In this example, we will use two source files and two tests, as follows:

.
├── CMakeLists.txt
├── CTestConfig.cmake
├── dashboard.cmake
├── src
│   ├── buggy.cpp
│   ├── buggy.hpp
│   └── CMakeLists.txt
└── tests
    ├── CMakeLists.txt
    ├── leaky.cpp
    └── use_after_free.cpp

The file buggy.cpp contains two buggy functions, as follows:

#include "buggy.hpp"

#include <iostream>

int function_leaky() {

  double *my_array = new double[1000];

  // do some work ...

  // we forget to deallocate the array
  // delete[] my_array;

  return 0;
}

int function_use_after_free() {

  double *another_array = new double[1000];

  // do some work ...

  // deallocate it, good!
  delete[] another_array;

  // however, we accidentally use the array
  // after it has been deallocated
  std::cout << "not sure what we get: " << another_array[123] << std::endl;

  return 0;
}

These functions are exposed in the corresponding header file (buggy.hpp):

#pragma once

int function_leaky();
int function_use_after_free();

The test source, leaky.cpp, verifies the return code from function_leaky:

#include "buggy.hpp"

int main() {
  int return_code = function_leaky();
  return return_code;
}

Correspondingly, use_after_free.cpp checks the return value of function_use_after_free, as follows:

#include "buggy.hpp"

int main() {
  int return_code = function_use_after_free();
  return return_code;
}