Testavimas yra svarbus ir imlus darbui aplikacijos kūrimo ir vystymo procesas. Norint suvaldyti augančias testavimo sąnaudas, siekiama dalį žiniatinklio aplikacijos testavimo proceso automatizuoti. Žiniatinklio aplikacijų atnaujinimo ciklas yra trumpas – žiniatinklio aplikacijos atnaujinamos dažnai, todėl žiniatinklio aplikacijos testavimas tampa ypač aktualus. Automatinis testavimas yra vertinamas nevienareikšmiškai, tai yra taip pat imlus darbui uždavinys, nes reikalauja apibrėžti testavimo procesą ir sukurti modelį automatiniam testų vykdymui. Vis dėlto, parengus automatinį testavimą, žiniatinklio aplikacijos testavimas palengvėja – sumažinamas rankinio testavimo kiekis. Darbo tikslas – parinkus geriausią žiniatinklio aplikacijų automatinio testavimo įrankį, pasiūlyti metodą žiniatinklio aplikacijų automatiniam testavimui. Uždaviniai – įvertinti žiniatinklio aplikacijų automatinio testavimo įrankius, parinkti geriausią įrankį, sukurti automatinio testavimo modelį. Rezultatai – magistro baigiamajame darbe nustatytas žiniatinklio aplikacijos automatiniam testavimui geriausiai tinkantis įrankis, kuriuo remiantis sukurtas ir įvertintas įrankis automatiniam žiniatinklio aplikacijų testavimui; apibrėžtas žiniatinklio aplikacijos automatinio testavimo modelis (sukurtas įrankis pritaikomas modelyje). / Testing is important and labor intensive part of application development. In order to manage growing testing efforts, we try to automate part of the web application testing process. Web application testing is important because the timespan between web application updates is short and web applications are updated often. Automated testing is a two-edged effort, it is also labour intensive task, because one has to define testing process and create a model for automated test execution. However after automated testing succeeds – testing takes less efforts, amount of manual testing is reduced. The objective of this work is after selecting the best automated web application testing tool to introduce an approach to automated web application testing. Goals: to evaluate web application automated testing tools, identify best tool, create a model of automated testing. Results: we identified the best tool fitting for automated testing of web application. Created and evaluated automated web application testing tool, defined and implemented a model of automated testing, implemented tool adapted in the model.

ECDL programa skirta kompiuterinio raštingumo įgūdžiams formuoti ir tobulinti. Šios programos pasiekimas yra tai, kad nustatytas standartas kompiuterio vartotojo žinioms ir įgūdžiams sertifikuoti. ECDL sertifikavimas iš pradžių buvo atliekamas rankiniu būdu. Vėliau buvo pereita prie kompiuterizuoto testavimo, tam sukuriant specialias testavimo sistemas. ECDL programoje įgytų įgūdžių įvertinimui Lietuvoje naudojama testavimo sistema dirba su interneto klientais. Teisę prie jos jungtis turi tik įgaliotieji ECDL testavimo centrai. Testavimo sistemos užduotys yra keleto tipų: pagrįstos tiesioginiu atlikimu, imitavimu, spragtelėjimo ir tempimo, „tiesa/netiesa“ pasirinkimu bei daugybiniu pasirinkimu. Iki šiol nėra spręsta problema, ar efektyviai realizuotoje testavimo sistemoje atliekamas įvairių užduočių tipų tikrinimas, ar testuojant atsižvelgiama į žmogaus įgūdžių kokybę. Kadangi ECDL testų tikrinimas vykdomas tiek rankiniu būdu, tiek automatizuotu būdu, todėl būtų labai naudinga išanalizuoti įvairius testavimo būdus ir įvertinti rankinio ir automatizuoto testavimo būdų galimybes kokybės požiūriu. Pasirinkta tyrimų sritis yra pakankamai plati, apimanti efektyviausių testavimo metodų, vartotojų reakcijos į testus problemas. Tiek užsienyje, tiek Lietuvoje yra atlikta nemažai tyrimų, siekiant sukurti efektyvų vartotojų testavimo metodą ar patobulinti jau esamus. Šio tyrimo objektas – ECDL rankinio ir automatizuoto testavimo būdų galimybės. Tyrimo tikslas – išanalizavus ir įvertinus... [toliau žr. visą tekstą] / The problem of ECDL testing methodologies in approach to quality is disputed in this master’s work. This work studies ECDL testing methodologies and their problems, analyses the pros and cons of automated test assembly and computer adaptive testing methods. It also analyses the advantages and disadvantages of automated and manual testing. The purpose of this work is to suggest a new ECDL testing system architecture, which ensures effective and efficient testing of people’s computer skills and competence, as the quality of those computer skills is very important. The objects of this paper are the possibilities of ECDL manual and automated testing. The main goal – is to suggest a new ECDL testing system architecture, after analyzing and evaluating ECDL testing possibilities. The main tasks of this master’s work are: to evaluate testing methods in approach to quality; to suggest a new ECDL testing system conceptual architecture model, which includes best practice management sub-system; to find best practice elements according to ECDL Advanced Syllabus 1.0; to perform experiments connected with the use of best practice elements in the ECDL Advanced testing. The main results of the research. The expert survey was conducted in order to find out, what experts think about the need to complete ECDL tasks correctly and with a touch of quality. The second purpose of this survey was to find out, if best practice elements are needed, and if yes, then for which ECDL modules. The results... [to full text]

