essay18.txt

Java Essay Serials 1 - Java Basics - 18, Can a char type variable store a chinese character?
Basically the answer is YES, a char type variable can store a chinese character.
The reason is that Java supports Unicode, so multibyte-character such as chinese character is not a problem. 

But we have to know, Unicode contains 17 planes, usually we use the plane 0, which the code range from u0000-uFFFF. We call this BMP (Basic Multilingual Plane). Theoreotically plane 0 can contain 65536 characters, but the actual amount is less than 65536 as the code range uD800-uDFFF is reserved for extension purposes. That means BMP can not represent all characters, for example,some ancient chinese charaters is out of this range.

Java adopts UTF-16 to store Unicode characters. Accroding to the standard ISO/IEC 10646-1, UTF-16 uses 2 bytes to store a BMP Unicode character, and for those characters which code are defined in the range u10000-u10FFFF, UTF-16 uses 4 bytes to store.

In Java, the length of a char type variable is 2 bytes, so a char type variable can store a UTF-16 encoded BMP character. As most of chinese characters are defined in the range from 0x4E00 to 0x9FBB, so Java works with chinese well. 
for example:
	char c1 = 'A';
	char c2 = 0x41;
	char c3 = '\u0041';
	char c4 = '郭';
	char c5 = '\u90ED';
	System.out.println(c1);
	System.out.println(c2);
	System.out.println(c3);
	System.out.println(c4);
	System.out.println(c5);
This program outputs:
A
A
A
郭
郭
We can see that chinese character works properly.

We now use another example to print the internal representation of a String:
	String s = "Author 郭";
	byte[] bytes = s.getBytes("UTF-16");
	for (int i = 0; i < bytes.length; i++) { 
		System.out.format("%02X ",bytes[i]);
        }  
The result would be:
FE FF 00 41 00 75 00 74 00 68 00 6F 00 72 00 20 90 ED
The last 2 bytes - 90 ED - is the UTF-16 code for chinese character '郭'.

For those extension characters in other planes, prior to JDK1.5, Java does not support, now Java uses 2 characters to represent them (refer to JSP 204).
for example:
	String s = "Hanzi 𠀡";
	byte[] bytes = s.getBytes("UTF-16");
	for (int i = 0; i < bytes.length; i++) { 
		System.out.format("%02X ",bytes[i]);
        }  
The result would be:
FE FF 00 48 00 61 00 6E 00 7A 00 69 00 20 D8 40 DC 21
The last 4 bytes - D8 40 DC 21 - is the UTF-16 code for the chinese character '𠀡'.
This chinese character can not be stored with a Java char. If we write the code like this:
char c= '𠀡';
or
char c='\u20021';
The Java compiler would prompt an error:Invalid character constant.

A program usually processes BMP characters, so we can simply say that a char type variable stores a chinese character.