C Unit Test Generator Project

Project Overview

This project demonstrates my expertise in C programming, AI-powered tools, and automated test case generation—skills essential for the position at AMD. The goal of this project is to develop a C Unit Test Generator leveraging AI techniques to automate the creation of test cases for C programs, significantly improving efficiency and coverage in testing complex systems.

The project aligns with AMD's mission to research and develop a framework for automated test case generation, particularly to address the challenges of firmware development in mission-critical systems, where manual test generation is both labor-intensive and difficult to scale.

Features

• AI-Driven Unit Test Case Generation: Uses AI models like Hugging Face and GPT to automatically generate test cases for C programs.
• CUnit Testing Framework: Incorporates the CUnit testing framework for defining and running unit tests.
• Mocks, Fakes, and Stubs: The project demonstrates the usage of mocks, fakes, and stubs in C, critical for unit testing complex firmware.
• Cross-Compatibility: Works across various C programs, adapting test cases to various codebases.
• Git Integration: Complete Git version control to track project changes and maintain a clean codebase.

Project Structure

├── CProgram
│   └── example_program.c    # Sample C program for test case generation
├── generated_tests
│   └── test_example_program.c  # AI-generated unit test cases
├── requirements.txt         # Python dependencies
├── README.md                # Project documentation
└── test_case_generator.py   # Python script for generating unit tests

Example Unit Test File

#include <CUnit/CUnit.h>
#include "../CProgram/example_program.c"

void test_function1() {
    // Test case for function1
    CU_ASSERT_EQUAL(function1(5), 10);
}

void test_function2() {
    // Test case for function2
    CU_ASSERT_STRING_EQUAL(function2("test"), "success");
}

int test_main() {
    CU_initialize_registry();
    CU_pSuite suite = CU_add_suite("Example Program Test Suite", NULL, NULL);

    CU_add_test(suite, "Test Case 1", test_function1);
    CU_add_test(suite, "Test Case 2", test_function2);

    CU_basic_run_tests();
    CU_cleanup_registry();
    return 0;
}

Technologies and Tools

• C Programming Language: Strong experience in C, including developing complex systems and firmware. Expert knowledge of unit tests, mocks, fakes, stubs, and Makefile.
• Python: Proficient in using Python for tool development and integrating AI models for test generation.
• AI Libraries (GPT): Utilizing AI techniques to simulate and automate the process of unit test generation.
• CUnit: Leveraged the CUnit testing framework to execute the generated test cases.
• Git: Extensive use of Git for version control, ensuring efficient project collaboration and code tracking.

How It Works

1. Input: The tool reads a C program (e.g., example_program.c).
2. AI Generation: Using AI models, it analyzes the program structure and generates unit test cases automatically.
3. Output: The generated test cases are saved in a new file (e.g., test_example_program.c), ready to be compiled and executed using the CUnit framework.
4. Run Tests: Execute the generated test cases by running the command:

   gcc -o test_program generated_tests/test_example_program.c -lcunit
   ./test_program

Key Achievements

• Automated Test Generation: Reduced manual test generation effort by leveraging AI, significantly improving coverage.
• Firmware Testing: Demonstrated proficiency in creating and running unit tests for mission-critical firmware.
• AI-Powered Tools: Applied Generative AI to automate tedious programming tasks, improving overall productivity.
• Project Management: Managed complex workflows using Git, ensuring efficient progress tracking and collaboration.