CODEWITHSUNDEEP
ctype-conversion
Hyeyeon
Hyeyeon

Reputation: 421

How to convert int to byte in C?

I would like to convert an int to a byte in C.

In Java, I'd write:

int num = 167;
byte b = num.toByte(); // -89

In C:

int num = 167;
???

Upvotes: 1

Views: 8767

Answers (4)

chqrlie
chqrlie

Reputation: 144550

byte is a java signed integer type with a range of -128 to 127. The corresponding type in C is int8_t defined in <stdint.h> for architectures with 8-bit bytes. It is an alias for signed char.

You can write:

#include <stdint.h>

void f() {
    int num = 167;
    int8_t b = num;  // or signed char b = num;
    ...

If your compiler emits a warning about the implicit conversion to a smaller type, you can add an explicit cast:

    int8_t b = (int8_t)num;  // or signed char b = (signed char)num;

Note however that it is much more common to think of 8-bit bytes as unsigned quantities in the range 0 to 255, for which one would use type uint8_t or unsigned char. The reason java byte is a signed type might be that there is no unsigned type in this language, but it is quite confusing for non-native readers.

byte can also be defined as a typedef:

typedef unsigned char byte;  // 0-255;

or

typedef signed char byte;    // -128-127;

Do not use type char because it is implementation defined whether this type is signed or unsigned by default. Reserve type char for the characters in C strings, although many functions actually consider these to be unsigned: strcmp(), functions from <ctype.h>...

Upvotes: 3

Lundin
Lundin

Reputation: 213319

In computer science, the term byte is well-defined as an 8 bit chunk of raw data. Apparently Java uses a different definition than computer science...

  • -89 is not the value 167 "converted to a byte". 167 already fits in a byte, so no conversion is necessary.

  • -89 is the value 167 converted to signed 2's complement with 8 bits representation.

  • The most correct type to use for signed 2's complement 8 bit integers in C is int8_t from stdint.h.

  • Converting from int to int8_t is done implicitly in C upon assignment. There is no need for a cast.

    int num = 167;
int8_t b = num;

Upvotes: 3

mousetail &#39;he-him&#39;
mousetail &#39;he-him&#39;

Reputation: 8001

You can simply cast to a byte:

unsigned char b=(unsigned char)num;

Note that if num is more than 255 or less than 0 C won't crash and simply give the wrong result.

Upvotes: 4

raphael.oester
raphael.oester

Reputation: 124

There is no such type as Byte in native C. Although if you don't want to import new libs, you can create one like this :

typedef unsigned char Byte

And then create any variable you'd like with it :

int bar = 15;
Byte foo = (Byte)bar

Upvotes: 3

Related Questions