Lesson test: Tests, and building a program

Sign in

Libre University uses your GitHub account. Signing in is only needed to sit a final test, so the score is kept on your profile.

Tests, and building a program

1.[3p]
Why does a hand written `check` helper compare with `JSON.stringify` rather than `===`?

=== on two arrays or objects with identical contents is false, since it compares identity rather than value

Correct

JSON.stringify is faster than ===

The answer is: `===` on two arrays or objects with identical contents is false, since it compares identity rather than value

=== cannot compare numbers to numbers

The answer is: `===` on two arrays or objects with identical contents is false, since it compares identity rather than value

JSON.stringify ignores the order of keys

The answer is: `===` on two arrays or objects with identical contents is false, since it compares identity rather than value

2.[3p]
What did Dijkstra's 1969 remark about testing say?

Testing can show the presence of bugs, but never their absence

Correct

Testing is only worthwhile for programs longer than a thousand lines

The answer is: Testing can show the presence of bugs, but never their absence

A program with tests is proved correct

The answer is: Testing can show the presence of bugs, but never their absence

Testing should be done by someone other than the author

The answer is: Testing can show the presence of bugs, but never their absence

3.[3p]
Which of these cases are worth writing a test for?

Select all that apply

Empty input

Correct

A single element

Correct

Input the function is expected to reject

Correct

The exact case that produced a bug you have just fixed

Correct

A hundred randomly generated arrays

The answer is: Empty input, A single element, Input the function is expected to reject, The exact case that produced a bug you have just fixed

4.[2p]
What is the name for a test written from a bug you have just fixed, so that it cannot come back unnoticed? Write one word.

CorrectNot quite: regression

5.[2p]
`Number("4.35") * 100` does not give a whole number. What does `Math.round(Number("4.35") * 100)` give?

CorrectNot quite: 435

6.[3p]
Put these steps of building a program by decomposition in order.

Move on to the next piece
Assemble the tested pieces into the finished program
Write the smallest piece that can be tested on its own
Write the tests that pin down what that piece does

Show the answer

b, c, d, a

7.[2p]
A new test should be seen to fail before the code that makes it pass is written, because a test that has never failed might not be exercising the behaviour at all.

True

Correct

False

The answer is: True

8.[3p]
Match each function of the expenses program to its job.

parseAmount
parseLine
parseAll
totalsByCategory
formatPence

groups the records and sums each group
turns whole pence back into a printable string
splits the text, drops blank lines, and parses the rest
validates the shape of one line and returns a record
turns a text amount into whole pence

Show the answer

parseAmount: turns a text amount into whole pence parseLine: validates the shape of one line and returns a record parseAll: splits the text, drops blank lines, and parses the rest totalsByCategory: groups the records and sums each group formatPence: turns whole pence back into a printable string

9.[3p]
Why is `largest([-5, -2])` a test worth writing?

It catches the common bug of starting the accumulator at 0 rather than at the first element

Correct

It checks that the array is sorted before the search

The answer is: It catches the common bug of starting the accumulator at 0 rather than at the first element

It is the only case where the empty array matters

The answer is: It catches the common bug of starting the accumulator at 0 rather than at the first element

Negative numbers are compared differently from positive ones

The answer is: It catches the common bug of starting the accumulator at 0 rather than at the first element

1.[3p] Why does a hand written check helper compare with JSON.stringify rather than ===?

2.[3p] What did Dijkstra's 1969 remark about testing say?

3.[3p] Which of these cases are worth writing a test for?

4.[2p] What is the name for a test written from a bug you have just fixed, so that it cannot come back unnoticed? Write one word.

5.[2p] Number("4.35") * 100 does not give a whole number. What does Math.round(Number("4.35") * 100) give?

6.[3p] Put these steps of building a program by decomposition in order.

7.[2p] A new test should be seen to fail before the code that makes it pass is written, because a test that has never failed might not be exercising the behaviour at all.

8.[3p] Match each function of the expenses program to its job.

9.[3p] Why is largest([-5, -2]) a test worth writing?

1.[3p]
Why does a hand written `check` helper compare with `JSON.stringify` rather than `===`?

2.[3p]
What did Dijkstra's 1969 remark about testing say?

3.[3p]
Which of these cases are worth writing a test for?

4.[2p]
What is the name for a test written from a bug you have just fixed, so that it cannot come back unnoticed? Write one word.

5.[2p]
`Number("4.35") * 100` does not give a whole number. What does `Math.round(Number("4.35") * 100)` give?

6.[3p]
Put these steps of building a program by decomposition in order.

7.[2p]
A new test should be seen to fail before the code that makes it pass is written, because a test that has never failed might not be exercising the behaviour at all.

8.[3p]
Match each function of the expenses program to its job.

9.[3p]
Why is `largest([-5, -2])` a test worth writing?