CheckPy

A Python tool for running tests on Python source files. Intended to be used by students whom are taking courses in the Minor Programming at the UvA.

Installation

pip install checkpy

Besides installing checkPy, you might want to download some tests along with it. Simply run checkPy with the -d arg as follows:

checkpy -d YOUR_GITHUB_TESTS_URL

Usage

usage: checkpy [-h] [-m MODULE] [-d GITHUBLINK] [-clean] [file]

checkPy: a simple python testing framework

positional arguments:
  file           name of file to be tested

optional arguments:
  -h, --help     show this help message and exit
  -m MODULE      provide a module name or path to run all tests from the
                 module, or target a module for a specific test
  -d GITHUBLINK  download tests from a Github repository and exit
  -clean         remove all tests from the tests folder and exit

To simply test a single file, call:

checkpy YOUR_FILE_NAME

If you are unsure whether multiple tests exist with the same name, you can target a specific test by specifying its module:

checkpy YOUR_FILE_NAME -m YOUR_MODULE_NAME

If you want to test all files from a module within your current working directory, then this is the command for you:

checkpy -m YOUR_MODULE_NAME

Features

• Support for ordering of tests
• Execution of tests can be made dependable on the outcome of other tests
• The test designer need not concern herself with exception handling and printing
• The full scope of Python is available when designing tests
• Full control over displayed information
• Support for importing modules without executing scripts that are not wrapped by if __name__ == "__main__"
• Support for overriding functions from imports in order to for instance prevent blocking function calls
• Support for grouping tests in modules, allowing the user to target tests from a specific module or run all tests in a module with a single command.

An example

Tests in checkPy are collections of abstract methods that you as a test designer need to implement. A test may look something like the following:

0| @t.failed(exact)
1| @t.test(1)
2| def contains(test):
3|     test.test = lambda : assertlib.contains(lib.outputOf(_fileName), "100")
4|     test.description = lambda : "contains 100 in the output"
5|     test.fail = lambda info : "the correct answer (100) cannot be found in the output"

From top to bottom:

• The decorator failed on line 0 defines a precondition. The test exact must have failed for the following tests to execute.
• The decorator test on line 1 prescribes that the following method creates a test with order number 1. Tests are executed in order, lowest first.
• The method definition on line 2 describes the name of the test (contains), and takes in an instance of Test found in test.py. This instance is provided by the decorator test on the previous line.
• On line 3 the test method is bound to a lambda which describes the test that is to be executed. In this case asserting that the print output of _fileName contains the number 100. _fileName is a magic variable that refers to the to be tested source file. Besides resulting in a boolean indicating passing or failing the test, the test method may also return a message. This message can be used in other methods to provide valuable information to the user. In this case however, no message is provided.
• On line 4 the description method is bound to a lambda which when called produces a string message describing the intent of the test.
• On line 5 the fail method is bound to a lambda. This method is used to provide information that should be shown to the user in case the test fails. The method takes in a message (info) which comes from the second returned value of the test method. This message can be used to relay information found during execution of the test to the user.

Writing tests

Test methods are discovered in checkPy by filename. If one wants to test a file foo.py, the corresponding test must be named fooTest.py. checkPy assumes that all methods in the test file are tests, as such one should not use the from ... import ... statement when importing modules.

A test minimally consists of the following:

import check.test as t
@t.test(0)
def someTest(test):
  test.test = lambda : False
  test.description = lambda : "some description"

Here the method someTest is marked as test by the decorator test. The abstract methods test and description are implemented as these are the only methods that necessarily require implementation. For more information on tests and their abstract methods you should refer to test.py. Note that besides defining the Test class and its abstract methods, test.py also provides several decorators for introducing test dependencies such as failed.

When providing a concrete implementation for the test method one should take a closer look at lib.py and assertlib.py. lib.py provides a collection of useful functions to help implement tests. Most notably getFunction and outputOf. These provide the tester with a function from the source file and the complete print output respectively. Calling getFunction makes checkPy evaluate only import statements and code inside definitions of the to be tested file. Effectively all other parts of code are wrapped by if __name__ == "__main__" and thus ignored. assertlib.py provides a collection of assertions that one may find usefull when implementing tests.

For inspiration inspect some existing collections of tests like the tests for progNS2016.

Distributing tests

CheckPy can download tests directly from Github repos. The only requirement is that a folder called tests exists within the repo that contains only tests and folders (which checkpy treats as modules). Simply call checkPy with the optional -d argument and pass your github repo url. Tests will then be automatically downloaded and installed.