problerm with string equals

[gaurav abbi]

hi,
i'm using junit for testing my http server client functionality.
here one test looks for the expected and actual Content-Length values.
exp = "66"
actual = "66"
Although the display shows same strings with values "66" and "66" for expected and actual, the test fails.
on further investigation, i found that the exp.equals(actual) returns false.
Moreover, when i printed the byte arrays for both the strings, it was returning same sequence of bytes.

i tried printing the char arrays, but my linux machine can't display some characters, so it was throwing exceptions.
byte[] b1 = new byte[exp.length()]
byte[] b2 = new byte[actual.length()]
but the

Arrays.equals(exp.getChars(0, contentLength.length(), b1, 0), actual.getChars(0,actual.length(), b2, 0)) 

returns false.
also their hascode are different
so from this i can conclude the charater sequence is different for both the strings which i'm not sure why?

is it because of different charsets or formatting or because of client server scenario?(client and server are on same machine).
i tried testing with different charsets and it works fine. equals() returns true.
i'm totally stuck
can anybody throw som light on this or is it a bug in java

[Amit Y Desai]

yes i also faced a similar problem

Integer i = new Integer(10);
String j= new String("10");
then i.equals(j) returned false...
but when i used i.toString().equals(j) it returned true..

really was surprising as in case of i.equals(j) internally should have used toString() of i and j only

[gaurav abbi]

they both are strings and it seems to be a charset problem.
trying to solve it.

[Campbell Ritchie]

The reason an Integer 10 and a String "10" don't return equal is that they are completely different kinds of Object.

Suggest you print out your char[] arrays using the %d or the %x formatting tag.

for(char c : charArray)
   System.out.printf("%x ", c);

See whether the two arrays are in fact equal. It makes me suspicious if you are getting Exceptions.

[edit]Try changing %x to %04x for better formatting[/edit]
[ February 22, 2008: Message edited by: Campbell Ritchie ]

[Joanne Neal]

Are there any spaces at the beginning or end of the strings ?

[Bill Shirley]

Some actual code, rather than code snippets, would help us debug your problem.

[camelo]

Want code? Here you go:

This segmet of code is an intent to encode a non ASCII String into an ASCII one:

 
public class ByteSeparatorEncoder implements SimpleStringEncoder
{
    private final String separator = "e";
 
    public String decodeComplex( String simpleString )
    {
        String[] byteRepresentation = simpleString.split( separator );
        byte[] bytes = new byte[ byteRepresentation.length ];
 
        for( int i = 0 ; i < bytes.length ; i++ )
            bytes[ i ] = Byte.parseByte( byteRepresentation[ i ] );
        return new String( bytes );
    }
 
    public String encodeSimple( String complexString )
    {
        byte[] bytes = complexString.getBytes();
        StringBuffer visibleString = new StringBuffer();
 
        for( int i = 0 ; i < bytes.length ; i++ )
        {
            visibleString.append( bytes[ i ] );
            visibleString.append( separator );
        }
        return visibleString.toString();
    }
}   

For test you can use the following JUnit:

  private static final int NUMBER_OF_TRYS = 10;
  private ByteSeparatorEncoder encoder = new ByteSeparatorEncoder();
 
  public void testSimpleEncoding2() throws Exception
    {
        SecureRandom random = new SecureRandom();
        ByteBuffer buffer = null;
        Charset charset = Charset.forName( "US-ASCII" );
        CharsetDecoder decoder = charset.newDecoder();
        byte[] complexBytes = null;
        String[] complexStrings = new String[ NUMBER_OF_TRYS ];
        String[] simpleStrings = new String[ NUMBER_OF_TRYS ];
 
        for( int i = 0 ; i < NUMBER_OF_TRYS ; i++ )
        {
            complexBytes = random.generateSeed( 200 );
            complexStrings[ i ] = new String( complexBytes );
            simpleStrings[ i ] = encoder.encodeSimple( complexStrings[ i ] );
            buffer = ByteBuffer.wrap( simpleStrings[ i ].getBytes() );
            // If the below line don't thorw exception then the simple string
            // returned is ASCII standard
            try
            {
                decoder.decode( buffer );
            }
            catch( Exception e )
            {
                fail( "The " + SimpleStringEncoder.class.getName()
                    + " implementation is not generating ASCII simple strings" );
            }
        }
 
        for( int i = 0 ; i < NUMBER_OF_TRYS ; i++ )
        // Check if the returned decoded string is the same as the recived
            assertEquals( "The decoded string is not the same as the "
                + "originally encoded, index: " + i , new String(complexStrings[ i ]) , encoder
                .decodeComplex( simpleStrings[ i ] ) );
 
    }

This test always fails, I tryed everything, both string before and after the conversation print the same sequence of characters, the getBytes() method returns exactly the same array. I'm really pissed for this, hope it can be resolved quickly.

[Campbell Ritchie]

Please check your displayed name "camelo" against the JavaRanch naming policy and change it to comply.

What does the assertEquals() method do?

[Garrett Rowe]

I can only conclude that there is some subset of characters that your encoding method doesn't work as expected on. It has nothing to do with Charsets or anything like that. I haven't figured out what characters it's failing on, but there is definitely a difference between the original String and the decoded String. Here's the tests I used:

   @Test
    public void testRandomString()//Fails 99% of the time, passes 1% of the time
    {
    String s = makeRandomString();
    testEncode(s);
    }
    
    @Test
    public void testSimpleString()//Always passes
    {
    String s = "Hello World!";
    testEncode(s);
    }
    
    private void testEncode(String s)
    {
    String encoded = encoder.encodeSimple(s);
    assertFalse(encoded.equals(s));
    String decoded = encoder.decodeComplex(encoded);
    assertEquals("The decoded String does not match the original String",
              s, decoded);    
    }
 
    private String makeRandomString()
    {
Random r = new Random();
    byte[] bytes = new byte[100];
    r.nextBytes(bytes);
    return new String(bytes);
    }

[ April 20, 2008: Message edited by: Garrett Rowe ]
[ April 20, 2008: Message edited by: Garrett Rowe ]