1 /*
2  * Copyright (C) 2011 The Android Open Source Project
3  *
4  * Licensed under the Apache License, Version 2.0 (the "License");
5  * you may not use this file except in compliance with the License.
6  * You may obtain a copy of the License at
7  *
8  *      http://www.apache.org/licenses/LICENSE-2.0
9  *
10  * Unless required by applicable law or agreed to in writing, software
11  * distributed under the License is distributed on an "AS IS" BASIS,
12  * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
13  * See the License for the specific language governing permissions and
14  * limitations under the License.
15  */
16 
17 package com.android.dex;
18 
19 import com.android.dex.util.ByteInput;
20 import java.io.UTFDataFormatException;
21 
22 /**
23  * Modified UTF-8 as described in the dex file format spec.
24  *
25  * <p>Derived from libcore's MUTF-8 encoder at java.nio.charset.ModifiedUtf8.
26  */
27 public final class Mutf8 {
Mutf8()28     private Mutf8() {}
29 
30     /**
31      * Decodes bytes from {@code in} into {@code out} until a delimiter 0x00 is
32      * encountered. Returns a new string containing the decoded characters.
33      */
decode(ByteInput in, char[] out)34     public static String decode(ByteInput in, char[] out) throws UTFDataFormatException {
35         int s = 0;
36         while (true) {
37             char a = (char) (in.readByte() & 0xff);
38             if (a == 0) {
39                 return new String(out, 0, s);
40             }
41             out[s] = a;
42             if (a < '\u0080') {
43                 s++;
44             } else if ((a & 0xe0) == 0xc0) {
45                 int b = in.readByte() & 0xff;
46                 if ((b & 0xC0) != 0x80) {
47                     throw new UTFDataFormatException("bad second byte");
48                 }
49                 out[s++] = (char) (((a & 0x1F) << 6) | (b & 0x3F));
50             } else if ((a & 0xf0) == 0xe0) {
51                 int b = in.readByte() & 0xff;
52                 int c = in.readByte() & 0xff;
53                 if (((b & 0xC0) != 0x80) || ((c & 0xC0) != 0x80)) {
54                     throw new UTFDataFormatException("bad second or third byte");
55                 }
56                 out[s++] = (char) (((a & 0x0F) << 12) | ((b & 0x3F) << 6) | (c & 0x3F));
57             } else {
58                 throw new UTFDataFormatException("bad byte");
59             }
60         }
61     }
62 
63     /**
64      * Returns the number of bytes the modified UTF8 representation of 's' would take.
65      */
countBytes(String s, boolean shortLength)66     private static long countBytes(String s, boolean shortLength) throws UTFDataFormatException {
67         long result = 0;
68         final int length = s.length();
69         for (int i = 0; i < length; ++i) {
70             char ch = s.charAt(i);
71             if (ch != 0 && ch <= 127) { // U+0000 uses two bytes.
72                 ++result;
73             } else if (ch <= 2047) {
74                 result += 2;
75             } else {
76                 result += 3;
77             }
78             if (shortLength && result > 65535) {
79                 throw new UTFDataFormatException("String more than 65535 UTF bytes long");
80             }
81         }
82         return result;
83     }
84 
85     /**
86      * Encodes the modified UTF-8 bytes corresponding to {@code s} into  {@code
87      * dst}, starting at {@code offset}.
88      */
encode(byte[] dst, int offset, String s)89     public static void encode(byte[] dst, int offset, String s) {
90         final int length = s.length();
91         for (int i = 0; i < length; i++) {
92             char ch = s.charAt(i);
93             if (ch != 0 && ch <= 127) { // U+0000 uses two bytes.
94                 dst[offset++] = (byte) ch;
95             } else if (ch <= 2047) {
96                 dst[offset++] = (byte) (0xc0 | (0x1f & (ch >> 6)));
97                 dst[offset++] = (byte) (0x80 | (0x3f & ch));
98             } else {
99                 dst[offset++] = (byte) (0xe0 | (0x0f & (ch >> 12)));
100                 dst[offset++] = (byte) (0x80 | (0x3f & (ch >> 6)));
101                 dst[offset++] = (byte) (0x80 | (0x3f & ch));
102             }
103         }
104     }
105 
106     /**
107      * Returns an array containing the <i>modified UTF-8</i> form of {@code s}.
108      */
encode(String s)109     public static byte[] encode(String s) throws UTFDataFormatException {
110         int utfCount = (int) countBytes(s, true);
111         byte[] result = new byte[utfCount];
112         encode(result, 0, s);
113         return result;
114     }
115 }
116