUnit framework in version 5 (including the usage of features). The last passage explains options of an extension model within the JUnit framework. The description of the JUnit framework is enhanced with relevant information to version 5 (the summary of changes and expected finalization of the version 5.0) and the description of the framework integration to its surroundings (the other test tools and frameworks) on Java platform. Additionally, the thesis is extended with these appendices: - The organization and concept of source code used by examples. - The complex example of usage of framework Junit. It covers everything from the creation the project (with usage of Spring Boot framework on Java platform) to the implementation of relevant unit tests. - The work-around to support JUnit framework of the version 5 in Eclipse tool.

No description available.

Syftet med studien har varit att undersöka vilka utmaningar utvecklare upplever vidgenomförande av enhetstester. Enhetstester är testkod som skrivs för att verifiera att produktionskod eller så kallad vanlig kod fungerar på ett bra sätt och uppfyller sitt syfte. Detta är den typ av testning vilket görs först i en utvecklingsprocess för att garantera att koden håller god kvalitét. Studien genomfördes genom att 5 utvecklare på ett IT konsultföretag i Luleå medverkat i individuella öppna kvalitativa intervjuer där de fått svara på frågor kring deras upplevelser av att arbeta med enhetstest. De slutsatser som studien har kommit fram till är att många av de identifierade problem som framkommit i denna studie stämmer överens med tidigare forskning inom detta ämne. Huvudsakligen var dessa problem av teknisk natur, där exempelvis underhåll, att veta vad som ska testas och att veta när en har testat tillräckligt, hörde till de vanligast förekommande gemensamma. Detta visar att denna problematik finns kvar än i dag. Vad som även framkommit i denna studie är att det är vanligt med avsaknad avutbildning gällande enhetstest. Även avsaknad av någon gemensam metod är något vi, speciellt i kombination med avsaknad av utbildning, anser förvärrar problematiken som redan existerar. Med hjälp av studiens kvalitativa ansats så har det identifierats en del potentiellabidragande faktorer för dessa upplevda problem. Utifrån dessa slutsatser har det formulerats rekommendationer för organisationer och för vidare forskning. / The purpose of the study has been to investigate what challenges developers experience when working with unit tests. Unit tests are test code written to verify that production code or so-called regular code works well and fulfills its purpose. This is the type of testing which is done first in a development process to guarantee that the code maintains good quality. The study was carried out by having 5 developers at an IT consulting company in Luleå participate in individual open qualitative interviews where they had to answer questions about their experiences of working with unit tests. The conclusions that the study has reached are that many of the identified problems that emerged in this study are consistent with previous research on this topic. Mainly these problems were of a technical nature, where, for example, maintenance, knowing what to test and knowing when one has tested enough, were among the most commonly occurring ones. This shows that these problems still exist today. What also emerged in this study is that it is common to lack education regarding unit tests. Also the lack of a common method is something we, especially in combination with the lack of education, consider to worsen the problems that already exist. With the help of the study's qualitative approach, some potential contributing factors for these perceived problems have also been identified. Based on these conclusions, recommendations have been formulated for organizations and for further research.

This study explores a solution to promote Test-Driven Development in the embedded development field, a field that has not adopted the use as widely as other software development fields. In this study, an evaluation process of Unit Test Framework was created. In Test-Driven Development, creating unit tests is an essential part where Unit Test Framework is a software and tool to help developers create such unit tests. The target group is junior embedded developers, catering to a niche group with the reason for introducing and promoting Test-Driven Development early on. This research was conducted specifically in the software development field for embedded microcontrollers in C++.  The first research area was about what junior embedded developers think are the most important features and criteria a Unit Test Framework should consist of and support. Those results, along with the author's own opinions from their experience of using Unit Test Framework, a list of features and criteria was compiled for the model to evaluate. The model tests and evaluates criteria using both an observational and experimental method. The result outlines each framework's different strengths and weaknesses and lowers the knowledge barrier, thereby promoting Test-Driven Development and the use of Unit Test Frameworks.

