Disclaimer

These tests don't test everything, namely, coding-style(see course guidelines), memory leaks(google valgrind), compilation without warnings/errors, proper project structure, etc - so ensure your project passes the presubmit tests first.

These tests might have bugs, e.g generate valid students as invalid, or treat cases that can be assumed not to occur, if so, see please report these issues via Github

Installation

The tests do not support Windows, and haven't been tested on OSX.

The tests are ran via python3 (version 3.7 or higher), and require the following libraries:

• hypothesis
• pytest

These are already installed by default in the Aquarium PCs, so if you're working there, you may skip this section and go to Configuration

If your personal PC is using Linux, and has python 3.7 or higher, you should install the libraries via the following command. python3 -m pip install --user pytest hypothesis. On some systems you may need to replace python3 with python, but either way, make sure your version is 3.7 or higher (via python3 --version or python --version)

If you have difficulties with this step, try running the tests on the Aquarium PCs.

Configuration

You need to place the random_tester.py within your project's root directory, or up to 1 folder inside it. for example, if your project's directory is named 'ex1', the following would work:

• ex1/
• ex1/CMakeLists.txt
• ex1/manageStudents.c
• ex1/cmake-build-debug/manageStudents
• ex1/random_tester.py

Before running the tests, make sure you've compiled manageStudents, and make sure there aren't multiple 'manageStudents' executables near the test folder or in the parent folder, otherwise the wrong executable might be tested.

Running

You can then run the tests via terminal, as follows:

python3 -m pytest -vv random_tester.py

If you wish to see debug information, such as which students were inserted, type:

python3 -m pytest -vv random_tester.py -s

Alternatively you may run the tester via PyCharm: you'll need to add a configuration under Run > Edit Configurations... > + > Python tests > pytest, then type random_tester under Target: module name. (This still requires pytest and hypothesis to be installed, if you're doing this on your personal PC)

How do the tests work, what is being tested

The tester uses the 'hypothesis' library, which can generate thousands of test cases, by randomly changing various parameters. This allows the tests to cover a lot of scenarios. Some of the things that are being tested are:

• Being able to input valid students without errors

• Finding the best student

• Merge-sort, which should be a stable sort

• Quick-sort

• Printing an ERROR message when given an invalid student

Some of the ways invalid students are generated:

• Invalid characters in character/integer fields

• Integers out of bounds/negative

• Missing fields

Problems with the tests and feedback

If you find any bugs with the tests (e.g, valid students are generated as part of invalid tests), please post the issue here