Cpp: Formatting Output

Decimal, Octal, and Hexadecimal Output

cout << "hex:" << hex << 11 << " " << 12 << " " << 13 << endl
     << "oct:" << oct << 11 << " " << 12 << " " << 13 << endl
     << "dec:" << dec << 11 << " " << 12 << " " << 13 << endl;

hex used for display the hexadecimal numbers in the stream, then oct displays octal numbers, dec restores the output to decimal. The keywords that change the behavior of the stream, such as hex and oct, are called manipulators.

Once the format of output is changed by manipulators like hex and oct, the numbers following them will be output in that base (persistent):

cout << hex << 10 << endl;  // outputs a
cout << 10 << endl;         // still outputs a

showbase displays the base prefix. It adds the prefix 0x when outputting hexadecimal numbers, and the prefix 0 when outputting octal numbers:

cout << showbase << hex << 11 << endl; // 0xb
cout << oct << 11 << endl; // 013

unshowbase cancels the prefix:

cout << showbase << hex;
cout << 11 << endl; // 0xb
cout << unshowbase;
cout << 11;         // b

Decimal, Octal, and Hexadecimal Input

hex, oct, and dec can also be used with cin:

int a, b, c;
cin >> a >> hex >> b >> oct >> c;
cout << a << " " << b << " " << c;

Input and output:

Input: 10 0xa 012
Output: 10 10 10
Input: 10 a 12
Output: 10 10 10

Floating-Point Output

There are three output formats:

• defaultfloat: Outputs at most six digits; too large numbers are output in scientific notation.
• fixed: Displays six digits after the decimal point.
• scientific: Scientific notation.

double num = 1234567890.123456789;
cout << num << "\n"                // defaultfloat: 1.23457e+09
     << fixed << num << "\n"       // fixed:        1234567890.123457
     << scientific << num << "\n"; // scientific:   1.234568e+09

Displays n digits after the decimal point: fixed + setprecision(), which requires the iomanip library:

#include <iomanip>
double num = 1234567890.123456789;
cout << setprecision(12) << num << "\n" // 1234567890.12
     << fixed << num << "\n"            // 1234567890.123456716537
     << scientific << num << "\n";      // 1.234567890123e+09

Setting Output Width

The setw() function from the <iomanip> library can be used to set the width of the output. It’s not persistent, meaning it only affects the next output statement. Here is an example:

const int w = 16;
cout << "|" << setw(w) << "Hello" << "|\n"; // |           Hello|
cout << "|" << setw(w) << "123"   << "|\n"; // |             123|
cout << "|" << setw(w) << "abc"   << "|\n"; // |             abc|

String Streams

The <sstream> header file provides the ability to manipulate strings as if they were input and output streams. This can be useful for converting between numeric and string representations or for formatting output.

Converting a string to a number can be done using an istringstream object, like so:

istringstream is { s };
double d;
is >> d;
if (!is) throw runtime_error("double format error: " + s);

Converting a number to a string can be done using an ostringstream object, like so:

ostringstream os;
int x = 12, y = 34;
os << "Point { " << x << ", " << y << " }";
string s = os.str();

Inputting by Line

The getline() function can be used to input a line of text from the standard input stream cin. The function returns a string that does not include the newline character.

string l;
getline(cin, l);