Question

double Money::rounding(int convert, Currency m)
{
    int tmp;
    if ( (convert % m) > (m / 2.0)) {
        tmp = convert + (m - (convert % m));
    } else {
        tmp = convert - (convert % m);
    }

    return convert / 100.0; // error happens here as well, 945 becomes 9.44
}

Money round(const Money target, Currency m)
{
    int tmp;
    double rounded;
    if (target.amount < 0) {
        tmp = -(target.amount * 100);
        rounded = -Money::rounding(tmp, m);

    } else {
        tmp = target.amount * 100;
        //error happens here
        std::cout << tmp << std::endl; // ouput 944, should be 945

        rounded = Money::rounding(tmp, m);
    }
    Money newM {0, 0};
    newM.setAmount(rounded);

    return newM;
}

int main(int argc, const char * argv[])
{
    Money a {9, 45}; // equal 9.45
    Money b = round(a, Currency::NICKLE); //argument a is 9.45

    cout << b << endl;

}

该函数是将Money实例目标舍入到最接近的货币，在本例中为NICKLE。因为货币a是9.45，它应该是9.45。但是在舍入之前的乘法中存在一个问题，tmp = target.amount * 100; ＆lt; - 变为944，但它应该是945，我知道舍入错误有时会发生，但这只会乘以100，为什么会出现舍入错误？

Answer 1

在功能rounding中定义tmp，但不要使用它。

应该是

return tmp / 100.0;

乘法舍入误差

1 个答案: