A simple tool to stress test competitive programming solutions (or anything really).
In each iteration, the generator file is run to generate a random test case. This is passed as input to the bad (target/interesting/suspicious) file, and the test case and the output of the bad file are passed to the checker file. If there is a single valid output, then the checker file can use a "good" file to directly compare the outputs. If there are several possible valid outputs, the checker is used to judge the output.
In main.py, set the appropriate file paths and run it. Sample good, bad and checker files have been included.
C++ is supported. Python and any scripting languages can be used with a shebang and executable permissions.
Linux. Might work with Windows.