File access in C++

Durga_Devi

Hi, While reading the double value from the written text file, the memory get corrupetd.Here is the sample code and its output for the scenario. Can anyone help in this.

#include ;
#include "stdafx.h"

int main(int argc, char* argv[])
{
double dbleArray[][4] = { -315.00000255999998,0.00000000000000000,0.00000000000000000,-166.19238000000001,
-329.19238000000001,315.00000255999998,-329.19238000000001,-329.19238000000001,
0.00000000000000000,0.00000000000000000,320.00000000000000,-536.00000000000000,
0.00000000000000000,0.00000000000000000,0.00000000000000000,1.0000000000000000};

//Write the data to the file
FILE\* pFile = NULL;
pFile = \_wfopen(L"C:\\\\Temp\\\\InputVolumeDetails.txt", L"w");
if(pFile)
{
	for(int nRow=0; nRow<4; nRow++)
	{
		for(int nCol=0; nCol<4; nCol++)
		{
			fwrite(&dbleArray\[nRow\]\[nCol\], sizeof(double), 1, pFile);
		}
	}
}
fclose(pFile);

//Read the data from the file
double dbleArray\_r\[4\]\[4\];
FILE\* pFile\_r = NULL;
pFile\_r = \_wfopen(L"C:\\\\Temp\\\\InputVolumeDetails.txt", L"r");
if(pFile\_r)
{
	for(int nRow=0; nRow<4; nRow++)
	{
		for(int nCol=0; nCol<4; nCol++)
		{
			fread(&dbleArray\_r\[nRow\]\[nCol\], sizeof(double), 1, pFile\_r);
			printf("%f\\n",dbleArray\_r\[nRow\]\[nCol\]);
		}
	}
}
fclose(pFile);

return 0;

}

The output for the above code is -315.000003 0.000000 0.000000 -92559631348757048000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000 -92559631349317831000000000000000000000000000000000000000000000.000000

Romualdas Cukuras

Try to change write and read modes to binary L"wb" and L"rb"

Lost User

A very subtle situation which is a great lesson for the future. Both your read and write loops contain no error checking. However having completed the write processing, a visual check of the output file suggests that the content is correct. The read loop reads sixteen values and prints them out, but does not check that you have read the full values every time. If you take a look at the file produced by this program you will see the hex value 'x1a' at character position 26. This character is taken by the fread() function as signifying no more data, so the last value read in is incomplete which gives the strange value printed by the program. Every subsequent read will fail so the value stored in memory is whatever was there previously; in this case the values that happen to be in the array storage space. The solution is to always check the results of your read and write statements, but in this case you also need to use "wb" and "rb" on your _wfopen statements to ensure the 'x1A' is not taken as end of file.

Durga_Devi

Thanks for all your replies. I tried with "wb" and "rb" mode and its working fine.

Lost User

Good news. But don't forget my comments about error checking, particularly when reading the file. Never assume that the file contents will be correct, however they were created.