Hey,
I have written a piece of code to do as the title says, but I seem to have a problem in that any odd number of digits is converted wrong. eg. it converts 25 correctly but not 125.
it gives me 1 less than the value I expect, output is 124 for input of "125"
Here is my code, can anyone see any mistakes I've been looking over it for ages now and can't figure it out.
Last edited by manutdfan; 01-06-2008 at 10:32 AM. Reason: no longer needed code
Works as expected for me. 125 gives 125.
Originally Posted by Adak
Thats really odd, i've tried re-writing it into a single function which i've posted below and I still get the same problem.
im using Dev C++ to compile in windows. Cheers for looking at it.
Code:#include <stdio.h> #include <string.h> #include <stdlib.h> #include <math.h> int stringToInt(char s[]) { int i = 0; int power = strlen(s) - 1; int value = 0, curNo = 0, sign = 0; /* test for negativity */ if(s[i] == '-') { sign = 1; i++; power--; } while(s[i] != '\0') { curNo = s[i] - 48; if(curNo > 9 || curNo < 0) { printf("error"); exit(1); } value = (value + (curNo * pow(10,power))); i++; power--; } if(sign == 1) { value = -(value); } return value; } int main(int argc, char* argv[]) { printf("%d", stringToInt(argv[1])); return 0; }
Your logic looks okay, but I suggest that there might be a floating point inaccuracy problem with your use of pow(10, power).
Look up a C++ Reference and learn How To Ask Questions The Smart Way
Indeed: (int)pow(10,2) prints as 99 on my system.
Edit: That's not true, what I just said; what I meant was that curNo*pow(10,power) prints as 99. See:
gives me 100 =? 99 =? 100 (using gcc/Dev-C++). I'm guessing the compiler can replace the first with the third, but not the second.Code:int main(void) { int power=2, curNo=1; printf("%d", (int)(1*pow(10.0,2))); printf(" =? %d", (int)(curNo*pow(10,power))); printf(" =? %d\n", (int)(10.0*10.0)); return 0; }
So maybe replace your pow(10,power) with a variable that starts at 100000 (or wherever is appropriate) and gets divided by 10 each time.
Last edited by tabstop; 01-06-2008 at 11:03 AM. Reason: accuracy
Prints 100 for me. But that's really poor; there's no reliable pow function for integers.
Indeed. I see two options:Prints 100 for me. But that's really poor; there's no reliable pow function for integers.
1. Write your own pow-like function for ints.
2. Reverse the portion of the numeric string consisting of digits, then iterate again, but with a multiplier that is itself multiplied by 10 on each iteration, starting with a multiplier value of 1.
Look up a C++ Reference and learn How To Ask Questions The Smart Way
Cheers guys, that was the problem have just resorted to using a for loop to calculate pow value instead.
Why do you compare with the minus sign by writing '-' but when you subtract that value of '0' you write it explicitly as 48?
If you're going to write '0' as 48 you might as well write '-' as 45, right?
Just use a character literal.
