Why is my assertEquals() deprecated in JUnit?

问题:

I was wondering why assertEquals(double, double) is deprecated.

I used import static org.junit.Assert.assertEquals; and I used JUnit 4.11.

Below is my code:

import org.junit.Test;
import static org.junit.Assert.assertEquals;


public class AccountTest {

@Test
public void test() {
    Account checking = new Account(Account.CHECKING);
    checking.deposit(1000.0);
    checking.withdraw(100.0);
    assertEquals(900.0, checking.getBalance());
   }
}

checking.getBalance() returns a double value.

What could be wrong?

回答1:

It's deprecated because of the double's precision problems.

If you note, there's another method assertEquals(double expected, double actual, double delta) which allows a delta precision loss.

JavaDoc:

Asserts that two doubles are equal to within a positive delta. If they are not, an AssertionError is thrown. If the expected value is infinity then the delta value is ignored.NaNs are considered equal: assertEquals(Double.NaN, Double.NaN, *) passes

...

delta - the maximum delta between expected and actual for which both numbers are still considered equal.

回答2:

People explain but don't give samples... So here goes what worked for me:

@Test
public void WhenMakingDepositAccountBalanceIncreases() {
    Account account = new Account();
    account.makeDeposit(10.0);
    assertEquals("Account balance was not correct.", 10.0, account.getBalance(), 0);
}

The 0 in the end;

回答3:

assertEquals(double, double) is deprecated because the 2 doubles may be the same but if they are calculated values, the processor may make them slightly different values.

If you try this it will fail: assertEquals(.1 + .7, .8). This was tested using an Intel® processor.

Calling the deprecated method will trigger fail("Use assertEquals(expected, actual, delta) to compare floating-point numbers"); to be called.

回答4:

Old question but this hasn't been said yet and might help someone.

You can use com.google.common.math.DoubleMath.fuzzyEquals(double a, double b, double tolerance) which allows you to specify how close the two doubles should be to each other.

I found it very handy for unit tests where I don't want to hardcode test result values with a lot of decimal places.