1 `Back To Debugging`_ - `Forward To Documentation`_
3 .. _Back To Debugging: https://github.com/thehackerwithin/UofCSCBC2012/tree/master/4-Debugging/
4 .. _Forward To Documentation: https://github.com/thehackerwithin/UofCSCBC2012/tree/master/6-Documentation/
8 **Presented By Anthony Scopatz**
10 **Based on materials by Katy Huff, Rachel Slaybaugh, and Anthony Scopatz**
12 .. image:: http://s3.amazonaws.com/inscight/img/blog/evo_sol1.png
14 http://memecreator.net/the-most-interesting-man-in-the-world/showimage.php/169/I-don%27t-always-test-my-code-But-when-I-do-I-do-it-in-production.jpg
19 Software testing is a process by which one or more expected behaviors and
20 results from a piece of software are exercised and confirmed. Well chosen
21 tests will confirm expected code behavior for the extreme boundaries of the
22 input domains, output ranges, parametric combinations, and other behavioral
27 Unless you write flawless, bug-free, perfectly accurate, fully precise, and
28 predictable code every time, you must test your code in order to trust it
29 enough to answer in the affirmative to at least a few of the following questions:
31 * Does your code work?
33 * Does it do what you think it does?
34 * Does it continue to work after changes are made?
35 * Does it continue to work after system configurations or libraries are upgraded?
36 * Does it respond properly for a full range of input parameters?
37 * What about edge or corner cases?
38 * What's the limit on that input parameter?
42 *Verification* is the process of asking, "Have we built the software correctly?"
43 That is, is the code bug free, precise, accurate, and repeatable?
47 *Validation* is the process of asking, "Have we built the right software?"
48 That is, is the code designed in such a way as to produce the answers we are
49 interested in, data we want, etc.
51 Uncertainty Quantification
52 **************************
53 *Uncertainty Quantification* is the process of asking, "Given that our algorithm
54 may not be deterministic, was our execution within acceptable error bounds?" This
55 is particularly important for anything which uses random numbers, eg Monte Carlo methods.
60 Say we have an averaging function:
62 .. code-block:: python
69 Tests could be implemented as runtime exceptions in the function:
71 .. code-block:: python
78 print "The number list was not a list of numbers."
80 print "There was a problem evaluating the number list."
84 Sometimes tests they are functions alongside the function definitions they are testing.
86 .. code-block:: python
93 print "The number list was not a list of numbers."
95 print "There was a problem evaluating the number list."
100 assert mean([0, 0, 0, 0]) == 0
101 assert mean([0, 200]) == 100
102 assert mean([0, -200]) == -100
103 assert mean([0]) == 0
106 def test_floating_mean():
107 assert mean([1, 2]) == 1.5
109 Sometimes they are in an executable independent of the main executable.
111 .. code-block:: python
116 length = len(numlist)
118 print "The number list was not a list of numbers."
120 print "There was a problem evaluating the number list."
124 Where, in a different file exists a test module:
126 .. code-block:: python
131 assert mean([0, 0, 0, 0]) == 0
132 assert mean([0, 200]) == 100
133 assert mean([0, -200]) == -100
134 assert mean([0]) == 0
137 def test_floating_mean():
138 assert mean([1, 2]) == 1.5
142 The three right answers are:
148 The longer answer is that testing either before or after your software
149 is written will improve your code, but testing after your program is used for
150 something important is too late.
152 If we have a robust set of tests, we can run them before adding something new and after
153 adding something new. If the tests give the same results (as appropriate), we can have
154 some assurance that we didn'treak anything. The same idea applies to making changes in
155 your system configuration, updating support codes, etc.
157 Another important feature of testing is that it helps you remember what all the parts
158 of your code do. If you are working on a large project over three years and you end up
159 with 200 classes, it may be hard to remember what the widget class does in detail. If
160 you have a test that checks all of the widget's functionality, you can look at the test
161 to remember what it's supposed to do.
