Gzip compression and http post from python to java

Question

I want to http post a gzip compressed data from python to java and I want to store it as a BLOB in database. Then I want to gzip decompress that BLOB in java. So I want to know howto post a BLOB in python and how to read a BLOB in java. I have given my python and java code below. In my code I gzip compress a string in python and store that compressed data in a file. Then I read that file in java and decompress it using GZIPInputStream. But I'm getting the below exception.

java.io.IOException: Not in GZIP format
    at java.util.zip.GZIPInputStream.readHeader(GZIPInputStream.java:154)
    at java.util.zip.GZIPInputStream.<init>(GZIPInputStream.java:75)
    at java.util.zip.GZIPInputStream.<init>(GZIPInputStream.java:85)
    at GZipFile.gunzipIt(GZipFile.java:60)
    at GZipFile.main(GZipFile.java:43)

If I print the byte array of the compressed data in python I get

[31, 139, 8, 0, 254, 213, 186, 87, 2, 255, 203, 72, 205, 201, 201, 231, 229, 42, 207, 47, 202, 73, 1, 0, 66, 102, 86, 48, 12, 0, 0, 0]

If I read and print that compressed data from that file in java I get as

[31, -17, -65, -67, 8, 0, -17, -65, -67, -42, -70, 87, 2, -17, -65, -67, -17, -65, -67, 72, -17, -65, -67, -17, -65, -67, -17, -65, -67, -17, -65, -67, -17, -65, -67, 42, -17, -65, -67, 47, -17, -65, -67, 73, 1, 0, 66, 102, 86, 48, 12, 0, 0, 0]

Has you can see there is difference. If I give the printed byte array in python as input to the java code it works fine. So please help me to know how to post a blob(the compressed data) in python and how to read that compressed data in java to decompress it.

This is the compression code in python:

import StringIO  
import gzip  
import base64  
import os  


m='hello'+'\r\n'+'world'  

out = StringIO.StringIO()  
with gzip.GzipFile(fileobj=out, mode="wb") as f:  

    f.write(m.encode('utf-8'))
print list(array.array('B',out.getvalue())[:])
f=open('comp_dump','wb')  
f.write(out.getvalue())  
f.close()

This is the decompression code in java:

//$Id$

import java.io.*;  
import java.io.FileInputStream;  
import java.io.FileOutputStream;  
import java.io.IOException;  
import java.util.zip.GZIPInputStream;  
import javax.xml.bind.DatatypeConverter;  
import java.util.Arrays;

public class GZipFile
{


public static String readCompressedData()throws Exception
{
        String compressedStr ="";
        String nextLine;
        BufferedReader reader = new BufferedReader(new InputStreamReader(new FileInputStream("comp_dump")));
        try
        {
                while((nextLine=reader.readLine())!=null)
                {
                        compressedStr += nextLine;
                }
        }
        finally
        {
                reader.close();
        }
        return compressedStr;
}

public static void main( String[] args ) throws Exception
{
        GZipFile gZip = new GZipFile();
        byte[] contentInBytes = readCompressedData().getBytes("UTF-8");

        System.out.println(Arrays.toString(contentInBytes));
        String decomp = gZip.gunzipIt(contentInBytes);
        System.out.println(decomp);
}

/**
 * GunZip it
 */
public static String gunzipIt(final byte[] compressed){

        byte[] buffer = new byte[1024];
        StringBuilder decomp = new StringBuilder() ;

        try{

                GZIPInputStream gzis = new GZIPInputStream(new ByteArrayInputStream(compressed));

                int len;
                while ((len = gzis.read(buffer)) > 0) {

                        decomp.append(new String(buffer, 0, len));

                }

                gzis.close();

        }catch(IOException ex){
                ex.printStackTrace();
        }
        return decomp.toString();
}
}

Answer 1

Have you checked this: gzip a file in Python ?

My guess is that your string

m='hello'+'\r\n'+'world' 

is possibly causing some issues with the whole process...

Have you considered replacing it with m="hello\r\nworld", using double quotes instead?

Answer 2

You can't read the compressed data to string directly.What you have done in readCompressedData method is reading the compressed data to literal(which lead to a wrong string) and then get it's bytes(in method main).After you doing this,the contentInBytes is not really the bytes stored in the file.

When you try to make a string with bytes that can't be transformed into String. The bytes that represent the string is different.

For example:

        byte bytesBefore[] = {-1,-2,65,76,79,80};
        try {
            String str = new String(bytesBefore);
            byte bytesAfter[] = str.getBytes();
            System.out.println("str is " + str);
            System.out.println("after");
            for(Byte b : bytesAfter){
                System.out.print(" " + b);
            }
        } catch (Exception e) {
            e.printStackTrace();
        }

OUTPUT:

str is ��ALOP
after
 -17 -65 -67 -17 -65 -67 65 76 79 80

Because bytes -1 and -2 here can't be transformed into string,when you new the string with bytesBefore,the bytes that stored in memory for str is bytesAfter,which change the -1 and -2 to -17 -65 -67 -17 -65 -67 .

Actually, the GZIPInputStream can be built with a FileInputStream,no need to get the bytes first.Just use the BufferedReader to read the GZIPInputStream which is built with a FileInputStream.

There is a solution:

import java.io.*;
import java.util.zip.GZIPInputStream;

public class GZipFile {
    public static void main(String[] args) throws Exception {
        BufferedReader reader = new BufferedReader(new InputStreamReader(
                new GZIPInputStream(new FileInputStream(
                        "comp_dump")), "UTF-8"));
        StringBuffer sb = new StringBuffer();
        String line;
        while ((line = reader.readLine()) != null) {
            sb.append(line).append("\r\n");
        }
        System.out.println(sb.toString());
    }
}

OUTPUT:

hello
world