FileInputStream alone VS BufferedInputStream wrapped FileInputStream

[Krick]

I've been trying to settle on an implementation for a fileCopy method
in one of my utility classes.

After searching the newsgroups, I've narrowed it down to two
implemenations (source for both are below). The first uses
FileInputStream directly, the second wraps the FileInputStream in a
BufferedInputStream. Both methods read in 8K chunks using their own
internal buffers.

My question is this:

Which implemetation is better, and why?


....
Krick





public static void fileCopy1(String src, String dst) {
try {
FileInputStream in = new FileInputStream(new File(src));
FileOutputStream out = new FileOutputStream(new File(dst));
byte[] buffer = new byte[8 * 1024];
int count = 0;
do {
out.write(buffer, 0, count);
count = in.read(buffer, 0, buffer.length);
}
while (count != -1);
in.close();
out.close();
} catch (IOException ex) {}
}

public static void fileCopy2(String src, String dst) {
try {
BufferedInputStream in = new BufferedInputStream(new
FileInputStream(src));
BufferedOutputStream out = new BufferedOutputStream(new
FileOutputStream(dst));
byte[] buffer = new byte[8 * 1024];
int count;
while ((count = in.read(buffer)) >= 0) {
out.write(buffer, 0, count);
}
in.close();
out.close();
} catch (IOException ex) {}
}

 

[Shripathi Kamath]

I've been trying to settle on an implementation for a fileCopy method
in one of my utility classes.

After searching the newsgroups, I've narrowed it down to two
implemenations (source for both are below). The first uses
FileInputStream directly, the second wraps the FileInputStream in a
BufferedInputStream. Both methods read in 8K chunks using their own
internal buffers.

My question is this:

Which implemetation is better, and why?

Benchmarking the two will likely give you the answer that is better for your
app. As far as one can see, the only difference is the buffering which is
usually meant to improve performance.

Of course, the output can be buffered as well.

HTH,

 

[Marco Schmidt]

Jack9:

I have no idea and every
tutorial on socket programming seems to implement network connections
differently without explaining WHY they choose the streams they do.
Any help would be appreciated!

Check out Roedy's overview at <http://mindprod.com/jgloss/io.html> and
<http://mindprod.com/fileio.html>.

You should also visit related terms like socket.

Regards,
Marco