ECDL CAD testų administravimo sąsajos tyrimas. / Research of ECDL CAD Test Administration Interface.

The results of empirical tests of the capital asset pricing model (CAPM) are discussed in this paper. A formidable problem here involves setting up an effective method for testing or test methodology. Many conceptual and statistical problems are inherent in tests of capital asset pricing model. It always has to be concerned the possible contaminating effects of the inevitable real-world violations of the model’s assumption. The tests reported in this paper are tests of how well the model fits history. The purpose of this work is to determine whether the CAPM fits the real world and, if it does not, to determine the source and size of the discrepancies between the model and the world. The data of Lithuanian firm’s stocks are used in order to test the model. The asset market of Lithuania is young and unstable. Thus the results are not the best. But recently situation is getting better. It will be shown in this paper that capital asset pricing model could be tested with real data of Lithuanian stock market. The results are quite good. Microsoft Visual Basic 6.3 equipment is used to solve this problem.

Program of Analysis of Machine Testing is the process of: • Creating additional test cases to increase coverage, and • Determining a quantitative measure of code coverage, which is an indirect measure of quality. • Identifying redundant test cases that do not increase coverage. The program automates this process. You use coverage analysis to assure quality of your set of tests, not the quality of the actual product. You do not generally use a coverage analyzer when running your set of tests through your release candidate. Coverage analysis requires access to test program source code and often requires recompiling it.

Software for mobile devices is becoming increasingly popular. The limited resources and limited user interface brings new issues in software development process for mobile devices. These limitations should be considered especially in design and testing phases of software development process. The software for mobile devices can not be thoroughly tested, because unit tests can not be stored in mobile device due to the lack of storage on it. Testing results can not be easily observed on small screens of mobile devices. This work depictures the influence of mobile technologies to software development process. The influence is exemplified with a project “Mobile Professor”. This work concentrates on problems encountered in software testing process for mobile devices especially in unit testing. The work presents other authors suggested solutions for earlier mentioned problems, and also suggest the new approach for performing unit testing for mobile devices, which does not suffer for these limitations of mobile devices. Existing and suggested methods are compared and evaluated experimentally.

Pagrindinis darbo tikslas - realizuoti programinę įrangą, gebančią automatiškai testuoti vartotojo sąsają ir testinius atvejus generuojančią iš duotų UML modelių. Kuriama programinė įranga turėtų būti realizuota naudojant Java technologiją. Taigi šio darbo tyrimo sritis apims vartotojo sąsajos testavimo automatizavimo būdų analizę. UML modeliai tapo ypač populiaria priemone modeliuojant programinės įrangos architektūrą. UML modeliai šiais laikais naudojami ne tik įprastoms klasių diagramoms, veiklos diagramoms, sekų diagramoms modeliuoti ir pan., bet tampa vis populiaresnia priemone ir yra taikomi daugelio kitų projektavimo uždavinių sprendimui. Vartotojo sąsajos atvaizdavimas UML diagramomis vis dar nėra populiarus būdas, nors literatūroje vis dažniau sutinkame siūlymų naudoti šias technologijas. Taigi literatūroje atsirandantys straipsniai, vykdomos konferencijos ir panašaus pobūdžio įvykiai byloja, kad ateityje ši technologija neaplenks ir vartojo sąsajos modeliavimo proceso. Rinkoje jaučiamas vartotojo sąsajos automatinio testavimo sistemos, kuri gebėtų testinius atvejus generuoti iš UML modelių ir automatiškai vykdyti testavimą, trūkumas. Tokia sistema galėtų būti naudojama efektyvesniam testavimui atlikti. Tokia sistema gebėtų testuoti vartotojo sąsają greičiau ir tiksliau. Greitis pasiekiamas tuo, kad automatinio testavimo procese nedalyvauja žmogus, o visą testavimą atlieka įrenginio procesorius. Testuojant sistemą skirtingose platformose testavimą galima atlikti... [toliau žr. visą tekstą] / In many cases, testing is an essential, but time and resource consuming activity in the software development process. In the case of model-based development, test construction and test execution can be partially automated. As the application size is constantly growing, the need for automated testing frameworks comes into place, particularly frameworks for automated testing of user interaction and graphical user interface. This document describes an implementation of the GUI test generator framework based on UML models where specific UML activity diagrams are used for test case generation. It is not a usual case to use UML activity diagrams for UI modeling. However the existing stereotypes of activity diagram elements are not suitable for UI modeling. With usual activity diagram it is complicated to define buttons, containers, pages and other UI elements in the diagram and find differences between them. Even more complicated is to model the navigation of the testing application. Using this approach the UI can be defined in a set of UI elements along with a set of UI navigation elements. This is an optimal and suitable approach in most cases. This document describes an implementation of the automated GUI tests runner framework as well. This framework is able to run the given application in test mode using the previously generated test cases. The framework collects all the information about each test case results and provides it to the tester. Future improvements: Find the... [to full text]

