Autograder for C++

Overview

The program expects that there is a directory containing all the files under test in following order -

base_path/
    |---- some_path1/testbenchi.txt
    |---- some_path2/testbencho.txt
    |---- some_path3/root_folder/
                |---- roll1/
                |       |---- file1.cpp
                |       |---- file2.cpp
                |       |---- some_filename_to_match.cpp
                |---- roll2/
                |       |---- file1.cpp
                    ...

In most cases base_path could be kept an empty string.

Roll is the unique identification string of each submitter.

The autograder looks for every folder/roll in the root folder and checks for the first file whose filename contains the keyword filename_to_match, a parameter to the helper function search_cpp_files(root_folder, filename_to_match). If found, the file is tested under the provided input and output cases. Otherwise, absent is marked.

Modes

• Summary: Only shows the net number of test cases passed
• Concise: Shows whether the testcase was passed or failed
• Standard: Shows passed if test case was passed, shows line-by-line comparision of actual and expected outputs if test case was failed
• Verbose: Shows line-by-line comparision of actual and expected outputs for each test case

User Instructions

1. Create an *.txt file containing sample inputs and separate the input test cases using a line containing #testcase_input as the beginnig keyword
2. Create an *.txt file containing expected outputs and separate the test cases using a line containing #testcase_output as the beginning keyword
3. Execute the following program with proper parameters

from autogradercpp.autograder import AutograderCpp

autograder = AutograderCpp(
        base_path=BASE_PATH, 
        run_mode=RUN_MODE, 
        timeout_sec=WAIT_TIMEOUT_SEC
    )

autograder.grade_root_dir(
        ROOT_FOLDER_PATH, 
        FILENAME_TO_MATCH, 
        TEST_INPUT_FILE_PATH, 
        TEST_OUTPUT_FILE_PATH 
    )

Some Helper Functions

Below are some helper functions whose usability, I believe, can be expanded beyond just this autograder.

• autograder.py contains grade_root_dir(args*) which runs the assembled process of finding files and grading every single roll.
• file_handler.py contains read_test_cases(str) which returns the test case inputs and expected outputs into an array readable by autograder. It has to be run separately for input file and output file.
• autograder.py contains autograde(args*) which can run the testbench on one file at a time
• diff.py contains find_and_print_differences(actual_text, expected_text) which compares the texts line by line