So far in the course, CodeRunner questions have asked you to write a program or function, which we then test. But have you thought about the code we write to test your answer? In this question, we turn things around by providing you with a set of (hidden) answers to some question and we ask you to write the tests.
Suppose that you are a teacher and you've set the following problem for your students:
Write a boolean-valued function is_old_enough(age) which returns True if the given age value is 18 or more. Otherwise the function should return False. You may assume that age is an int. The return value must be a boolean, not the strings "True" or "False".
How would you test the code that the student submits? That's what you're going to do in this question!
You have to write the tests for this function in the answer box below. Each test should consist of a bit of code that should be run together with the expected output ('Result'). Obviously your test code should call the function is_old_enough which we supply. Do not include an is_old_enough function in your test.
A few sample values have been pre-filled into the answer box. In row 1 you should enter the expected result for the given test. In rows 2 through 4 you should insert tests that will produce the given result. Then consider what additional tests you might need to make.
We will firstly run your tests with a correct answer to the question. All your tests should pass in this case, that is the output should be that which you've specified in the Result column.
We will then run your tests with several different incorrect answers. At least one of your tests should fail with each of the different incorrect answers. Only the first failing test will be shown to you. Although the tests we've given you in most questions so far in this course are often a single line of code, that's just to make it easy for you to read them. The tests you write here can be multiple lines of code if you wish.
Now ... go for it! See if you can find all our bugs. The penalty regime has been set to zero so you can get a feel for the how these questions work. There will be more of them in future labs.
]]>def is_old_enough(age):
"""True if age greater than or equal to 18, False otherwise"""
if age >= 18:
return "True"
else:
return "False"
The chances are that this wrong answer passes all your tests in the previous question. So in this question we've added that case to the set of answers you have to test. See if you can figure out how!
Hint: what is the output from
print(type("True"))
versus
print(type(True))]]>
We will provide a number of initially hidden different answers: one correct one and several incorrect ones. You must write one or more tests. When run with a correct answer all your tests should pass (that is, the output from each of your tests should equal the corresponding result). But when run on an incorrect answer at least one of your tests should fail.
The function you have to test in this question is the following:
Write a function alarm_clock(day, on_vacation) that takes an int day (encoded as 0=Sun, 1=Mon, 2=Tue, ...6=Sat) and a boolean on_vacation and returns a string of the form "7:00" indicating when the alarm clock should ring. Weekdays, the alarm should be "7:00" and on the weekend it should be "10:00". Unless we are on vacation -- then on weekdays it should be "10:00" and weekends it should be "off".
If you reject the right answer, you get zero marks. Otherwise your mark depends on what percentage of the incorrect answers you reject. You will be shown only the first failing case.
]]>If the student rejects the right answer, they get zero marks. Otherwise, depending on the setting of all-or-nothing, they may get partial marks if they reject some of the wrong answers.
The author-supplied test cases should contain several solutions to the specified question, at least one of which (usually the first) should be correct. The 'Output' of each test must be either 'Yes' if the supplied test case code is a valid answer or 'No' otherwise.
]]>