A number of software that supports multiple interactions such as synergistic use of speech and gesture is increasing. With power and versatility of multimodal interfaces also comes and new issues like increased complexity of software design. These issues should be considered in early design and testing stages of software development process. In this work we are researching a complex use of users multiple interactions with software. We will represent a method for the specification of multimodal user interfaces offered by other authors. This method is based on CARE properties. The CARE properties provide a formal framework for reasoning design of multimodal systems. Multimodal systems are using fusion mechanisms to combine different modalities. Later in this work we will take a closer look at testing process of multimodal applications. A new approach for testing modalities fusion mechanism in multimodal applications is offered. This approach will be experimentally demonstrated, evaluated and compared to alternative method.

We present methodology, which allows automatically test database schemes. Such methodology allows you to create new one or expand existing CASE tool. You can integrate database tests in generating physical database model process. In this way you can make scheme validation much easier, as a result it lowers the possibility of errors in early steps of modelling. Acording to our analysis, there are very few tools that covers such methodotlogy. So during our work we have designed and created a new tool – „DbTestAddin“, tool that tests database schemes. It works as Microsoft Office Visio 2003 Add-in. With this tool you can easily test any database scheme you want, compare it with a real database etc. In future, there are several updates and functional enhacements which can make this tool more attractive to use.

Daugelis testavimo procesų modelių neužtikrina visų programinės įrangos testavimo aspektų. Darbo tikslas – sumodeliuoti testavimo procesą, kuris nurodytų testavimo proceso vietą programų sistemų kūrimo gyvavimo ciklo kontekste, organizacijai, kuriančiai didelės apimties programinę įrangą. Uždaviniai: ištirti, kaip nuo programų sistemų kūrimo modelių ir metodų („Krioklio“ modelis, iteratyvus gyvavimo ciklo modelis, „V-modelis“, RAD, Scrum) priklauso testavimo proceso modeliai; Ištirti pasirinkto organizacijos tipo testavimo proceso specifiką, problematiką; Sukurti testavimo proceso modelį organizacijai, kuriančiai programinę įrangą; Įvertinti sukurtą modelį pagal testavimo proceso brandos modelį. Baigiamajame darbe remtasi literatūros analize, išnagrinėta didelės apimties PS testavimo specifika, taikytini metodai, įrankiai, geriausios praktikos. Darbo rezultatas – sukurtas ir įvertintas bei optimizuotas programų sistemų testavimo proceso modelis, kuris sujungia geriausias esamų modelių savybes ir yra lengvai plečiamas. / Many testing process models do not assure software quality in all aspects. The goal of this work is to make a testing process model showing testing process place in the software development lifecycle for the organization, developing large scope software. Objectives: check how testing process model depends on software development process model (Waterfall model, iterative processes, V-model, RAD, Scrum); Analyze testing process specifics for the chosen organization type; Create testing process model for the organization that creates software; Evaluate new model according to the testing process maturity model. Work is based on literature analysis: large scope program systems testing process specifics, applicable methods, tools and best practices are analyzed. The resulting newly created testing process model is evaluated and optimized. The model is based on the best qualities of existing models and is easily adjustable.