165 In a collaborative coding environment, where many developers contribute to the same code base,
166 developers should be responsible individually for testing the functions they create and
167 collectively for testing the code as a whole.
169 Professionals often test their code, and take pride in test coverage, the percent
170 of their functions that they feel confident are comprehensively tested.
172 How are tests written?
173 ======================
174 The type of tests that are written is determined by the testing framework you adopt.
175 Don't worry, there are a lot of choices.
179 **Exceptions:** Exceptions can be thought of as type of runttime test. They alert
180 the user to exceptional behavior in the code. Often, exceptions are related to
181 functions that depend on input that is unknown at compile time. Checks that occur
182 within the code to handle exceptional behavior that results from this type of input
183 are called Exceptions.
185 **Unit Tests:** Unit tests are a type of test which test the fundametal units of a
186 program's functionality. Often, this is on the class or function level of detail.
187 However what defines a *code unit* is not formally defined.
189 To test functions and classes, the interfaces (API) - rather than the implmentation - should
190 be tested. Treating the implementation as a ack box, we can probe the expected behavior
191 with boundary cases for the inputs.
193 **System Tests:** System level tests are intended to test the code as a whole. As opposed
194 to unit tests, system tests ask for the behavior as a whole. This sort of testing involves
195 comparison with other validated codes, analytical solutions, etc.
197 **Regression Tests:** A regression test ensures that new code does change anything.
198 If you change the default answer, for example, or add a new question, you'll need to
199 make sure that missing entries are still found and fixed.
201 **Integration Tests:** Integration tests query the ability of the code to integrate
202 well with the system configuration and third party libraries and modules. This type
203 of test is essential for codes that depend on libraries which might be updated
204 independently of your code or when your code might be used by a number of users
205 who may have various versions of libraries.
207 **Test Suites:** Putting a series of unit tests into a collection of modules creates,
208 a test suite. Typically the suite as a whole is executed (rather than each test individually)
209 when verifying that the code base still functions after changes have been made.
213 **Behavior:** The behavior you want to test. For example, you might want to test the fun()
216 **Expected Result:** This might be a single number, a range of numbers, a new fully defined
217 object, a system state, an exception, etc. When we run the fun() function, we expect to
218 generate some fun. If we don't generate any fun, the fun() function should fail its test.
219 Alternatively, if it does create some fun, the fun() function should pass this test.
220 The the expected result should known *a priori*. For numerical functions, this is
221 result is ideally analytically determined even if the fucntion being tested isn't.
223 **Assertions:** Require that some conditional be true. If the conditional is false,
226 **Fixtures:** Sometimes you have to do some legwork to create the objects that are
227 necessary to run one or many tests. These objects are called fixtures as they are not really
228 part of the test themselves but rather involve getting the computer into the appropriate state.
230 For example, since fun varies a lot between people, the fun() function is a method of
231 the Person class. In order to check the fun function, then, we need to create an appropriate
232 Person object on which to run fun().
234 **Setup and teardown:** Creating fixtures is often done in a call to a setup function.
235 Deleting them and other cleanup is done in a teardown function.
237 **The Big Picture:** Putting all this together, the testing algorithm is often:
239 .. code-block:: python
246 But, sometimes it's the case that your tests change the fixtures. If so, it's better
247 for the setup() and teardown() functions to occur on either side of each test. In
248 that case, the testing algorithm should be:
250 .. code-block:: python
264 ----------------------------------------------------------
266 Nose: A Python Testing Framework
267 ================================
268 The testing framework we'll discuss today is called nose. However, there are several
269 other testing frameworks available in most language. Most notably there is `JUnit`_
270 in Java which can arguably attributed to inventing the testing framework.
272 .. _nose: http://readthedocs.org/docs/nose/en/latest/
273 .. _JUnit: http://www.junit.org/
275 Where do nose tests live?
276 *************************
277 Nose tests are files that begin with ``Test-``, ``Test_``, ``test-``, or ``test_``.
278 Specifically, these satisfy the testMatch regular expression ``[Tt]est[-_]``.
279 (You can also teach nose to find tests by declaring them in the unittest.TestCase
280 subclasses chat you create in your code. You can also create test functions which
281 are not unittest.TestCase subclasses if they are named with the configured
282 testMatch regular expression.)
