CODEWITHSUNDEEP
javaioarraylistarrays
lazyprogrammer
lazyprogrammer

Reputation: 629

Breaking binary files

With reference to my previous question.

I have made the program with the following approach:
The program first reads 2k of data from the file and stores it into a byte array.
Then the data to be added to each packet is also stored in an array and both are added to an array list. The array list is then written to an output stream for the file.

The Code is here:

File bin=chooser.getSelectedFile();
int filesize=(int)bin.length();
int pcount=filesize/2048; 
byte[] file=new byte[filesize];
byte[] meta=new byte[12];
int arraysize=pcount*12+filesize;
byte[] rootfile=new byte[46];
ArrayList al = new ArrayList();
String root;
prbar.setVisible(true);
int mark=0;
String metas;
try{
    FileInputStream fis=new FileInputStream(bin);
    FileOutputStream fos=new FileOutputStream(bin.getName().replace(".bin", ".xyz"));
    ObjectOutputStream os=new ObjectOutputStream(fos);
    root="46kb"+"5678"+"0000"+pcount+"MYBOX"+"13"+"S208";
    rootfile=root.getBytes();
    for(int n=0;n<=pcount;n++)
    {
        fis.read(file, 0, 2048);
        mark=mark+2048;
        int v=(mark/filesize)*100;
        prbar.setValue(v);
        metas="02KB"+"1234"+n;
        meta=metas.getBytes();

        al.add(rootfile);
        al.add(meta);
        al.add(file);
    }
    os.writeObject(al.toArray());
}
catch(Exception ex){
    erlabel.setText(ex.getMessage());
}

The program runs without any errors but the file is not created correctly. Either the approach is wrong or the code.

Please Help

Upvotes: 0

Views: 445

Answers (2)

Peter Lawrey
Peter Lawrey

Reputation: 533492

You appear to be writing your own binary format but you are using ObjectOutputStream which has it own header. writeObject write an Object not data in a manner that lets a Java process deserialize that object e.g. with it class hierarchy and field names.

For binary, I suggest you use a plain DataOutputStream with a BufferedOutputStream which will be more efficient and do what you want.

I also suggest you write the data as you generate it rather than using an ArrayList. This will use less memory, make the code simpler and be faster.

I would write the code more like this

File bin = chooser.getSelectedFile();
int filesize = (int) bin.length();
int pcount = (filesize + 2048 - 1) / 2048;
byte[] file = new byte[2048];

FileInputStream fis = new FileInputStream(bin);
String name2 = bin.getName().replace(".bin", ".xyz");
OutputStream os = new BufferedOutputStream(new FileOutputStream(name2));
byte[] rootfile = ("46kb" + "5678" + "0000" + pcount + "MYBOX" + "13" + "S208").getBytes("UTF-8");

for (int n = 0; n < pcount; n++) {
    os.write(rootfile);
    byte[] metas = ("02KB" + "1234" + n).getBytes("UTF-8");
    os.write(metas);

    int len = fis.read(file);

    os.write(file, 0, len);
    int percent = 100 * n / pcount;
    prbar.setValue(percent);
}
ow.close();

Upvotes: 2

Joop Eggen
Joop Eggen

Reputation: 109547

With the smallest thing first:

int v=(mark/filesize)*100;

Is using integer division yielding always 0 I think.

int v = mark * 100 / filesize;

The byte[] object (file for instance) is created once and many times added to the list. You get n copies of the last overwrite.

Upvotes: 1

Related Questions