C programming involving files and structures - c

Question

how to read a negative number from a file using fgetc?

Answer 1

fgetc only reads one character at a time. If you are trying to read in a negative number from a file -- or any number -- use fscanf.
#include <stdio.h>
main()
{
int v;
fscanf (stdin, "%d", &v);
printf ("v = %d\n", v);
}

Answer 2

If the "structures" in the title implies binary, then you likely want to use fread(), but if you really are chasing the question of what the layout of the integer stored in a binary file is, you can use fgetc().
This code shows how to use a union to map a series of read bytes back into an integer.
#include <stdio.h>
#include <stdlib.h>
#include <assert.h>
union Integer
{
int intPart_;
char charPart_[4];
};
int main(int argc, char* argv[])
{
FILE* pFile = fopen("integerFile.dat", "w");
int intWritten = -257;
size_t bytesWritten = fwrite(&intWritten, 1, sizeof(int), pFile);
assert(bytesWritten == sizeof(int));
fclose(pFile);
pFile = fopen("integerFile.dat", "r");
int intRead = 0;
size_t bytesRead = fread(&intRead, 1, sizeof(int), pFile);
assert(bytesRead == sizeof(int));
printf("%d\n", intRead);
fclose(pFile);
pFile = fopen("integerFile.dat", "r");
Integer intToRead;
for(int i = 0;
i != sizeof(int);
++i)
{
int byteRead = fgetc(pFile);
intToRead.charPart_[i] = byteRead;
printf("%d\n", byteRead );
}
printf("%d\n", intToRead.intPart_);
fclose(pFile);
return 0;
}

Answer 3

How's the number encoded?
If it's ASCII, remember it takes more than one character. You can write a loop for it, but you might find fscanf more help.
If it's binary data, remember fgetc is only going to read 8 bits anyway -- again, you need to think about other functions to do it efficiently.
The point here really is that unless you're doing it just to prove you can, fgetc is probably the wrong answer. Maybe fgets?

Answer 4

fgetc() interprets the character as an "unsigned char", but casts it to an int (but returns EOF, which is -1, on end of file).
If your source file contains some representation of signed values, then you need to decode that.

Answer 5

'0' has the ascii value 0x30、'1' is 0x31... and so far...
Assuming you have only one number in your file:
FILE * pFile;
int c, n = 0;
bool negative;
pFile=fopen ("myfile.txt","r");
if (pFile==NULL){
perror("Error opening file");
}else{
c = fgetc (pFile);
negative = (c == '-');
do {
c = fgetc (pFile);
if (c>=0x30 && c<=0x39) {
n = n*10 + (c-0x30);
}
} while (c != EOF);
fclose (pFile);
}
if(negative==true) n=n*-1;
printf ("Your number: %d\n",n);

Answer 6

I am with Charlie on the point that there are far better ways to do this than with what would need to be multiple calls to fgetc but if you insist on using that function you would need to run through a loop that evaluates each char. It also does depend on how the data is encoded. If it is ascii (as using a func that returns char would imply) you would check to see if the first char is "-" and then convert each subsequent char to int with atoi(const char*) one at a time, and multiplying your result value by ten each iteration before adding the new value to it. A better way would be to read several chars (using fgets or something) and then convert the char* with atoi(const char*). Maybe if you described more clearly what you were trying to do a better answer could be provided. Beware that using fscanf will fail if your data is not formatted exactly as you specify. But fscanf is really your answer for this problem it seems.

Answer 7

Without over/underflow-check you can use fscanf WITH return-value check:
#include <stdio.h>
int main()
{
int v;
if( 1==fscanf(yourfilepointer, "%d", &v) )
printf ("v = %d\n", v);
else
fputs("error on reading v",stderr);
return 0;
}