Reputation: 181
I need to convert 2 byte array ( byte[2] ) to integer value in java. How can I do that?
Upvotes: 16
Views: 53398
Reputation: 1
import java.io.*;
public class ByteArray {
public static void main(String[] args) throws IOException {
File f=new File("c:/users/sample.txt");
byte[]b={1,2,3,4,5};
ByteArrayInputStream is=new ByteArrayInputStream(b);
int i;
while((i=is.read())!=-1) {
System.out.println((int)i);
FileOutputStream f1=new FileOutputStream(f);
FileOutputStream f2=new FileOutputStream(f);
ByteArrayOutputStream b1=new ByteArrayOutputStream();
b1.write(6545);
b1.writeTo(f1);
b1.writeTo(f2);
b1.close();
}
Upvotes: 0
Reputation: 3783
Simply do this:
return new BigInteger(byte[] yourByteArray).intValue();
Works great on Bluetooth command conversions etc. No need to worry about signed vs. unsigned conversion.
Upvotes: 3
Reputation: 2266
In Java, Bytes are signed, which means a byte's value can be negative, and when that happens, @MattBall's original solution won't work.
For example, if the binary form of the bytes array are like this:
1000 1101 1000 1101
then myArray[0] is 1000 1101
and myArray[1] is 1000 1101
, the decimal value of byte 1000 1101
is -115
instead of 141(= 2^7 + 2^3 + 2^2 + 2^0)
if we use
int result = (myArray[0] << 8) + myArray[1]
the value would be -16191
which is WRONG.
The reason why its wrong is that when we interpret a 2-byte array into integer, all the bytes are unsigned, so when translating, we should map the signed bytes to unsigned integer:
((myArray[0] & 0xff) << 8) + (myArray[1] & 0xff)
the result is 36237
, use a calculator or ByteBuffer to check if its correct(I have done it, and yes, it's correct).
Upvotes: 11
Reputation: 359816
You can use ByteBuffer
for this:
ByteBuffer buffer = ByteBuffer.wrap(myArray);
buffer.order(ByteOrder.LITTLE_ENDIAN); // if you want little-endian
int result = buffer.getShort();
See also Convert 4 bytes to int.
Upvotes: 25
Reputation: 7318
Also, if you can use the Guava library:
Ints.fromByteArray(0, 0, myArray[1], myArray[0]);
It was worth mentioning since a lot of projects use it anyway.
Upvotes: 3
Reputation: 126203
Well, each byte is an integer in the range -128..127, so you need a way to map a pair of integers to a single integer. There are many ways of doing that, depending on what you have encoded in the pair of bytes. The most common will be storing a 16-bit signed integer as a pair of bytes. Converting that back to an integer depends on whether you store it big-endian form:
(byte_array[0]<<8) + (byte_array[1] & 0xff)
or little endian:
(byte_array[1]<<8) + (byte_array[0] & 0xff)
Upvotes: 3