Earticle

철저한 단위테스트를 통해 소프트웨어 단위부품의 품질을 조기에 확보하여야 통합테스트 기간을 단축하고 프로젝트 비용을 절감하여 궁극적으로 정해진 기간 내에 시스템의 구축을 완료할 수 있게 된 다. 그러나 업무 지원 응용 시스템은 대부분이 데이터베이스를 사용하므로 그렇지 않은 소프트웨어보다 단위테스트를 수행하기가 훨씬 더 복잡하고 어려워 단위부품의 품질 확인이 부실해지거나 지연됨으로써 개발에 차질을 초래하는 경우가 많다. 지금까지 테스트에 대한 많은 연구가 있었지만 데이터베이스를 사용하는 대규모 응용시스템을 테스트하기 위한 연구는 상대적으로 미약하였다. 이는 데이터베이스로 인하여 증폭된 복잡성이문제에 대한 분석과 해결을 어렵게 한 것이 원인으로 보인다. 참조무결성(referential integrity)을 사용하는 경우 테이블에 데이터를 등록하거나 삭제하려 할 때 등록을 못하거나 삭제를 못하는 문제가 발생한다. 이러한 문제를 해결하기 위해 여분의 데이터를 작성하면 등록이나 삭제가 가능해질 수 있다. 본 논문에서는 이 데이터를 수작업으로 작성하는 것이 아니라 참조관계를 이용하여 자동으로 생성하는 알고리즘에 대해 연구한다. 연구의 결과는 테스트 데이터 작성에 드는 노력을 줄여 데이터베이스 응용시스템의 단위테스트를 원활하게 수행할 수 있는 자동화 도구를 구축하는 기반이 될 것으로 기대한다.

Careful cleaning and disinfection of pigpens is essential to prevent disease spread and avoid the resultant economic losses. Hygiene in pigpens is generally evaluated by visual monitoring supplemented with bacteriological monitoring, which includes counting the total aerobic bacteria (TAB) and/or fecal indicator bacteria (FIB). However, these methods present drawbacks such as time and labor requirements. As adenosine triphosphate (ATP) is ubiquitous in all living organisms including microorganisms, this study aimed to directly compare the results of microbial assessment and ATP quantification, and to suggest possible detailed application methods of the ATP test for hygiene evaluation in pigpens of a farrowing unit. Before and after standard cleaning procedures, samples were collected from the floor corner, floor center, and feeding trough of four pigpens at different time points. No FIB were detected and both the TAB and ATP levels were significantly decreased in the floor center area after cleaning. FIB were continuously detected after cleaning and disinfection of the floor corners, and there was no significant ATP level reduction. The feeding trough did not show any significant difference in these values before and after cleaning, indicating insufficient cleaning of this area. The levels of TAB and ATP after cleaning were significantly correlated and the average ATP value was significantly lower in the absence of FIB than in their presence. In the absence of standard references, a more thorough hygiene management could be achieved evenly by supplementing cleaning or disinfection based on the lowest ATP results obtained at the cleanest test site, which in the present study was the floor center. Overall, these results indicate that the on-farm ATP test can be used to determine the cleanliness status, in addition to visual inspection, as an alternative to laboratory culture-based testing for the presence of microorganisms.

소프트웨어 개발을 성공적으로 수행하기 위하여서 유닛 테스팅은 아주 필수 불가결한 요소이다. Python, Java, C# 등의 현대 컴퓨터 개발 언어에서는 유닛 테스팅을 용이하게 하기 위하여 다양한 기능을 제공하고 있다. 하지만, C++ 언어에 있어서는 워낙 많은 프레임워크를 제공하고 있는 관계로, 유닛 테스팅을 위한 프레임워크 선택이 결코 용이하지않다. 이 논문에서는 C++ 언어에서의 유닛 테스팅을 위한 프레임워크를 집약된 test runner와 분리된 test runner의 두그룹으로 나눠서 고찰하였다.

Unit testing is proved to be vital for a successful software development process. Modern languages, such as Python, Java and C#, have a great support and tools for unit testing. But when it comes to C++, there are a big number of C++ frameworks available [List], and it becomes hard to make a choice of unit testing framework to use. This paper presents a survey of open source C++ unit testing frameworks by dividing open source C++ unit testing frameworks into two groups: frameworks with an integrated test runner and frameworks with a separate test runner.

The popularity of Agile Development has been increasing over the last several years. Many Agile techniques include unit testing as a basic activity. Unit testing has two characteristics: it is performed by programmers, and it is normally based on source code structures. However, the problem arises that programmers are unlikely to design structure-based tests. Test design process draws graphs that represent structures; and applies coverage criteria these graphs. The coverage criteria concept is very important in unit testing. Support tools are required, especially by programmers performing unit testing. However, the existing tools do not cover all the steps of the process. This study develops a method that fills the gap between the graph generation and the test design step. This paper We analyzes the extent to which the method decreases the effort required for unit testing, and show that the method effectively reduces the effort. The effort is measured in ManMonths.

Software visualization techniques fall into two categories: visualization of software component relationships and visualization of software metrics. In this paper, we propose a new hybrid method based on both categories. The proposed method visualizes the coincidence between specification and implementation from two aspects: static checking and ordinal testing by test suites. In our method, each ratio of the coincidence is shown by pie charts which represent classes of the target software. The whole software is represented in a weighted digraph structure. We have prototyped a tool to implement our proposed method. We have evaluated the utility of the proposed method by applying the tool to two kinds of software: a warehouse management program and a telephone directory management program. We conclude that the proposed method yields informative results.

ATTOL is a software unit testing tool produced by the ATTOL Testware SA in France. It automates the entire software unit testing process: test plan template and test program generation, test program execution, test result analysis and test report generation. ATTOL is suited for the development of embedded software as it allows programmers to operate in native and cross development environments. Particularly, it is used for the development of the flight software which is embedded in the Communication Ocean Meteorological Satellite (COMS). As the flight software is mission-critical, it requires the strict software quality and high testing constraints. The flight software of COMS is verified by ATTOL in native and cross platforms. In 2002, ATTOL was taken over by the IDM Rational Software and has been supplied with the name of Test RealTime (TestRT). The test process of TestRT becomes different from that of ATTOL as TestRT provides the new functionalities that were absent from ATTOL. TestRT provides the new features in the test script language, as well. In this paper, we compare the test process of ATTOL to TestRT With an example of COMS and explain what has been changed in the test script language.

Phasor Measurement Unit (PMU) devices, based on the synchronized measurement technology, are basic building blocks of Wide Area Monitoring System (WAMS) and Wide Area Monitoring Analysis Protection-control (WAMAP). Similar to the Remote Terminal Unit (RTU) in Energy Management System (EMS), PMU is becoming very important in the WAMS or WAMAP. This paper introduces the architecture of PMU testing and evaluation (T&E) in our laboratory. The integrated evaluation schemes include testing environment, performance testing method, conformance testing method and evaluation.

We performed an importance analysis of In-Service Testing (157) components for Ulchin Unit 3 using the integrated evaluation method for categorizing component safety significance developed in this study. The developed method is basically aimed at having a PSA expert perform an importance analysis using PSA and its related information. The importance analysis using the developed method is initiated by ranking the component importance using quantitative PSA information. The importance analysis of the IST components not modeled in the PSA is performed through the engineering judgment, based on the expertise of PSA, and the quantitative and qualitative information for the 157 components. The PSA scope for importance analysis includes not only Level 1 and 2 internal PSA but also Level 1 external and shutdown/low power operation PSA. The importance analysis results of valves show that 167 (26.55%) of the 629 IST valves are HSSCs and 462 (73.45%) are LSSCs. Those of pumps also show that 28 (70%)of the 40157 pumps are HSSCs and 12 (30%) are LSSCs.

An ARIMA signal observed with measurement error is shown to have another ARIMA representation with nonlinear restrictions on parameters. For this model, the restricted Newton-Raphson estimator(RNRE) of the unit root is shown to have the same limiting distribution as the ordinary least squares estimator of the unit root in an AR(1) model tabulated by Dickey and Fuller (1979). The RNRE of parameters of the ARIMA(p,1,k) process and unit root tests base on the RNRE are developed.

This paper describes a design and temporal analysis of a hardware-in-the-loop (HIL) simulation environment for testing a motor control unit (MCU). The design concepts and main characteristics including unavoidable time delays of each component module are described. From temporal analysis results according to the module integration method, an appropriate solution is proposed to fix and minimize time delays. In order to verify the effectiveness of the proposed solution, the HIL test results are compared with the results of experiments and an offline simulation.

Careful cleaning and disinfection of pigpens is essential to prevent disease spread and avoid the resultant economic losses. Hygiene in pigpens is generally evaluated by visual monitoring supplemented with bacteriological monitoring, which includes counting the total aerobic bacteria (TAB) and/or fecal indicator bacteria (FIB). However, these methods present drawbacks such as time and labor requirements. As adenosine triphosphate (ATP) is ubiquitous in all living organisms including microorganisms, this study aimed to directly compare the results of microbial assessment and ATP quantification, and to suggest possible detailed application methods of the ATP test for hygiene evaluation in pigpens of a farrowing unit. Before and after standard cleaning procedures, samples were collected from the floor corner, floor center, and feeding trough of four pigpens at different time points. No FIB were detected and both the TAB and ATP levels were significantly decreased in the floor center area after cleaning. FIB were continuously detected after cleaning and disinfection of the floor corners, and there was no significant ATP level reduction. The feeding trough did not show any significant difference in these values before and after cleaning, indicating insufficient cleaning of this area. The levels of TAB and ATP after cleaning were significantly correlated and the average ATP value was significantly lower in the absence of FIB than in their presence. In the absence of standard references, a more thorough hygiene management could be achieved evenly by supplementing cleaning or disinfection based on the lowest ATP results obtained at the cleanest test site, which in the present study was the floor center. Overall, these results indicate that the on-farm ATP test can be used to determine the cleanliness status, in addition to visual inspection, as an alternative to laboratory culture-based testing for the presence of microorganisms.

Overuse of laboratory and X-ray testing is common in the intensive care unit (ICU). This review highlights focused strategies for critical care clinicians as outlined by the Critical Care Societies Collaborative (CCSC) as part of the American Board of Internal Medicine Foundation’s Choosing Wisely® campaign. The campaign aims to promote the use of judicious testing and decrease unnecessary treatment measures in the ICU. The CCSC outlines five specific recommendations for reducing unnecessary testing in the ICU. First, reduce the use of daily or regular interval diagnostic testing. Second, do not transfuse red blood cells in hemodynamically stable, non-bleeding ICU patients with a hemoglobin concentration greater than 7 mg/dl. Third, do not use parenteral nutrition in adequately nourished critically ill patients within the first 7 days of ICU stay. Fourth, do not deeply sedate mechanically ventilated patients without a specific indication and without daily attempts to lighten sedation. Finally, do not continue life support for patients at high risk of death without offering patients and their families the alternative of comfort focused care. A number of strategies can be used to reduce unnecessary testing in the ICU, including educational campaigns, audit and feedback, and implementing prompts in the electronic ordering system to allow only acceptable indications when ordering routine testing. Greater awareness of the lack of outcome benefit and associated costs can prompt clinicians to be more mindful of ordering tests and procedures in order to reduce unnecessary testing in the ICU.

This paper considers the size distortion problem when testing the hypothesis on the cointegrating vector. The test on the cointegrating vector tends to be erroneously rejected. We propose constructing a valid confidence interval by application of the Bonferroni’s inequality, and compare the performance to another alternative procedure.

최근 사물인터넷 서비스가 점차 보편화되면서 사물인터넷에 사용되는 임베디드 시스템의 신뢰성 및 안전성 이슈가 부각되고 있다. 사물인터넷 시스템 등과 같은 임베디드 시스템 개발의 경우, 개발과정에서 소프트웨어 가상 수행환경을 구축하는 것이 쉽지 않아 단위테스트를 거의 수행하지 않고 있어 기능 신뢰성을 보장하기 어려운 실정이다. 이 연구에서는 실제 플랜트 환경에서 로그 기반 단위테스트 수행방법을 제안한다. 실제 시나리오 수행과정에서 함수별 수행경로를 고려하여 중복된 수행로그를 제거하여 저장되는 로그를 최소화하고 나중에 대표 수행경로만을 분석하여 단위테스트를 진행하는 방식이다. 제안된 방식은 시스템 테스트를 진행하는 과정에서 로그 기반 단위테스트를 효율적으로 수행할 수 있는 장점이 있다.

