Conversion between binary and decimal number

Question

I woud like to ask, what bin-'0' means in this piece of code which convert binary number to decimal. Thanks.

#include <stdio.h>
#include <stdlib.h>

int main(){
    char bin;
    int dec = 0;

    printf("Binary: \n");
    bin = getchar();

    while((bin != '\n')){
        if((bin != '0') && (bin != '1')){
            printf("Wrong!\n");
            return 0;
        }
        printf("%c",bin-'0');  // ?

        dec = dec*2+(bin-'0'); // ?
        bin = getchar();
    }

    printf("Decimal: %d\n", dec);

    return 0;
}

The statement `printf("%c",bin-'0');` does nothing useful. Try `"%d\n"` there, instead. — Robᵩ, Oct 30 '12 at 14:53

tomahh · Answer 1 · 2012-10-30T15:07:15.347

4

bin - '0' converts the ASCII value of bin to its integer value. Given bin = '1', bin - '0' = 1

edited Oct 30 '12 at 15:07

answered Oct 30 '12 at 14:50

tomahh

13,441
3
49
70

Is this guaranteed ('1' being 49)? – Luchian Grigore Oct 30 '12 at 14:50
2

@LuchianGrigore I'm pretty sure that ascii table is standardized. But I dont want to affirm that. – tomahh Oct 30 '12 at 14:51
@LuchianGrigore According to [wikipedia](http://en.wikipedia.org/wiki/ASCII), yes, It is sure. – tomahh Oct 30 '12 at 14:53
3

ASCII stands for American Standard Code for Information Interchange, so I guess it's standardized :) – Andreas Brinck Oct 30 '12 at 14:53
wikipedia isn't a solid reference. – Luchian Grigore Oct 30 '12 at 14:53
1

Does the standard say C++ has to use ASCII? – Luchian Grigore Oct 30 '12 at 14:55
It will surely depend on what encoding user's terminal is using. – Oct 30 '12 at 15:00
I posted a new question [here](http://stackoverflow.com/questions/13141776/is-0-guaranteed-to-be-48) because I'm not completely sure of the validity of your statement. – Luchian Grigore Oct 30 '12 at 15:01
@LuchianGrigore The (nonsensical) argument here is whether the integer value for char '0' is 48. I don't see how that is relevant, the only assumption being made by the code is that the character encodings for ['0' - '9'] are contiguous. – Praetorian Oct 30 '12 at 15:03
To be technical, it converts a *character reprsentation* of a **digit** to its numerical equivalent, *independent of the character's encoding*. – Thomas Matthews Oct 30 '12 at 16:13

score 1 · Answer 2 · answered Oct 30 '12 at 14:51

This code is taking advantage of the fact that C++ chars are really just special ints. It's using getchar to take in a char that is either '0' or '1'. Now it needs to convert that into 0 or 1 (note that these are numbers, not chars). Given that the char '0' is one before '1', subtracting the value of char '0' from both will turn '0' into 0 and '1' into 1.

'0' - '0' = 0
'1' - '0' = 1

Conversion between binary and decimal number

2 Answers2

Linked