Compression?

[Dillinger4]

Ive been checking out some of the compression classes in java.util.zip and i was wondering if any extra information is added to a file when it is compressed. I slapped the following code together which just compresses and decompresses files from the command line. It seems that the more a file is compressed the larger it becomes. I took a file starting at 143 bytes and compressed it down to 72 bytes but after that each subsequent compression inflated the file size. 143,72,83,94,105,114. Now when a noncompressed file is attempted to be decompressed a java.util.zip.ZipException incorrect header check is thrown. So does file compression add headers to the file? It seems so.

  import java.io.*; 
  import java.util.zip.*;
   
  public class Compression{
   public static void main(String[] args){
    
   BufferedReader buff = new BufferedReader(new InputStreamReader(System.in)); 
   for(;;){ 
    try{
    System.out.println("enter mode" + " -c Or -d");  
    String mode = buff.readLine();
    System.out.println("enter file");
    String file = buff.readLine();
    if(mode.equals("") || file.equals("")){ 
      System.out.println("enter mode file");
      System.exit(0);
    }
    if(mode.equals("-c")){
         compress(file); 
    }
    else if(mode.equals("-d")){
         decompress(file);
    }
    else{
    System.out.println("Correct mode needed"); 
    }
    
    System.out.println("Like to continue? no to exit"); 
    String con = buff.readLine(); 
    if(con.equals("no")) break; 
 
    }catch(IOException e){;}
   }
  }
 
 public static void compress(String file){
   File f = new File(file);
   FileInputStream is = null;   
   DeflaterOutputStream dos = null;
 
   try{
       is = new FileInputStream(f); 
       byte[] buffer = new byte[(int)f.length()]; 
     while(true){
       int bytesread = is.read(buffer);     
         if(bytesread < 0){
            is.close(); 
            break;
        }
      }
     dos = new DeflaterOutputStream(new FileOutputStream(f));
     dos.write(buffer); 
     System.out.println("File compressed"); 
     }catch(Exception e){System.err.println(e);}
    finally{
       try{
       dos.flush();
       dos.close();
       }catch(Exception ie){System.err.println(ie);}
    }
   
   }
 public static void decompress(String file){
    File f = new File(file);
    FileOutputStream fos = null;   
    InflaterInputStream iis = null; 
    try{
    iis = new InflaterInputStream(new FileInputStream(f));   
    byte[] buffer = new byte[(int)f.length()];
    while(true){
      int bytesread = iis.read(buffer); 
     if(bytesread < 0){
        iis.close(); 
        break;
        }
    }
    fos = new FileOutputStream(f); 
    fos.write(buffer);
    System.out.println("File decompressed");  
    }catch(Exception ie){System.err.println(ie);} 
   finally{
       try{
       fos.flush();
       fos.close();
       }catch(Exception ie){System.err.println(ie);}
    }
   }
  }

[marnitzg]

Yeah, stuff like zip version, dictionary used etc is added as a header.

[Dillinger4]

Interesting. Id like to try it out on larger files to see how far down the file is compacted.

[CornedBee]

The header is one of two reasons why the file gets larger if you compress already compressed content.
The other lies in the compression algorithm.