When running the TestCaseGenerator, the generated test cases are output to the TC folder. However, test cases from previous runs are not cleaned up. This is confusing to the user, and with a mixed set of generated test cases, there will be test coverage in ways that the user had not intended.
Expected behavior:
After running the TestCaseGenerator, the number of test cases in the output directory should be equal to the number specified in the properties file.
Actual behavior:
The test cases from previous runs of the TestCaseGenerator remain in the output directory, and are mixed with subsequent runs.
Platforms:
This bug is reproducible on Linux and Windows
Steps to reproduce:
* Build the TestCaseGenerator, supply an EFG
* Change the max number of test cases in the properties file to 4
* Run the TestCaseGenerator
* Change the man number of test cases in the properties file to 3
* Run the TestCaseGenerator
After running the TestCaseGenerator the second time, there are still 4 test cases in the output directory. This bug was discovered while repeatedly testing the TestCaseGenerator
Discovered & reported by the TestCaseGenerator team (7 members)