问题
I have a value I read in from a file and is stored as a char*. The value is a monetary number, #.##, ##.##, or ###.##. I want to convert the char* to a number I can use in calculations, I've tried atof and strtod and they just give me garbage numbers. What is the correct way to do this, and why is the way I am doing it wrong?
This is essentially what I am doing, just the char* value is read in from a file. When I print out the temp and ftemp variables they are just garbage, gigantic negative numbers.
Another Edit:
I am running exactly this in gcc
#include <stdio.h>
int main()
{
char *test = "12.11";
double temp = strtod(test,NULL);
float ftemp = atof(test);
printf("price: %f, %f",temp,ftemp);
return 0;
}
and my output is price: 3344336.000000, 3344336.000000
Edit: Here is my code
if(file != NULL)
{
char curLine [128];
while(fgets(curLine, sizeof curLine, file) != NULL)
{
tempVal = strtok(curLine,"|");
pairs[i].name= strdup(tempVal);
tempVal = strtok(NULL,"|");
pairs[i].value= strdup(tempVal);
++i;
}
fclose(file);
}
double temp = strtod(pairs[0].value,NULL);
float ftemp = atof(pairs[0].value);
printf("price: %d, %f",temp,ftemp);
my input file is very simple name, value pairs like this:
NAME|VALUE
NAME|VALUE
NAME|VALUE
with the value being dollar amounts
SOLVED: Thank you all, I was using %d instead of %f and didn't have the right headers included.
回答1:
You are missing an include :
#include <stdlib.h>
, so GCC creates an implicit declaration of atof
and atod
, leading to garbage values.
And the format specifier for double is %f
, not %d
(that is for integers).
#include <stdlib.h>
#include <stdio.h>
int main()
{
char *test = "12.11";
double temp = strtod(test,NULL);
float ftemp = atof(test);
printf("price: %f, %f",temp,ftemp);
return 0;
}
/* Output */
price: 12.110000, 12.110000
回答2:
Code posted by you is correct and should have worked. But check exactly what you have in the char*
. If the correct value is to big to be represented, functions will return a positive or negative HUGE_VAL
. Check what you have in the char*
against maximum values that float
and double
can represent on your computer.
Check this page for strtod reference and this page for atof reference.
I have tried the example you provided in both Windows and Linux and it worked fine.
回答3:
printf("price: %d, %f",temp,ftemp);
^^^
This is your problem. Since the arguments are type double
and float
, you should be using %f
for both (since printf
is a variadic function, ftemp
will be promoted to double
).
%d
expects the corresponding argument to be type int
, not double
.
Variadic functions like printf
don't really know the types of the arguments in the variable argument list; you have to tell it with the conversion specifier. Since you told printf
that the first argument is supposed to be an int
, printf will take the next sizeof (int)
bytes from the argument list and interpret it as an integer value; hence the first garbage number.
Now, it's almost guaranteed that sizeof (int)
< sizeof (double)
, so when printf
takes the next sizeof (double)
bytes from the argument list, it's probably starting with the middle byte of temp
, rather than the first byte of ftemp
; hence the second garbage number.
Use %f
for both.
来源:https://stackoverflow.com/questions/10605653/converting-char-to-float-or-double