C++/freelancer rates

[jaunty surge]

Hi.

I am not sure what is wrong with my code. From my side the exercise is done, however when I run tests, I am getting error displayed on attached screenshot.

But if I am not wrong 126.0 == 126.0. Has anyone had similar issue?
Thanks!

[ancient wing]

Hello mer, Would you please supply your code as text, so we have more information?

[random hearth]

Increase your chance of getting help and look like a pro by sharing codeblocks not images. For example, you can type the following. Note, the ``` must be on their own line.
```
for number in range(10):
total += number;
```
Discord will render that as so:

for number in range(10):
    total += number;

Click here to learn more about codeblocks: https://exercism.org/docs/community/being-a-good-community-member/writing-support-requests and http://bit.ly/howto-ask

[jaunty surge]

so i resolved the issue, however i am not sure why this worked.
my original solution looked like this:

double apply_discount(double before_discount, double discount) {
    return before_discount / 100.00 * (100 - discount);
}

the updated solution, looks like this (and all tests passed):

double apply_discount(double before_discount, double discount) {
    return before_discount * (100 - discount) / 100.00;
}

maybe my question is why this change helped?

[clear bridge]

It could be a rounding issue

[next parcel]

I believe it is a rounding error. I made this code to check:

#include <iostream>
#include <iomanip>
#include <limits>

int main() {
    const double before = 140.0;
    const double discount = 10;

    const double opt1 = before / 100.00 * (100 - discount);
    const double opt2 = before * (100 - discount) / 100.00;
    const double target = 126.00;

    // Print booleans as true/false
    std::cout << std::boolalpha;

    std::cout << "opt1 == opt2: "   << (opt1 == opt2) << "\n"; 
    std::cout << "opt1 == target: " << (opt1 == target) << "\n"; 
    std::cout << "opt2 == target: " << (opt2 == target) << "\n"; 
    
    // Print doubles with high precision and scientific notation
    std::cout << std::setprecision(std::numeric_limits<double>::digits10 + 1) << std::scientific;
    
    std::cout << "opt1: "   << opt1 << "\n"; 
    std::cout << "opt2: "   << opt2 << "\n"; 
    std::cout << "target: " << target << "\n"; 
}

The output is consistent with what we see when using gcc 13.2 x86-64:

opt1 == opt2: false
opt1 == target: false
opt2 == target: true
opt1: 1.2599999999999999e+02
opt2: 1.2600000000000000e+02
target: 1.2600000000000000e+02

Proof that it is a bad idea to test doubles for equality!

[ancient wing]

I used a matcher for one for the test cases with double, but not for the others. I will open an issue to correct that:
https://github.com/exercism/cpp/blob/b6b3993315dee5c163c38f1db24655e716c165e4/exercises/concept/freelancer-rates/freelancer_rates_test.cpp#L26C19-L26C64

[ancient wing]

That took a while, but there is a PR now. https://github.com/exercism/cpp/pull/936