|
| 1 | +/* |
| 2 | + * Copyright (C) 2005 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 ANDROID_UNICODE_H |
| 18 | +#define ANDROID_UNICODE_H |
| 19 | + |
| 20 | +#define __STDC_LIMIT_MACROS 1 |
| 21 | + |
| 22 | +#include <sys/types.h> |
| 23 | +#include <stdint.h> |
| 24 | + |
| 25 | +extern "C" { |
| 26 | + |
| 27 | +// char32_t and char16_t are built-in types as of c++0x. |
| 28 | +#if !defined(__GXX_EXPERIMENTAL_CXX0X__) && __cplusplus < 201103L |
| 29 | +typedef uint32_t char32_t; |
| 30 | +typedef uint16_t char16_t; |
| 31 | +#endif |
| 32 | + |
| 33 | +// Standard string functions on char16_t strings. |
| 34 | +int strcmp16(const char16_t *, const char16_t *); |
| 35 | +int strncmp16(const char16_t *s1, const char16_t *s2, size_t n); |
| 36 | +size_t strlen16(const char16_t *); |
| 37 | +size_t strnlen16(const char16_t *, size_t); |
| 38 | +char16_t *strcpy16(char16_t *, const char16_t *); |
| 39 | +char16_t *strncpy16(char16_t *, const char16_t *, size_t); |
| 40 | + |
| 41 | +// Version of comparison that supports embedded nulls. |
| 42 | +// This is different than strncmp() because we don't stop |
| 43 | +// at a nul character and consider the strings to be different |
| 44 | +// if the lengths are different (thus we need to supply the |
| 45 | +// lengths of both strings). This can also be used when |
| 46 | +// your string is not nul-terminated as it will have the |
| 47 | +// equivalent result as strcmp16 (unlike strncmp16). |
| 48 | +int strzcmp16(const char16_t *s1, size_t n1, const char16_t *s2, size_t n2); |
| 49 | + |
| 50 | +// Version of strzcmp16 for comparing strings in different endianness. |
| 51 | +int strzcmp16_h_n(const char16_t *s1H, size_t n1, const char16_t *s2N, size_t n2); |
| 52 | + |
| 53 | +// Standard string functions on char32_t strings. |
| 54 | +size_t strlen32(const char32_t *); |
| 55 | +size_t strnlen32(const char32_t *, size_t); |
| 56 | + |
| 57 | +/** |
| 58 | + * Measure the length of a UTF-32 string in UTF-8. If the string is invalid |
| 59 | + * such as containing a surrogate character, -1 will be returned. |
| 60 | + */ |
| 61 | +ssize_t utf32_to_utf8_length(const char32_t *src, size_t src_len); |
| 62 | + |
| 63 | +/** |
| 64 | + * Stores a UTF-8 string converted from "src" in "dst", if "dst_length" is not |
| 65 | + * large enough to store the string, the part of the "src" string is stored |
| 66 | + * into "dst" as much as possible. See the examples for more detail. |
| 67 | + * Returns the size actually used for storing the string. |
| 68 | + * dst" is not null-terminated when dst_len is fully used (like strncpy). |
| 69 | + * |
| 70 | + * Example 1 |
| 71 | + * "src" == \u3042\u3044 (\xE3\x81\x82\xE3\x81\x84) |
| 72 | + * "src_len" == 2 |
| 73 | + * "dst_len" >= 7 |
| 74 | + * -> |
| 75 | + * Returned value == 6 |
| 76 | + * "dst" becomes \xE3\x81\x82\xE3\x81\x84\0 |
| 77 | + * (note that "dst" is null-terminated) |
| 78 | + * |
| 79 | + * Example 2 |
| 80 | + * "src" == \u3042\u3044 (\xE3\x81\x82\xE3\x81\x84) |
| 81 | + * "src_len" == 2 |
| 82 | + * "dst_len" == 5 |
| 83 | + * -> |
| 84 | + * Returned value == 3 |
| 85 | + * "dst" becomes \xE3\x81\x82\0 |
| 86 | + * (note that "dst" is null-terminated, but \u3044 is not stored in "dst" |
| 87 | + * since "dst" does not have enough size to store the character) |
| 88 | + * |
| 89 | + * Example 3 |
| 90 | + * "src" == \u3042\u3044 (\xE3\x81\x82\xE3\x81\x84) |
| 91 | + * "src_len" == 2 |
| 92 | + * "dst_len" == 6 |
| 93 | + * -> |
| 94 | + * Returned value == 6 |
| 95 | + * "dst" becomes \xE3\x81\x82\xE3\x81\x84 |
| 96 | + * (note that "dst" is NOT null-terminated, like strncpy) |
