errno constant in C++

errno is a preprocessor macro used for error indication. 

• The value of errno is set to zero at program startup, and any function of the standard C++ library are allowed to write positive integers to errno whether or not an error occurred.
• Once the value of errno is changed from zero to non zero then no other function in the C++ standard library can change its value to zero.The errno is defined in cerrno header file.
• The value of errno is set to 33 when there is an error in mathematical argument. In C++ error of mathematical argument is represented by EDOM whose value is 33.

The same header that declares errno () also declares at least the following macro constants with values different from zero: 

• EDOM - Domain error: Some mathematical functions are only defined for certain real values, which is called its domain, for example, the square root and log function is only defined for non-negative numbers so if we pass negative argument in these function they set errno to EDOM
• ERANGE - Range error: The range of values that can be represented by a variable is limited. For example, mathematical functions such as pow can easily outbound the range representable by a floating point variable, or functions such as strtod can encounter sequences of digits longer than the range representable values. In these cases, errno is set to ERANGE.
• EILSEQ - Illegal sequence: Multibyte character sequence may have a restricted set of valid sequences. When a set of multibyte characters is translated by functions such as mbrtowc, errno is set to EILSEQ when an invalid sequence is encountered.

Below are the programs to implement the working of errno:
Program 1: This program detect error when negative value is passed in log function.  

#include <cerrno>
#include <clocale>
#include <cmath>
#include <cstring>
#include <iostream>
using namespace std;

int main()
{
    // log function doesn't take negative value
    // thus it changes value of errno to some positive number
    double not_valid = log(-1.0);

    // check if value of errno same as value of EDOM i.e. 33
    if (errno == EDOM) {
        cout << " Value of errno is : " << errno << '\n';
        cout << " log(-1) is not valid : "
             << strerror(errno) << '\n';
    }
    return 0;
}

Value of errno is : 33
 log(-1) is not valid : Numerical argument out of domain

Program 2: This program detect error when negative value is passed in square root function.

#include <cerrno>
#include <clocale>
#include <cmath>
#include <cstring>
#include <iostream>
using namespace std;
int main()
{
    // sqrt function doesn't take negative value
    // thus it changes value of errno to some positive number
    double not_valid = sqrt(-100);

    // check if value of errno same as value of EDOM i.e. 33
    if (errno == EDOM) {
        cout << " Value of errno is : " << errno << '\n';
        cout << " -100 is not valid argument for square"
             << " root function : " << strerror(errno) << '\n';
    }
    return 0;
}

Value of errno is : 33
 -100 is not valid argument for square root function : Numerical argument out of domain

Program 3: This program set errno to ERANGE.

#include <bits/stdc++.h>
using namespace std;

// Driver code
int main()
{
    double x;
    double res;

    x = 5.000000;
    res = log(x);

    if (errno == ERANGE) {
        cout << "Log(" << x << ") is out of range\n";
    }
    else {
        cout << "Log(" << x << ") = " << res << endl;
    }

    x = 10.00000;
    res = log(x);

    if (errno == ERANGE) {
        cout << "Log(" << x << ") is out of range\n";
    }
    else {
        cout << "Log(" << x << ") = " << res << endl;
    }

    x = 0.000000;
    res = log(x);

    if (errno == ERANGE) {
        cout << "Log(" << x << ") is out of range\n";
    }
    else {
        cout << "Log(" << x << ") = " << res << endl;
    }

    return 0;
}

Log(5) = 1.60944
Log(10) = 2.30259
Log(0) is out of range