Ask Questions?

View Latest Questions



Posted on: July 26, 2006 at 12:00 AM
Java automatically converts decimal numbers you write in your source program into binary numbers internally.

Java Notes


Integers are whole numbers, for example, -35, 0, 2048, .... Integers are represented in binary inside the computer, and in decimal in Java source programs. Java automatically converts decimal numbers you write in your source program into binary numbers internally.

Four (or five) kinds of primtive integers and two integer classes.

Primitive types. The are four types of integers in Java: byte, short, int, long. The most common is int. All integers are stored in signed, two's-complement, format.

char! Technically, char is an unsigned integer type altho it is almost exclusively used to store characters. Making it integer is largely because of Java's legacy from C++. Don't use char for integers unless you are sure of what you're doing.

Classes. In addition to the primitive types, there are two classes used for integers.

  • Integer - Primarily useful for utility methods and to put in the Collections data structure classes.
  • BigInteger - Used where unbounded arithmetic is important.

How Java stores integers in memory

Java stores all integers in memory as binary numbers.

byte 18 -128+127
int 432-2,147,483,648+2,147,483,647
long 864-9,223,372,036,854,775,808+9,223,372,036,854,775,807

How to write integer literals

Here is how to write decimal integer literals (constants).

  • int literals are written in the usual decimal notation, like 34 or -222.
  • long literals are written by adding an L (or lowercase l altho this is almost impossible to distinguish from the digit 1), eg, 34L or -222L.
  • There is no way to write a literal byte or short, altho sometimes Java will automatically cast an int literal to the appropriate type.

Hexadecimal literals

You can write an int in hexadecimal by prefixing the hexadecimal number with the digit zero followed by the letter x, "0x" or "0X". The hexadecimal digits are 0-9 and the letters a-f in upper- or lowercase.

   int i;
   i = 0x2A;  // assigns decimal 42 to i.

The shame of integer arithmetic

Operations may produce numbers which are too large (overflow) to be stored in an int. No error is caused in this case; the result is simply an incorrect number (one of the shames of modern computer arithmetic). Division by zero will cause an execution exception (ArithmeticException). Use BigInteger to prevent arithmetic overflow.


Copyleft 2005 Fred Swartz MIT License