Formatting Numbers with C++ Output Streams
David Kieras, EECS Dept., Univ. of Michigan
Revised for EECS 381, Winter 2004.
Using the output operator with C++ streams is generally easy as pie, with the only hard part being controlling the format of
the output. As you will see, this is relatively clumsy compared to C stdio's printf function, but printf is not type-safe. An
output stream has a bunch of member variables that control the details of what output looks like, and these variables can be
controlled by a set of member functions and special objects called manipulators. Most of these variables retain their values
until they are changed, but a few only act temporarily - for the next output operation only.
Throughout this document, the output stream cout is used in the examples. However, everything works for any text output
stream, such as an output file stream.
The prototypical output statement in C++ is:
cout << "Hello, world!" << endl;
This contains a manipulator, endl. This is an object, which when supplied to operator<<, causes a newline character to
be put into the output stream, followed by a call of cout's flush function, which causes the internal buffer to be immediately
emptied. This makes sure all of the output is displayed before the program goes on to the next statement. So manipulators are
objects that cause the output stream object to do something, either to its output, or to its member variables.
The manipulators with no arguments, like endl, are included in <iostream>. If you use manipulators that take arguments
(like setprecision and setw) you need to #include <iomanip>.
Like the input operator, operator<< has overloaded definitions for all of the built-in types. By default, characters and
strings are simply output as is, and this is usually satisfactory. So the main complexity is controlling the format of numbers.
There are a zillion variables that control numeric and other formats. This document focuses on how to control the most
important and common aspects of numeric output.
You can control the format by using member functions or manipulators to change a variety of member variables. This
document summarizes the most important and common things you might want to do; check a complete reference to do more.
Quite a few things can be done with the stream formatting controls, and not just with numbers; if you find yourself tempted to
write complicated code to make the output look neat, stop and consult a good reference first - you might be about to reinvent
the wheel.
Default output format for integers and doubles
All of the digits of an integer will be printed using decimal (base 10), with no leading zeros and a leading minus if it is
negative, using as many characters as needed, but no more.
A double value will be printed using a general format that usually works fine to represent a number in a small amount of
space. The basic amount of space used is determined by the precision. The default precision is 6, which means up to 6
significant digits are used to represent the number. This counts digits both to the left and the right of the decimal point. Fewer
digits are printed if the result is accurate; e.g. trailing zeros at the right of the decimal point are not printed. As shown in the
example below, as the number increases in size, places to the right of the decimal point will be dropped, and the result
rounded off as needed to stay within 6 digits. If the number cannot be represented with six digits to the left of the decimal
point after all to the right have been dropped, the output will flip into exponential (scientific) notation showing 6 significant
1