286 To write a nose test, we make assertions.
288 .. code-block:: python
290 assert should_be_true()
291 assert not should_not_be_true()
293 Additionally, nose itself defines number of assert functions which can be used to
294 test more specific aspects of the code base.
296 .. code-block:: python
298 from nose.tools import *
301 assert_almost_equal(a, b)
304 assert_raises(exception, func, *args, **kwargs)
305 assert_is_instance(a, b)
308 Moreover, numpy offers similar testing functions for arrays:
310 .. code-block:: python
312 from numpy.testing import *
314 assert_array_equal(a, b)
315 assert_array_almost_equal(a, b)
318 Exersize: Writing tests for mean()
319 **********************************
320 There are a few tests for the mean() function that we listed in this lesson.
321 What are some tests that should fail? Add at least three test cases to this set.
322 Edit the ``test_mean.py`` file which tests the mean() function in ``mean.py``.
324 *Hint:* Think about what form your input could take and what you should do to handle it.
325 Also, think about the type of the elements in the list. What should be done if you pass
326 a list of integers? What if you pass a list of strings?
330 nosetests test_mean.py
332 Test Driven Development
333 =======================
334 Test driven development (TDD) is a philosophy whereby the developer creates code by
335 **writing the tests fist**. That is to say you write the tests *before* writing the
338 This is an iterative process whereby you write a test then write the minimum amount
339 code to make the test pass. If a new feature is needed, another test is written and
340 the code is expanded to meet this new use case. This continues until the code does
343 TDD operates on the YAGNI principle (You Ain't Gonna Need It). People who diligently
344 follow TDD swear by its effectiveness. This development style was put forth most
345 strongly by `Kent Beck in 2002`_.
347 .. _Kent Beck in 2002: http://www.amazon.com/Test-Driven-Development-By-Example/dp/0321146530
351 Say you want to write a fib() function which generates values of the
352 Fibinacci sequence fof given indexes. You would - of course - start
353 by writing the test, possibly testing a single value:
355 .. code-block:: python
357 from nose import assert_equal
364 assert_equal(obs, exp)
366 You would *then* go ahead and write the actual function:
368 .. code-block:: python
371 # you snarky so-and-so
374 And that is it right?! Well, not quite. This implementation fails for
375 most other values. Adding tests we see that:
377 .. code-block:: python
382 assert_equal(obs, exp)
388 assert_equal(obs, exp)
392 assert_equal(obs, exp)
394 This extra test now requires that we bother to implement at least the intial values:
396 .. code-block:: python
404 However, this function still falls over for ``2 < n``. Time for more tests!
406 .. code-block:: python
411 assert_equal(obs, exp)
417 assert_equal(obs, exp)
421 assert_equal(obs, exp)
427 assert_equal(obs, exp)
431 assert_equal(obs, exp)
433 At this point, we had better go ahead and try do the right thing...
435 .. code-block:: python
442 return fib(n - 1) + fib(n - 2)
444 Here it becomes very tempting to take an extended coffee break or possibly a
445 power lunch. But then you remember those pesky negative numbers and floats.
446 Perhaps the right thing to do here is to just be undefined.
448 .. code-block:: python
453 assert_equal(obs, exp)
459 assert_equal(obs, exp)
463 assert_equal(obs, exp)
469 assert_equal(obs, exp)
473 assert_equal(obs, exp)
479 assert_equal(obs, exp)
483 assert_equal(obs, exp)
485 This means that it is time to add the appropriate case to the funtion itself:
487 .. code-block:: python
490 # sequence and you shall find
491 if n < 0 or int(n) != n:
492 return NotImplemented
493 elif n == 0 or n == 1:
496 return fib(n - 1) + fib(n - 2)
498 And thus - finally - we have a robust function together with working tests!