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 #ifndef ART_LIBDEXFILE_DEX_UTF_H_
18 #define ART_LIBDEXFILE_DEX_UTF_H_
19
20 #include "base/macros.h"
21
22 #include <stddef.h>
23 #include <stdint.h>
24
25 #include <string>
26
27 /*
28 * All UTF-8 in art is actually modified UTF-8. Mostly, this distinction
29 * doesn't matter.
30 *
31 * See http://en.wikipedia.org/wiki/UTF-8#Modified_UTF-8 for the details.
32 */
33 namespace art {
34
35 /*
36 * Returns the number of UTF-16 characters in the given modified UTF-8 string.
37 */
38 size_t CountModifiedUtf8Chars(const char* utf8);
39 size_t CountModifiedUtf8Chars(const char* utf8, size_t byte_count);
40
41 /*
42 * Returns the number of modified UTF-8 bytes needed to represent the given
43 * UTF-16 string.
44 */
45 size_t CountUtf8Bytes(const uint16_t* chars, size_t char_count);
46
47 /*
48 * Convert from Modified UTF-8 to UTF-16.
49 */
50 void ConvertModifiedUtf8ToUtf16(uint16_t* utf16_out, const char* utf8_in);
51 void ConvertModifiedUtf8ToUtf16(uint16_t* utf16_out, size_t out_chars,
52 const char* utf8_in, size_t in_bytes);
53
54 /*
55 * Compare two modified UTF-8 strings as UTF-16 code point values in a non-locale sensitive manner
56 */
57 ALWAYS_INLINE int CompareModifiedUtf8ToModifiedUtf8AsUtf16CodePointValues(const char* utf8_1,
58 const char* utf8_2);
59
60 /*
61 * Compare a null-terminated modified UTF-8 string with a UTF-16 string (not null-terminated)
62 * as code point values in a non-locale sensitive manner.
63 */
64 int CompareModifiedUtf8ToUtf16AsCodePointValues(const char* utf8, const uint16_t* utf16,
65 size_t utf16_length);
66
67 /*
68 * Convert from UTF-16 to Modified UTF-8. Note that the output is _not_
69 * NUL-terminated. You probably need to call CountUtf8Bytes before calling
70 * this anyway, so if you want a NUL-terminated string, you know where to
71 * put the NUL byte.
72 */
73 void ConvertUtf16ToModifiedUtf8(char* utf8_out, size_t byte_count,
74 const uint16_t* utf16_in, size_t char_count);
75
76 /*
77 * The java.lang.String hashCode() algorithm.
78 */
79 template<typename MemoryType>
ComputeUtf16Hash(const MemoryType * chars,size_t char_count)80 int32_t ComputeUtf16Hash(const MemoryType* chars, size_t char_count) {
81 uint32_t hash = 0;
82 while (char_count--) {
83 hash = hash * 31 + *chars++;
84 }
85 return static_cast<int32_t>(hash);
86 }
87
88 int32_t ComputeUtf16HashFromModifiedUtf8(const char* utf8, size_t utf16_length);
89
90 // Compute a hash code of a modified UTF-8 string. Not the standard java hash since it returns a
91 // uint32_t and hashes individual chars instead of codepoint words.
92 uint32_t ComputeModifiedUtf8Hash(const char* chars);
93
94 /*
95 * Retrieve the next UTF-16 character or surrogate pair from a UTF-8 string.
96 * single byte, 2-byte and 3-byte UTF-8 sequences result in a single UTF-16
97 * character (possibly one half of a surrogate) whereas 4-byte UTF-8 sequences
98 * result in a surrogate pair. Use GetLeadingUtf16Char and GetTrailingUtf16Char
99 * to process the return value of this function.
100 *
101 * Advances "*utf8_data_in" to the start of the next character.
102 *
103 * WARNING: If a string is corrupted by dropping a '\0' in the middle
104 * of a multi byte sequence, you can end up overrunning the buffer with
105 * reads (and possibly with the writes if the length was computed and
106 * cached before the damage). For performance reasons, this function
107 * assumes that the string being parsed is known to be valid (e.g., by
108 * already being verified). Most strings we process here are coming
109 * out of dex files or other internal translations, so the only real
110 * risk comes from the JNI NewStringUTF call.
111 */
112 uint32_t GetUtf16FromUtf8(const char** utf8_data_in);
113
114 /**
115 * Gets the leading UTF-16 character from a surrogate pair, or the sole
116 * UTF-16 character from the return value of GetUtf16FromUtf8.
117 */
118 ALWAYS_INLINE uint16_t GetLeadingUtf16Char(uint32_t maybe_pair);
119
120 /**
121 * Gets the trailing UTF-16 character from a surrogate pair, or 0 otherwise
122 * from the return value of GetUtf16FromUtf8.
123 */
124 ALWAYS_INLINE uint16_t GetTrailingUtf16Char(uint32_t maybe_pair);
125
126 // Returns a printable (escaped) version of a character.
127 std::string PrintableChar(uint16_t ch);
128
129 // Returns an ASCII string corresponding to the given UTF-8 string.
130 // Java escapes are used for non-ASCII characters.
131 std::string PrintableString(const char* utf8);
132
133 } // namespace art
134
135 #endif // ART_LIBDEXFILE_DEX_UTF_H_
136