如何在Java中将字符转换为整数？

Question

I am a beginner at Java, trying to figure out how to convert characters from a text file into integers. In the process, I wrote a program which generates a text file showing what characters are generated by what integers.

package numberchars;

import java.io.FileWriter;
import java.io.IOException;
import java.io.FileReader;
import java.lang.Character;

public class Numberchars {

    public static void main(String[] args) throws IOException {

FileWriter outputStream = new FileWriter("NumberChars.txt");

//Write to the output file the char corresponding to the decimal
// from 1 to 255
int counter = 1;

while (counter <256)
{
outputStream.write(counter);

outputStream.flush();
counter++;
}
outputStream.close();

This generated NumberChars.txt, which had all the numbers, all the letters both upper and lower case, surrounded at each end by other symbols and glyphs.

Then I tried to read this file and convert its characters back into integers:

FileReader inputStream = new FileReader("NumberChars.txt");
FileWriter outputStream2 = new FileWriter ("CharNumbers.txt");
int c;

while ((c = inputStream.read()) != -1)
            {
                outputStream2.write(Character.getNumericValue(c));
                outputStream2.flush();
            }  

}


}

The resulting file, CharNumbers.txt , began with the same glyphs as NumberChars.txt but then was blank. Opening the files in MS Word, I found NumberChars had 248 characters (including 5 spaces) and CharNumbers had 173 (including 8 spaces).

So why didn't the Character.getNumericValue(c) result in an integer written to CharNumbers.txt? And given that it didn't, why at least didn't it write an exact copy of NumberChars.txt? Any help much appreciated.

Answer 1

Character.getNumericValue doesn't do what you think it does. If you read the Javadoc :

Returns the int value that the specified character (Unicode code point) represents. For example, the character '\Ⅼ' (the Roman numeral fifty) will return an int with a value of 50 .

On error it returns -1 (which looks like 0xFF_FF_FF_FF in 2s complement).

Most characters don't have such a "numeric value," so you write the ints out, each padded to 2 bytes (more on that later), read them back in the same way, and then start writing a whole lot of 0xFFFF ( -1 truncated to 2 bytes) courtesy of a misplaced Character.getNumericValue . I'm not sure what MS Word is doing, but it's probably getting confused what the encoding of your file is and glomming all those bytes into 0xFF_FF_FF_FF (because the high bits of each byte are set) and treating that as one character. (Use a text editor more suited to this kind of stuff like Notepad++, btw.) If you were to measure your file's size on disk in bytes it will probably still be 256 chars * 2 bytes/chars = 512 bytes .

I'm not sure what you meant to do here, so I'll note that InputStreamReader and OutputStreamWriter work on a (Unicode) character basis, with an encoder that defaults to the system one. That's why your ints are padded/truncated to 2 bytes. If you wanted pure byte IO, use FileInputStream / FileOutputStream . If you wanted to read and write the int s as String s, you need to use FileWriter / FileReader , but not like you did.

// Just bytes
// This is a try-with-resources. It executes the code with the decls in it
// but is also like an implicit finally block that calls `close()` on each resource.
try(FileOutputStream fos = new FileOutputStream("bytes.bin")) {
    for(int b = 0; b < 256; b++) { // Bytes are signed so we use int.
          // This takes an int and truncates it for the lowest byte
          fos.write(b);
          // Can also fill a byte[] and dump it all at once with overloaded write.
    }
}
byte[] bytes = new bytes[256];
try(FileInputStream fis = new FileInputStream("bytes.bin")) {
    // Reads up to bytes.length bytes into bytes
    fis.read(bytes);
}

// Foreach loop. If you don't know what this does, I think you can figure out from the name.
for(byte b : bytes) {
    System.out.println(b);
}

// As Strings
try(FileWriter fw = new FileWriter("strings.txt")) {
    for(int i = 0; i < 256; i++) {
        // You need a delimiter lest you not be able to tell 12 from 1,2 when you read
        // Uses system default encoding
        fw.write(Integer.toString(i) + "\n");
    }
}
byte[] bytes = new byte[256];
try(
    FileReader fr = new FileReader("strings.txt");
    // FileReaders can't do stuff like "read one line to String" so we wrap it
    BufferedReader br = new BufferedReader(fr);
) {
    for(int i = 0; i < 256; i++) {
        bytes[i] = Byte.valueOf(br.readLine());
    }
}

for(byte b : bytes) {
    System.out.println(b);
}

Answer 2

public class MyCLAss {

    public static void main(String[] args) 
    {
        char x='b';
        System.out.println(+x);//just by witting a plus symbol before the variable you can find it's ascii value....it will give 98.
    }

}