Programinės įrangos testavimas yra kritinė viso kūrimo proceso dalis. Kokybe privalo rūpintis ne vien testuotojai. Programuotojai turi naudoti įvairias kokybės užtikrinimo technikas, tokias kaip statinė kodo analizė (kodo peržiūra) ir vienetų testavimas. Kartais programuotojai praleidžia kodo testavimą arba atlieka tik paviršutinį testavimą, nes tai užimą jų laiką. Automatiniai testavimo įrankiai suteikia galimybes testavimo kaštų sumažinimui. Sukurta daug įvairių testavimo atvejų generavimo technikų naudojant specializuotus įrankius. Testavimo atvejų generavimo tikslas yra pasiekti kuo didesnį kodo padengimą ir rasti netikėtas klaidas. Daugumoje atvejų yra neįmanoma pasiekti visišką kodo padengimą, naudojant vien generuotus testavimo atvejus. Reikalingi testavimo atvejai, parašyti žmogaus. Logiška pakartotinai naudoti testus, kuriais buvo ištestuoti panašūs metodai anksčiau. Šiame darbe analizuojama testų pakartotino panaudojimo problema. Siūlomas sprendimas yra paremtas vienetų testų saugojimu ir pakartotinu panaudojimu lyginant testuojamą metodą ir jau ištestuotą, kuris turi išsaugotus vienetų testus. Siūlomas metodas turėtų būti naudojamas kartu su vienetų testų generavimo įrankiais. / Software testing is critical part in software development. Not only testers are responsible for software quality. Programmers should use quality assurance techniques as code static analysis (code review) and unit testing. Sometimes developers omit testing their code or perform just superficial testing because unit testing consumes development time. Automated unit testing tools provide the way for reducing of testing costs. There are various techniques for generating test cases by using specialized tools. The goal of generation of tests cases is to achieve higher code coverage and expose unexpected errors. In many cases it is impossible to achieve full code coverage by using only generated test cases. Then manual efforts are needed. It is reasonable to reuse unit tests for method similar to that already tested. This work helps to analyse unit test reusability problem. The approach presented is based on unit test storing and reusing by comparing current testing method and method that has saved tests. This method should be used in combination with particular unit test generation tool.

This master’s thesis deals with the design and implementation of part of test environment for automatic testing of electronic control unit in the production. The electronic control unit is manufactured by UNIS, a.s. and is used as a control and power electronics for the hydraulic unit. The theoretical part of the thesis deals with the development and automatic testing of electronics in the field of aerospace engineering. The practical part describes the proposed automatic test environment and implementation of testing software using hardware and software products by National Instruments. The main platforms used for software development are LabVIEW and TestStand. Created test environment is finally verified by testing several control units.

In this thesis the application of software defect prediction to predict unit test failure is investigated. Data for this purpose was collected from a Continuous Integration development environment. Experiments were performed using semantic features from the source code. As the data was imbalanced with defective samples being in minority different degrees of oversampling were also evaluated. The data collection process revealed that even though several different code commits were available few ever failed a unit test. Difficulties with linking a failure to a specific file were also encountered. The machine learning model used in the project produced poor results when compared against related work, from which it was based on. In F-measure, it on average achieve 53% of the mean performance of state-of-the-art for software defect prediction on bugs in Java source files. Specifically, it would appear that very little information was available for the model to learn defects in files not present in training data. / I denna avhandling undersöks applikationen av prognos för mjukvarudefekter för att förutse enhetstestfel. Data för detta syfte samlades in från en utvecklingsmiljö med kontinuerlig integration. Experimenten utfördes med användning av semantiska särdrag samlade från källkod. Då data var obalanserat med defekta exempel i minoritet evaluerades olika grader av översampling. Datainsamlingsprocessen visade att även om det fanns många kodinlämningar så misslyckades få någonsin ett enhetstest. Svårigheter med att länka testmisslyckanden till en specifik fil påträffades också. Den använda maskininlärningsmodellen uppvisade också dåliga resultat i jämförelse med relaterade värk. Mätt i F-measure uppnåddes i genomsnitt 53% av genomsnittlig prestandan av bästa möjliga prognos av mjukvarudefekter av buggar i Java källkod. Specifikt så framträdde det att väldigt lite information verkar finnas för modellen att lära sig defekter i filer som ej fanns med i träningsdata.