[Dynamic Language] Python Library: unittest

时间：2010-10-12 来源：※ABeen※

Python Library: unittest

对于熟悉 NUnit 的人来说，Python unittest 可以很快上手。

TestCase, FunctionTestCase: 创建包含测试方法的测试用例。
TestSuite: 组合多个测试用例，用于批量测试。
TestLoader: 可通过多种手段自动查找测试用例，返回 TestSuite。
TestRunner, TextTestRunner: 执行测试用例，并输出测试结果。
TestResult: 保存测试结果信息。

简单实例

abeen@localhost:~/learn_test/unittest$ cat test.py
#! /usr/bin/env python
# _*_ coding: utf-8 _*_

import sys
import unittest
from unittest import TestCase, FunctionTestCase, TestSuite, TextTestRunner, defaultTestLoader

class MyClass(object):
    def __init__(self, x, y):
        self.x = x
        self.y = y

    def func1(self):
        return self.x + self.y

    def func2(self):
        return self.x - self.y


class TestMyClass(TestCase):
    def setUp(self):
        print "setUp", hex(id(self))
        self.instance = MyClass(1,2)

    def tearDown(self):
        print "tearDown"

    def test_func1(self):
        self.assertEqual(self.instance.func1(), 3)

    def test_func2(self):
        self.assertEqual(self.instance.func2(), -1)


if __name__ == "__main__":
    unittest.main()

运行查看结果

abeen@localhost:~/learn_test/unittest$ python test.py
EsetUp 0x7f752e9731d0
tearDown
.setUp 0x7f752e973250
tearDown
.
----------------------------------------------------------------------
Ran 3 tests in 0.000s
FAILED (errors=0)

(1) 默认测试对象(TestMyClass)中以 "test" 开头的都被认为是测试方法，相当于 NUnit 当中添加 "[Test]" 特性的方法。
(2) setUp / tearDown 在每个测试函数执行前后被调用。
(3) 测试对象不会重复使用，每次都创建了新实例。

用suite来组织一个有前后顺序依赖的完整测试逻辑。

abeen@localhost:~/learn_test/unittest$ cat test.py
...
if __name__ == "__main__":
    """unittest.main()"""
    suite = TestSuite()
    suite.addTest(TestMyClass("test_func1"))
    suite.addTest(TestMyClass("test_func2"))

    TextTestRunner().run(suite)

运行测试

abeen@localhost:~/learn_test/unittest$ python test.py
setUp 0x7f15a0ac1cd0
tearDown
.setUp 0x7f15a0ac1d10
tearDown
.
----------------------------------------------------------------------
Ran 2 tests in 0.000s

OK

还有更简便的方法创建Suite, TestLoader提供了多种方法“自动、批量”完成这类工作....


if __name__ == "__main__":
    """unittest.main()"""
    """
    suite = TestSuite()
    suite.addTest(TestMyClass("test_func1"))
    suite.addTest(TestMyClass("test_func2"))

    TextTestRunner().run(suite)
    """

    suite = defaultTestLoader.loadTestsFromTestCase(TestMyClass)
    TextTestRunner().run(suite)

运行测试

abeen@localhost:~/learn_test/unittest$ python test.py
setUp 0x7f3d073cecd0
tearDown
.setUp 0x7f3d073ced50
tearDown
.
----------------------------------------------------------------------
Ran 2 tests in 0.000s

OK

我们自己该造一个更好用的方法来试试看

abeen@localhost:~/learn_test/unittest$ cat test.py
#! /usr/bin/env python
# _*_ coding: utf-8 _*_

import sys
import unittest
from unittest import TestCase, FunctionTestCase, TestSuite, TextTestRunner, defaultTestLoader

class MyClass(object):
    def __init__(self, x, y):
        self.x = x
        self.y = y

    def func1(self):
        return self.x + self.y

    def func2(self):
        return self.x - self.y

def MyFunc(x, y):
    return x + y

"""自己定义测试类"""
class MyFunctionTestCase(FunctionTestCase):
    def _func(self, name):
        attr_name = "_{0}__{1}Func".format(self.__class__.__base__.__name__, name)
        return getattr(self, attr_name)

    def setUp(self):
        func = self._func("setUp")
        if func is not None: func(self)

    def tearDown(self):
        func = self._func("tearDown")
        if func is not None: func(self)

    def runTest(self):
        func = self._func("test")
        func(self)

if __name__ == "__main__":
    def setUp(self):
        self.o = MyClass(1, 2)

    test1 = MyFunctionTestCase(lambda self: self.assertEqual(MyFunc(1,2), 3), None)
    test2 = MyFunctionTestCase(lambda self: self.assertEqual(self.o.func1(), 3), setUp)
    test3 = MyFunctionTestCase(lambda self: self.assertEqual(self.o.func2(), -1), setUp)

    suite = TestSuite()
    suite.addTests((test1, test2))

    TextTestRunner().run(suite)

测试结果

abeen@localhost:~/learn_test/unittest$ python test.py
..
----------------------------------------------------------------------
Ran 2 tests in 0.000s

OK