这段时间将项目中一个模块参照C++源代码，实现一个JAVA版。主要功能是将一些字段信息转换为String类型，传输后可以进行解析。
Integer、Long转为String，Java本身提供了这种转换方法。
Example:
int a = 127;
String s = new String(a);
System.out.println(s);
//输出 127
而我的项目需求是Integer、Long字段在String类型中所占大小为固定自己。比如Integer类型的127在内存中应该为0x7f,0x00,0x00,0x00(Little Endian)，在String类型输出可能为乱码，但需要在解析后可以还原为Integer类型的127。
C++中存在无符号数，而JAVA中没有无符号数。所以以byte[]替代unsigned char[]会出现小问题。见下：
int n = 0xff7f0012;
byte[] b = new byte[4];
b[3] = (byte) (n & 0xff);
b[2] = (byte) (n >> 8 & 0xff);
b[1] = (byte) (n >> 16 & 0xff);
b[0] = (byte) (n >> 24 & 0xff);
String s = new String(b);
for (byte i : b) {
System.out.printf("%02x ", i);
}
System.out.println();
System.out.println(s.length());
for (byte i : s.getBytes()) {
System.out.printf("%02x ", i);
}
//输出如下
ff 7f 00 12
3
3f 00 12
由上可见，byte[]转换String类型时，会发生问题，应为byte表示的范围是-128~127，当byte[]中元素大于127时，会被转换成63。尝试使用不同的编码方案解决。
测试发现UTF-8,GBK,US-ASCII均未得到正确结果。尝试charset为8859_1得到正确结果。
将上面程序中的String s = new String(b);替换为下面程序段：
String s = "";
try {
s = new String(b, "8859_1");
} catch (UnsupportedEncodingException e) {
// TODO Auto-generated catch block
e.printStackTrace();
}
即可得到正确结果，下面是程序功能的简单封装。
public class Test{
private static byte[] intToByte(int n) {
byte[] b = new byte[4];
b[3] = (byte) (n & 0xff);
b[2] = (byte) (n >> 8 & 0xff);
b[1] = (byte) (n >> 16 & 0xff);
b[0] = (byte) (n >> 24 & 0xff);
return b;
}
private static byte[] longToByte(long n) {
byte[] b = new byte[8];
b[7] = (byte) (n & 0xff);
b[6] = (byte) (n >> 8 & 0xff);
b[5] = (byte) (n >> 16 & 0xff);
b[4] = (byte) (n >> 24 & 0xff);
b[3] = (byte) (n >> 32 & 0xff);
b[2] = (byte) (n >> 40 & 0xff);
b[1] = (byte) (n >> 48 & 0xff);
b[0] = (byte) (n >> 56 & 0xff);
return b;
}
private static int byteToInt(byte[] b, int beginPos) {
return b[beginPos + 3] & 0xff |
(b[beginPos + 2] & 0xff) << 8 |
(b[beginPos + 1] & 0xff) << 16 |
(b[beginPos] & 0xff) << 24;
}
private static int byteToLong(byte[] b, int beginPos) {
return b[beginPos + 7] & 0xff |
(b[beginPos + 6] & 0xff) << 8 |
(b[beginPos + 5] & 0xff) << 16 |
(b[beginPos + 4] & 0xff) << 24 |
(b[beginPos + 3] & 0xff) << 32 |
(b[beginPos + 2] & 0xff) << 40 |
(b[beginPos + 1] & 0xff) << 48 |
(b[beginPos] & 0xff) << 56;
}
private static String intToString(int n) {
byte[] b = intToByte(n);
String s = "";
try {
s = new String(b, "8859_1");
} catch (UnsupportedEncodingException e) {
e.printStackTrace();
}
return new String(s);
}
private static String longToString(long n) {
byte[] b = longToByte(n);
String s = "";
try {
s = new String(b, "8859_1");
} catch (UnsupportedEncodingException e) {
e.printStackTrace();
}
return new String(s);
}
private static int stringToInt(String str) {
byte[] b = new byte[4];
try {
b = str.getBytes("8859_1");
} catch (UnsupportedEncodingException e) {
e.printStackTrace();
}
return byteToInt(b, 0);
}
private static int stringToLong(String str) {
byte[] b = new byte[8];
try {
b = str.getBytes("8859_1");
} catch (UnsupportedEncodingException e) {
e.printStackTrace();
}
return byteToLong(b, 0);
}
}