Printable View
Hi all,
The job is to read a .mp3 binary file and create(write) a new .mp3 file which is same as the original file. So, the basic thing we will be trying to do is to read/write of binary file(.doc, .pdf, .xls, .mp3). So the code should finally create a binary file(say a .mp3), which when opened with a player, should be recognized and played correctly.
Has anyone worked on this?
I mean, it will be appreciating if they can help me in getting started with this.
Regards
Swapna
Are you just trying to make an exact duplicate of a file?
If so, there are many options in the methods of read/write. Why don't you give it a shot and if you have trouble along the way, someone here can take a look at your code.
Check out the functions fread() and fwrite()
If you only care about Windows, then CopyFile may be the right choice.
--
Mats
There is a better way of doing this (read,write) but here is the code I used when making that program:
Code:#include <stdio.h>
#include <stdlib.h>
unsigned char byte;
int main(void) {
FILE *fin, *fout;
int val;
fin = fopen ("test.txt","rb");
if (fin == NULL) {
printf("Error opening input file\n");
system("pause");
exit(1);
}
fout = fopen ("newtext.txt","wb");
if (fout == NULL) {
printf("Error opening output file\n");
system("pause");
exit(1);
}
printf("Files opened, copying beginning.\n");
while(1) { //infinant loop
val = fgetc(fin);
if (val == EOF)
break;
byte = (unsigned char) val;
fputc(byte,fout);
}
printf("Copy successful!\n");
system("pause");
}
Hardly very efficient tho' - if the file is large, you'll be making MANY calls to fgetc()/fputc(), where a fread/fwrite loop could easily reduce the amount of calling-overhead by a factor of a few thousand or more (you could read 1MB at a time, which would reduce the overhead by a factor of 1 million, roughly).
--
Mats
That's what I meant by it could be better.
THought it would be useful for someone, so heres the code for copying a binary file.
Code:#include <iostream>
#include <fstream>
using namespace std;
int main () {
filebuf *pbuf;
ifstream sourcestr;
ofstream desstr;
long size;
char * buffer;
sourcestr.open("C:\\E_Drive\\Projects\\I-Radio\\Jashnebahaara.mp3",ios::in | ios::binary);
desstr.open("C:\\E_Drive\\Projects\\I-Radio\\Jashnebahaara1.mp3", ios::out | ios::binary);
// get pointer to associated buffer object
pbuf=sourcestr.rdbuf();
// get file size using buffer's members
size=pbuf->pubseekoff (0,ios::end,ios::in);
pbuf->pubseekpos (0,ios::in);
// allocate memory to contain file data
buffer=new char[size];
// get file data
pbuf->sgetn (buffer,size);
sourcestr.close();
// write content to Jashnebahaara1.mp3
desstr.write(buffer,size);
desstr.close();
return 0;
}
Shame that's not C...
The downside to your method is you need at least size bytes free in memory to copy a file size bytes big. Waste of resources in my opinion. Do what matsp suggested.
Yes, and on a windows system, a file bigger than 2GB won't be possible to copy (unless you have large amounts of RAM _AND_ a 64-bit version of Windows, and compiled the code for 64-bit).Quote:
Originally Posted by zacs7
--
Mats