Blogger Scott

一个utf8转换程序

据说是一个通用的手机上使用的UTF8转换程序，先记下来。

 1private final String readUnicodeFileUTF8(String filename) {
 2        StringBuffer sb = new StringBuffer(256);
 3        try {
 4            int[] surrogatePair = new int[2];
 5            InputStream is = this.getClass().getResourceAsStream(filename);
 6
 7            int val = 0;
 8            int unicharCount = 0;
 9            while ((val = readNextCharFromStreamUTF8(is))!=-1) {
10                unicharCount++;
11                if (val <= 0xFFFF) {
12                    // if first value is the Byte Order Mark (BOM), do not add
13                    if (! (unicharCount == 1 && val == 0xFEFF)) {
14                        sb.append((char)val);
15                    }
16                } else {
17                    supplementCodePointToSurrogatePair(val, surrogatePair);
18                    sb.append((char)surrogatePair[0]);
19                    sb.append((char)surrogatePair[1]);
20                }
21            }
22            is.close();
23        } catch (Exception e) {};
24
25        return new String(sb);
26    }
27   
28    private final static int readNextCharFromStreamUTF8(InputStream is) {
29        int c = -1;
30        if (is==null) return c;
31        boolean complete = false;
32       
33        try {
34            int byteVal;
35            int expecting=0;
36            int composedVal=0;
37           
38            while (!complete && (byteVal = is.read()) != -1) {
39                if (expecting > 0 && (byteVal & 0xC0) == 0x80) {  /* 10xxxxxx */
40                    expecting--;
41                    composedVal = composedVal | ((byteVal & 0x3F) << (expecting*6));
42                    if (expecting == 0) {
43                        c = composedVal;
44                        complete = true;
45                        //System.out.println("appending: U+" + Integer.toHexString(composedVal) );
46                    }
47                } else {
48                    composedVal = 0;
49                    expecting = 0;
50                    if ((byteVal & 0x80) == 0) {    /* 0xxxxxxx */
51                        // one byte character, no extending byte expected
52                        c = byteVal;
53                        complete = true;
54                        //System.out.println("appending: U+" + Integer.toHexString(byteVal) );
55                    } else if ((byteVal & 0xE0) == 0xC0) {  /* 110xxxxx */
56                        expecting = 1;  // expecting 1 extending byte
57                        composedVal = ((byteVal & 0x1F) << 6);
58                    } else if ((byteVal & 0xF0) == 0xE0) {  /* 1110xxxx */
59                        expecting = 2;  // expecting 2 extending bytes
60                        composedVal = ((byteVal & 0x0F) << 12);
61                    } else if ((byteVal & 0xF8) == 0xF0) {  /* 11110xxx */
62                        expecting = 3;  // expecting 3 extending bytes
63                        composedVal = ((byteVal & 0x07) << 18);
64                    } else {
65                        // non conformant utf-8, ignore or catch error
66                    }
67                }
68            }
69           
70        } catch (Exception e) {
71            System.out.println(e.toString());
72        }
73       
74        return c;
75    }
76
77    private final static void supplementCodePointToSurrogatePair(int codePoint, int[] surrogatePair) {
78        int high4 = ((codePoint >> 16) & 0x1F) - 1;
79        int mid6 = ((codePoint >> 10) & 0x3F);
80        int low10 = codePoint & 0x3FF;
81
82        surrogatePair[0] = (0xD800 | (high4 << 6) | (mid6));
83        surrogatePair[1] = (0xDC00 | (low10));
84    }

posted on 2009-06-07 16:37 江天部落格 阅读(285) 评论(0)  编辑  收藏 所属分类: Android 、Java


只有注册用户登录后才能发表评论。
网站导航: