Zeiss, B. 2008, June 3–6, Analysis of TTCN-3 test cases. Unpublished paper presented at ETSI TTCN-3 User Conference 2008, Madrid (Spain).
Added by: Deleted user (8/20/08, 3:01 PM)
|Resource type: Conference Paper
BibTeX citation key: Zeiss
View all bibliographic details
Publisher: MTP, ETSI (Madrid (Spain))
Collection: ETSI TTCN-3 User Conference 2008
As in any other programming language, it is easily possible to create code in TTCN-3 that contains execution faults or code anomalies that remain undetected by today’s compilers and analysis tools. Some of these issues can be detected by conventional static code analysis (e.g. the detection of magic values) without the necessity of an actual test run. Other anomalies remain undetected though. If incorrect test cases are executed against the system under test, fail scenarios might occur. It must be decided in a next step whether the observed fault is caused by the system or relates to the test case. This test result analysis can cost tremendous efforts and occurs late in the test process, which hurts in particular if the fault is caused by an incorrect test case.
In this presentation, we propose to construct behavioral models of TTCN-3 test cases based on simulation and test code instrumentation in order to detect faults prior to test execution. We demonstrate how to use the resulting models for the verification of properties that directly relate to specific quality characteristics of test cases. A simple example for such a characteristic is checking the existence of an execution path where no test verdict is set. We will present additional characteristics and discuss further uses of reverse-engineered test behavioral models.
Added by: Deleted user Last edited by: Deleted user