As Internet of Things (IoT) is recently serviced in several fields, the reliability and safety issues for IoT embedded systems are emerged. During the development of embedded systems, it is not easy to build the virtual execution environment and to test the developing version. Therefore, it is difficult to ensure its reliability due to lack of unit testing. In this paper, we propose a log-based unit testing framework for embedded software, which performs on real target board by extracting information of function execution. And, according to execution paths, duplicated logs are eliminated to keep a minimal log size. As a result, during system testing, testers can efficiently decide whether the executed paths of each function are correctly performed or not.

최근 세계 자동차 업계는 AUTOSAR 소프트웨어 플랫폼 표준을 제정하여 소프트웨어의 신뢰성과 재사용성을 향상키고 있다. 그러나 테스트 시스템 및 테스트 프로세스의 신뢰성 확보와 테스트에 소요되는 기간 단축을 위하여 테스트의 표준화 및 자동화가 절실하다. 특히 RTE 및 베이직 소프트웨어 모듈은 표준화를 만족하는 설계 도구를 사용함으로써 최소한의 신뢰성이 보장되지만, 여전히 응용 소프트웨어 컴포넌트에 대해서는 철저한 테스트가 필요하다. 본 논문에서는 표준화된 AUTOSAR 설계 정보를 이용하여 소프트웨어 컴포넌트의 유닛 테스트를 위한 테스트 시스템의 생성 및 테스트 케이스 생성 자동화 방법을 제안하였으며, 테스트 시스템은 TTCN-3 테스트 표준을 적용하였다. 본 논문에서 제안한 방법을 적용함으로써 차량 소프트웨어의 개발 기간 단축 및 신뢰성을 향상시킬 수 있었다.

AUTOSAR, a standard software platform for automotive, has been developed to manage software complexity and improve software reuseability. However Automated and standardized testing is needed to improve reliability and reduce time and effort on testing. Fundamental functionality of AUTOSAR RTE and basic software modules is guaranteed by using an AUTOSAR tool, but application software components have to be tested thoroughly. In this paper, we suggest a test system for the AUTOSAR software component using TTCN-3, a standardized testing language. Test execution system and test cases for the software component are generated automatically from AUTOSAR XML containing software design information. With the proposed testing techniques we can reduce time and effort to build the testing system.

내장형 시스템의 요구사항이 증가함에 따라 내장형 시스템 소프트웨어의 복잡도가 증가하고 있다. 그러므로 내장형 시스템 소프트웨어 테스팅의 필요성이 높아지고 있다. 본 논문은 내장형 시스템 소프트웨어의 테스트를 효과적으로 실행하기 위해 교차 개발 환경에 적합한 테스팅 도구를 제안한다. 본 도구는 호스트/타겟 구조로 이루어져있으며 호스트에서 작성한 테스트 케이스를 타겟에서 실행할 수 있는 직관적인 환경을 제공한다. 그리고 제안하는 도구의 테스트 케이스는 XML 기반의 테스트 스크립트를 이용한다. 또한, 직관적인 테스트 케이스를 작성하기 위해 트리와 테이블 기반의 테스트 스크립트 편집기를 갖는다. 제안하는 테스트 도구는 테스트 케이스 작성에서부터 결과확인까지 직관적인 GUI를 제공하여 테스트 케이스 작성에 대한 부담을 경감시키는 장점을 가진다.

According to increasing requirments in embedded systems, embedded software has been more complicated then before, a software developer is required to test her/his software to make a efficient embedded system software in both time and space. This paper suggests a testing tool with which a software developer can easily test the embedded system software in cross-development environments. The suggested tool is designed based on host/target architecture, to provide an intuitive test environment in which a test case can be executed in a target board. The tool uses an XML-based test script to generate an appropriate test case. It includes a tree-based test script editor with which a developer can easily make a test case. Therefore, with the suggested tool, a develop can put down a burden on an software testing and get more productivity in software development related on embedded system.