从Unicode到GB2312转换表制作程式

本章介绍了UnicodeGB2312.java源程序。它可以用来制作Unicode编码到GB2312编码转换 表。

在我发表了GB2312到Unicode的转换表以后,收到了许多读者信件,寻求Unicode到GB2312 的转换表。

下面的程式便可以用来制作这样的转换表。程式的输出结果收入下一章之中。

/* UnicodeGB2312.java
 - Copyright (c) 2016, HerongYang.com, All Rights Reserved.
 */
import java.io.*;
import java.nio.*;
import java.nio.charset.*;
class UnicodeGB2312 {
   static OutputStream out = null;
   static char hexDigit[] = {'0', '1', '2', '3', '4', '5', '6', '7',
                             '8', '9', 'A', 'B', 'C', 'D', 'E', 'F'};
   static int b_out[] = {201,267,279,293,484,587,625,657,734,782,827,
      874,901,980,1001,5590,8801};
   static int e_out[] = {216,268,280,294,494,594,632,694,748,794,836,
      894,903,994,1594,5594,9494};
   public static void main(String[] a) {
      try {
         out = new FileOutputStream("unicode_gb2312.gb");
         writeCode();
         out.close();
      } catch (IOException e) {
         System.out.println(e.toString());
      }
   }
   public static void writeCode() throws IOException {
      CharsetEncoder gbec = Charset.forName("GBK").newEncoder();
      char[] ca = new char[1];
      CharBuffer cb = null;
      ByteBuffer gbbb = null;
      writeHeader();
      int count = 0;
      for (int i=0; i<0x010000; i++) {
         ca[0] = (char) i;
         cb = CharBuffer.wrap(ca);
         try {
            gbbb = gbec.encode(cb);
         } catch (CharacterCodingException e) {
            gbbb = null;
         }
         if (validGB(gbbb)) {
            count++;
            writeHex((byte) (ca[0] >>> 8));
            writeHex((byte) (ca[0] & 0xff));
            writeString(" ");
            writeByteBuffer(gbbb,2);
            writeString(" ");
            writeByte(gbbb.get(0));
            writeByte(gbbb.get(1));
            if (count%5 == 0) writeln();
            else writeString("   ");
         }
      }
      if (count%5 != 0) writeln();
      writeFooter();
      System.out.println("Number of GB characters wrote: "+count);
   }
   public static boolean validGB(ByteBuffer gbbb) {
      if (gbbb==null) return false; 
      else if (gbbb.limit()!=2) return false; 
      else {
         byte hi = gbbb.get(0);
         byte lo = gbbb.get(1);
         if ((hi&0xFF)<0xA0) return false;
         if ((lo&0xFF)<0xA0) return false;
         int i = (hi&0xFF) - 0xA0;
         int j = (lo&0xFF) - 0xA0;
         if (i<1 || i>94) return false;
         if (j<1 || j>94) return false;
         for (int l=0; l<b_out.length; l++) {
            if (i*100+j>=b_out[l] && i*100+j<=e_out[l]) return false;
         }
      }
      return true;
   }
   public static void writeHeader() throws IOException {
      writeString("<pre>");
      writeln();
      writeString("Uni. GB   ");
      writeGBSpace();
      writeString("   ");
      writeString("Uni. GB   ");
      writeGBSpace();
      writeString("   ");
      writeString("Uni. GB   ");
      writeGBSpace();
      writeString("   ");
      writeString("Uni. GB   ");
      writeGBSpace();
      writeString("   ");
      writeString("Uni. GB   ");
      writeGBSpace();
      writeln();
      writeln();
   }
   public static void writeFooter() throws IOException {
      writeString("</pre>");
      writeln();
   }
   public static void writeln() throws IOException {
      out.write(0x0D);
      out.write(0x0A);
   }
   public static void writeGBSpace() throws IOException {
      out.write(0xA1);
      out.write(0xA1);
   }
   public static void writeByteBuffer(ByteBuffer b, int l)
      throws IOException {
      int i = 0;
      if (b==null) {
      	 writeString("null");
      	 i = 2;
      } else {
	for (i=0; i<b.limit(); i++) writeHex(b.get(i));
      }
      for (int j=i; j<l; j++) writeString("  ");
   }
   public static void writeString(String s) throws IOException {
      if (s!=null) {
         for (int i=0; i<s.length(); i++) {
            out.write((int) (s.charAt(i) & 0xFF));
         }
      }         
   }
   public static void writeHex(byte b) throws IOException {
      out.write((int) hexDigit[(b >> 4) & 0x0F]);
      out.write((int) hexDigit[b & 0x0F]);
   }
   public static void writeByte(byte b) throws IOException {
      out.write(b & 0xFF);
   }
}

上面的程式发表后,又有许多读者来信要求对程式加以说明,以便理解。其实这个程式的 逻辑很简单,阅读时仅需注意以下几点:

一, Unicode字符集的全体编码都在0x0000和0xFFFF之间,所以子程式writeCode()使用 了一个循环复句,以变量i走遍了Unicode的全体可能编码。

二,把单个Unicode编码转换成GB2312编码的关键语句是:gbec.encode(cb),它使用了 JDK中CharsetEncoder的中文编码功能。注意,GBK是由GB2312扩张而成。JDK只提供GBK编 码功能。

三,由于Unicode字符集比GB2312大,gbec.encode(cb)输出的编码有许多是废码,或者是 GBK的扩张码,所以要用子程式validGB()进行验证。

四,程式的其它部分主要是用于输出的列表制作。

Table of Contents

 说明与摘要

 国标GB2312字符集和编码说明

 从GB2312到Unicode转换表制作程式

 GB2312符号区:01-09区

 GB2312一级汉字区: 16-55区

 GB2312二级汉字区: 56-87区

从Unicode到GB2312转换表制作程式

 Unicode编码中的GB2312汉字: 7,445个

 参考文献

 PDF打印版