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 #include <sys/types.h> 21 #include <stdint.h> 22 23 extern "C" { 24 25 // Standard string functions on char16_t strings. 26 int strcmp16(const char16_t *, const char16_t *); 27 int strncmp16(const char16_t *s1, const char16_t *s2, size_t n); 28 size_t strlen16(const char16_t *); 29 size_t strnlen16(const char16_t *, size_t); 30 char16_t *strcpy16(char16_t *, const char16_t *); 31 char16_t *strncpy16(char16_t *, const char16_t *, size_t); 32 char16_t *strstr16(const char16_t*, const char16_t*); 33 34 // Version of comparison that supports embedded nulls. 35 // This is different than strncmp() because we don't stop 36 // at a nul character and consider the strings to be different 37 // if the lengths are different (thus we need to supply the 38 // lengths of both strings). This can also be used when 39 // your string is not nul-terminated as it will have the 40 // equivalent result as strcmp16 (unlike strncmp16). 41 int strzcmp16(const char16_t *s1, size_t n1, const char16_t *s2, size_t n2); 42 43 // Version of strzcmp16 for comparing strings in different endianness. 44 int strzcmp16_h_n(const char16_t *s1H, size_t n1, const char16_t *s2N, size_t n2); 45 46 // Standard string functions on char32_t strings. 47 size_t strlen32(const char32_t *); 48 size_t strnlen32(const char32_t *, size_t); 49 50 /** 51 * Measure the length of a UTF-32 string in UTF-8. If the string is invalid 52 * such as containing a surrogate character, -1 will be returned. 53 */ 54 ssize_t utf32_to_utf8_length(const char32_t *src, size_t src_len); 55 56 /** 57 * Stores a UTF-8 string converted from "src" in "dst", if "dst_length" is not 58 * large enough to store the string, the part of the "src" string is stored 59 * into "dst" as much as possible. See the examples for more detail. 60 * Returns the size actually used for storing the string. 61 * dst" is not null-terminated when dst_len is fully used (like strncpy). 62 * 63 * Example 1 64 * "src" == \u3042\u3044 (\xE3\x81\x82\xE3\x81\x84) 65 * "src_len" == 2 66 * "dst_len" >= 7 67 * -> 68 * Returned value == 6 69 * "dst" becomes \xE3\x81\x82\xE3\x81\x84\0 70 * (note that "dst" is null-terminated) 71 * 72 * Example 2 73 * "src" == \u3042\u3044 (\xE3\x81\x82\xE3\x81\x84) 74 * "src_len" == 2 75 * "dst_len" == 5 76 * -> 77 * Returned value == 3 78 * "dst" becomes \xE3\x81\x82\0 79 * (note that "dst" is null-terminated, but \u3044 is not stored in "dst" 80 * since "dst" does not have enough size to store the character) 81 * 82 * Example 3 83 * "src" == \u3042\u3044 (\xE3\x81\x82\xE3\x81\x84) 84 * "src_len" == 2 85 * "dst_len" == 6 86 * -> 87 * Returned value == 6 88 * "dst" becomes \xE3\x81\x82\xE3\x81\x84 89 * (note that "dst" is NOT null-terminated, like strncpy) 90 */ 91 void utf32_to_utf8(const char32_t* src, size_t src_len, char* dst); 92 93 /** 94 * Returns the unicode value at "index". 95 * Returns -1 when the index is invalid (equals to or more than "src_len"). 96 * If returned value is positive, it is able to be converted to char32_t, which 97 * is unsigned. Then, if "next_index" is not NULL, the next index to be used is 98 * stored in "next_index". "next_index" can be NULL. 99 */ 100 int32_t utf32_from_utf8_at(const char *src, size_t src_len, size_t index, size_t *next_index); 101 102 103 /** 104 * Returns the UTF-8 length of UTF-16 string "src". 105 */ 106 ssize_t utf16_to_utf8_length(const char16_t *src, size_t src_len); 107 108 /** 109 * Converts a UTF-16 string to UTF-8. The destination buffer must be large 110 * enough to fit the UTF-16 as measured by utf16_to_utf8_length with an added 111 * NULL terminator. 112 */ 113 void utf16_to_utf8(const char16_t* src, size_t src_len, char* dst); 114 115 /** 116 * Returns the length of "src" when "src" is valid UTF-8 string. 117 * Returns 0 if src is NULL or 0-length string. Returns -1 when the source 118 * is an invalid string. 119 * 120 * This function should be used to determine whether "src" is valid UTF-8 121 * characters with valid unicode codepoints. "src" must be null-terminated. 122 * 123 * If you are going to use other utf8_to_... functions defined in this header 124 * with string which may not be valid UTF-8 with valid codepoint (form 0 to 125 * 0x10FFFF), you should use this function before calling others, since the 126 * other functions do not check whether the string is valid UTF-8 or not. 127 * 128 * If you do not care whether "src" is valid UTF-8 or not, you should use 129 * strlen() as usual, which should be much faster. 130 */ 131 ssize_t utf8_length(const char *src); 132 133 /** 134 * Measure the length of a UTF-32 string. 135 */ 136 size_t utf8_to_utf32_length(const char *src, size_t src_len); 137 138 /** 139 * Stores a UTF-32 string converted from "src" in "dst". "dst" must be large 140 * enough to store the entire converted string as measured by 141 * utf8_to_utf32_length plus space for a NULL terminator. 142 */ 143 void utf8_to_utf32(const char* src, size_t src_len, char32_t* dst); 144 145 /** 146 * Returns the UTF-16 length of UTF-8 string "src". 147 */ 148 ssize_t utf8_to_utf16_length(const uint8_t* src, size_t srcLen); 149 150 /** 151 * Convert UTF-8 to UTF-16 including surrogate pairs. 152 * Returns a pointer to the end of the string (where a null terminator might go 153 * if you wanted to add one). 154 */ 155 char16_t* utf8_to_utf16_no_null_terminator(const uint8_t* src, size_t srcLen, char16_t* dst); 156 157 /** 158 * Convert UTF-8 to UTF-16 including surrogate pairs. The destination buffer 159 * must be large enough to hold the result as measured by utf8_to_utf16_length 160 * plus an added NULL terminator. 161 */ 162 void utf8_to_utf16(const uint8_t* src, size_t srcLen, char16_t* dst); 163 164 /** 165 * Like utf8_to_utf16_no_null_terminator, but you can supply a maximum length of the 166 * decoded string. The decoded string will fill up to that length; if it is longer 167 * the returned pointer will be to the character after dstLen. 168 */ 169 char16_t* utf8_to_utf16_n(const uint8_t* src, size_t srcLen, char16_t* dst, size_t dstLen); 170 171 } 172 173 #endif 174