Search

Copyright of Revista Facultad de Ingeniería Universidad de Antioquia is the property of Universidad de Antioquia and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Software testing focuses on detecting defects or failures during code execution. Testing is a challenging creative task, requiring automation. The adoption of good practices and testing strategies contributes to increasing the efficiency of software development companies. MTest.search model for automatic unit test generation has defined domain model extension, test, and execution mechanisms. In this work, mechanisms to extend the search-based reduction model are presented. The proposed extensions take into account the objects and sets involved in the source code, and enhance the detection of defects or failures based on the significance of the values and paths / scenarios involved in the test. To validate the proposal, three case studies were defined using classical methods and real projects. [ABSTRACT FROM AUTHOR]

This paper presents a proposal for assigning priorities to test cases in agile development environments. Through mathematical methods, a function is obtained to calculate the priority of the test cases of the software projects and their relationship with the phases of the project development life cycle. The defined function is based on four indicators whose information is possible to obtain in each of the project's iterations. A case study is also presented that allows comparing the results of ordering of the test cases, according to the priority with their effectiveness for the detection of potential errors and get expected result. [ABSTRACT FROM AUTHOR]

The test phase is a difficult process that consumes a large percentage of the cost in terms of the time of the process of software development. The purpose of the tests is to determine if the developed products meet the requirements accord with the users and customers on specifications. Therefore, the processes, methods and tools that allow obtain good sets of test of a system are needed. This article introduces a component that automatically combines values to perform unit tests and for that were applied metaheuristics algorithms. The proposed solution has been tested in a case study, and was compared with values obtained through other algorithms proposed by authors who work the issue in the scientific community. The component allows to obtain a reduced set of test values, with less time of execution and a coverage of 100%. [ABSTRACT FROM AUTHOR]

In models and standard international existence spite that allow to evaluate the quality that it possess the developed software products, this it constitutes an aspect that continues being a problem for many organizations that develop software. The software tests possess great importance inside the product development process, due to the time and effort that it bears to carry out them. To contribute the time and effort that it is used for the tests realization to diminish it is had tools that automate in some measure the evaluation process. However, a big number of the tools are de-signed to evaluate traditional software systems, obviating those smaller components that conform these systems. This work is centered in the creation of a plugin to carry out performance tests to software components using aspect-oriented programming in Java, given the necessity to automate this system test type that it lose protagonism before other tests types in many occasions like they could be the functional tests. [ABSTRACT FROM AUTHOR]

The process of finding and processing information is very important for many professionals. Therefore, information retrieving and sharing are key issues. At present, the quantity of information is growing and it is difficult to find it, due to its heterogeneity, its constant change and the diversity of data sources. Technological observatories solve this problem in some way, although they do not always take into account the necessity to make the monitoring process proactively. This paper shows how the inclusion of proactive behaviors into a technological observatory using the agent's philosophy and the implementation patterns can improve its performance and keep users updated if their interests are known. The test pilot results are described, where the users received mails according to their interests. [ABSTRACT FROM AUTHOR]