rs.getBytes() rarity

[Jonas Wille]

We have a RDB database with one column in utf-8 encoding for Chinese text. Now I want to know how many bytes the text takes. Well, we know how much it takes, but we want to see it in software.

So, I use the getBytes()-method of the ResultSet. For all Chinese characters it returns 2, but I doubt this is correct because when I use the getString().getBytes() of the ResultSet it returns 3 and with the SQL function LENGTHB() the result is also 3.

Finally, I try to make a String from the bytearray that is returned by the getBytes()-method of the ResultSet and compared it with the one that is returned by the getString() from the ResultSet.
If the text isn't Chinese, the 2 Strings are equals. But otherwise they aren't.


Is it possible that the getBytes() method of the ResultSet isn't correct? (Driver is a JDBC-RDB driver)

Hope you can understand what i'm trying to say.

ResultSet rs = statement.executeQuery("SELECT RTASK, TRIM(OTASK) AS OTASK, LENGTHB(TRIM(OTASK)) AS BYTECOUNT FROM TSYTEST;");
while (rs.next()) {
	byte[] otaskbytes = rs.getBytes("OTASK");
	String otaskstring = rs.getString("OTASK");

	System.out.printf("%s\t%s\t%s\t%s\t%s\n", 
					otaskstring,
					otaskbytes.length, 
					otaskstring.getBytes().length,
					rs.getInt("BYTECOUNT"),
					new String(otaskbytes)
);
}

output

test     4  4  4  test
qwerty   6  6  6  qwerty
零       2  3  3  