Need help creating function to double the digits of an integer individually - c

Question

So I'm trying to make a recursive function that takes an integer, let's say 123, and gives back double every single digit in the integer. So 123 would become 248, not 246. The main difference obviously is that instead of doing 123x2, you do (100x2)+(20x2)+(3x2). There is also the condition that if any of the numbers are equal to or greater than 5, you replace it with a 9, so 345 becomes 689. I am trying to create it iteratively before making it recursively and I am running into an issue with the conditions. Heres what I have so far:
int double_digit(int g) {
int dubl = 0;
int mod = 10;
int i = 1;
while (g > 0) {
mod = pow(10, i);
if (g % 10 < 5) {
dubl = dubl + g % mod * 2;
g = g - (g % mod);
i++;
}
else {
dubl = dubl + (9 * (mod / 10));
g = g - (g % mod);
i++;
}
}
printf("%d", dubl);
}
If you run the code, you will see that it works for numbers under 5, but not greater than or equal to, how can I fix it?

Answer 1

WAY too complicated... A lookup table will shrink the code down to almost nothing. This even tries to protect against receiving a negative number.
int double_digit(int g) {
int dd = 0;
for( int w = abs(g), pow = 1; w; w /= 10, pow *= 10 )
dd += pow * "\0\02\04\06\10\011\011\011\011\011"[ w % 10 ];
return g < 0 ? 0 - dd : dd;
}
int main() {
int tests[] = { 42, 256, 123, 578, -3256, };
for( int i = 0; i < sizeof tests/sizeof tests[0]; i++ )
printf( "%d = %d\n", tests[i], double_digit( tests[i] ) );
return 0;
}
42 = 84
256 = 499
123 = 246
578 = 999
-3256 = -6499
EDIT:
C's understanding of octal bytes specified in a character array (aka string) may be new.
Here is an abridged alternative that may not be as confrontational:
int lut[] = { 0, 2, 4, 6, 8, 9, 9, 9, 9, 9, };
for( int w = abs(g), pow = 1; w; w /= 10, pow *= 10 )
dd += pow * lut[ w % 10 ];
And, then there are branchless techniques (overkill in this instance, but useful to consider in other applications):
for( int d, w = abs(g), pow = 1; w; w /= 10, pow *= 10 )
d = (w%10)<<1, dd += pow * (d*(d<9)+(d>8)*9);
In other words, there are lots of ways to achieve objectives. A good exercise is to try to discover those different ways and then to consider where each may be used to the most benefit (and where other methods may not appropriate or as clear.) Write clean, clear, concise code to the best of your abilities.
I've written the recursive version. Without accounting for negative values, the recursive version requires, to the best of my abilities, more code and takes a 'linear' (iterative) problem into other dimensions. Make the method fit the problem, not vice versa.