Thread: String to Double function

Hi to all!,

I have a function that convert a string numeral (entered by a user) to a double value, having rounded its decimal value to (in this case) 4 significant places.

here is the code

Code:

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

int isNumber(const char str[])
{
	char *ok;
	strtod(str,&ok);
	return !isspace(*str) && strlen(ok)==0;
}


double stringToDouble(char *digit , size_t decPlace) // int dec is the number of decimal places
{

	double result = 0;
	char *ptr;
	char temp1[10]="";
	char temp2[15]="";

	if(digit != NULL && isNumber(digit)) // check if string is not NULL and a valid number
	{
		result = strtod(digit,&ptr);

		sprintf(temp1,"%%.%df",decPlace);
		sprintf(temp2,temp1,result);
		
		result = strtod(temp2,&ptr);

		
	}else
	{
		printf("FAILED:The String (%s) cannot be converted to double: result -> %g\n",digit, result);
	}

	return result;
}



int main()
{
	double num ;
	char numberInString[100];
	int decPlace = 4;
	
	//for testing purpose only
	printf("Enter a number\n");
	fgets(numberInString, sizeof(numberInString), stdin);
		if (numberInString[strlen(numberInString)-1] == '\n') 
		{   
			numberInString[strlen(numberInString)-1] = '\0';//remove the new-line
		}

	printf("the string %s\n",numberInString);
	
	num =stringToDouble(numberInString,decPlace);

	printf("is converted to double with value of %f with %d significant decimal places\n",num,decPlace);


	return 0;

}

OUTPUT

Enter a number
34.56789
the string 34.56789
is converted to double with value of 34.567900 with 4 significant decimal places

My question is with regards to stringToDouble().
I've used two temporary character array, temp1 and temp2 respectively, in order to manipulate the output of stringToDouble
in order for the stringToDouble to return the right value (with decimal place included).
Does this solution valid?

strtod() only returns the string in double form.AFAIK it can't round off value (decimal places).

Regards,
jaro

Originally Posted by citizen

I must say I don't see how this is supposed to be different or better than strtod(), especially since you call it in your function a couple times. Instead of trying to force a number to have four decimal places only, numbers usually round themselves if you specify the precision for output.

the function is not really for display purposes only.
the program that I'm currently working on is suppose to read a file and store it to the database. and I need to convert some string to numbers (in this case double) in order to be saved to the database.

thanks for the quick reply
-jaro

In that case, since your solution works, you can use it. As an alternative, you can study how strtod() really works (and also float-point representation in binary) and roll your own function to stop a certain number of places after a decimal, but you don't need to do that.