| 97 | + */ |
| 98 | +void utf32_to_utf8(const char32_t* src, size_t src_len, char* dst); |
| 99 | + |
| 100 | +/** |
| 101 | + * Returns the unicode value at "index". |
| 102 | + * Returns -1 when the index is invalid (equals to or more than "src_len"). |
| 103 | + * If returned value is positive, it is able to be converted to char32_t, which |
| 104 | + * is unsigned. Then, if "next_index" is not NULL, the next index to be used is |
| 105 | + * stored in "next_index". "next_index" can be NULL. |
| 106 | + */ |
| 107 | +int32_t utf32_from_utf8_at(const char *src, size_t src_len, size_t index, size_t *next_index); |
| 108 | + |
| 109 | + |
| 110 | +/** |
| 111 | + * Returns the UTF-8 length of UTF-16 string "src". |
| 112 | + */ |
| 113 | +ssize_t utf16_to_utf8_length(const char16_t *src, size_t src_len); |
| 114 | + |
| 115 | +/** |
| 116 | + * Converts a UTF-16 string to UTF-8. The destination buffer must be large |
| 117 | + * enough to fit the UTF-16 as measured by utf16_to_utf8_length with an added |
| 118 | + * NULL terminator. |
| 119 | + */ |
| 120 | +void utf16_to_utf8(const char16_t* src, size_t src_len, char* dst); |
| 121 | + |
| 122 | +/** |
| 123 | + * Returns the length of "src" when "src" is valid UTF-8 string. |
| 124 | + * Returns 0 if src is NULL or 0-length string. Returns -1 when the source |
| 125 | + * is an invalid string. |
| 126 | + * |
| 127 | + * This function should be used to determine whether "src" is valid UTF-8 |
| 128 | + * characters with valid unicode codepoints. "src" must be null-terminated. |
| 129 | + * |
| 130 | + * If you are going to use other utf8_to_... functions defined in this header |
| 131 | + * with string which may not be valid UTF-8 with valid codepoint (form 0 to |
| 132 | + * 0x10FFFF), you should use this function before calling others, since the |
| 133 | + * other functions do not check whether the string is valid UTF-8 or not. |
| 134 | + * |
| 135 | + * If you do not care whether "src" is valid UTF-8 or not, you should use |
| 136 | + * strlen() as usual, which should be much faster. |
| 137 | + */ |
| 138 | +ssize_t utf8_length(const char *src); |
| 139 | + |
| 140 | +/** |
| 141 | + * Measure the length of a UTF-32 string. |
| 142 | + */ |
| 143 | +size_t utf8_to_utf32_length(const char *src, size_t src_len); |
| 144 | + |
| 145 | +/** |
| 146 | + * Stores a UTF-32 string converted from "src" in "dst". "dst" must be large |
| 147 | + * enough to store the entire converted string as measured by |
| 148 | + * utf8_to_utf32_length plus space for a NULL terminator. |
| 149 | + */ |
| 150 | +void utf8_to_utf32(const char* src, size_t src_len, char32_t* dst); |
| 151 | + |
| 152 | +/** |
| 153 | + * Returns the UTF-16 length of UTF-8 string "src". |
| 154 | + */ |
| 155 | +ssize_t utf8_to_utf16_length(const uint8_t* src, size_t srcLen); |
| 156 | + |
| 157 | +/** |
| 158 | + * Convert UTF-8 to UTF-16 including surrogate pairs. |
| 159 | + * Returns a pointer to the end of the string (where a null terminator might go |
| 160 | + * if you wanted to add one). |
| 161 | + */ |
| 162 | +char16_t* utf8_to_utf16_no_null_terminator(const uint8_t* src, size_t srcLen, char16_t* dst); |
| 163 | + |
| 164 | +/** |
| 165 | + * Convert UTF-8 to UTF-16 including surrogate pairs. The destination buffer |
| 166 | + * must be large enough to hold the result as measured by utf8_to_utf16_length |
| 167 | + * plus an added NULL terminator. |
| 168 | + */ |
| 169 | +void utf8_to_utf16(const uint8_t* src, size_t srcLen, char16_t* dst); |
| 170 | + |
| 171 | +} |
| 172 | + |
| 173 | +#undef __STDC_LIMIT_MACROS |
| 174 | +#endif |
0 commit comments