Java char变量

Question

The following code within a program allows 90 to be assigned to the variable 'ch'. 'Z' is then printed to the console.

char ch;
ch = 90;
System.out.println(ch);

However, the following code, that lies within a program, does not compile. If the following code requires the input to the ch variable to be a character type, ie (char) System.in.read(); , then why does the same not apply when 90 is assigned to ch above? Why doesn't it have to be ch = (char) 90 ?

char ch;
ch = System.in.read();

Answer 1

The compiler knows that 90 is a valid value for char . However, System.in.read() can return any int , which may be out of the valid range for char s.

If you change 90 to 90000, the code won't compile:

 char ch;
 ch = 90000;

Answer 2

Whenever you are dealing with system io you need to ensure that you can handle all valid byte input values. However, you then need a mechanism to indicate that the stream has been exhausted. The javadoc for InputStream.read() ( System.in is a global InputStream ) says (emphasis added),

Reads the next byte of data from the input stream. The value byte is returned as an int in the range 0 to 255. If no byte is available because the end of the stream has been reached, the value -1 is returned.

If you were to cast -1 to char you would get 65535 because char is unsigned. With byte , it's even worse in that -1 is a valid value. Regardless, you aren't reading char values; you are reading byte (s) encoded as int in the range 0-255 plus -1 to indicate end of stream. If you want char (s) I suggest you look at an InputStreamReader which is described in the javadoc as

An InputStreamReader is a bridge from byte streams to character streams