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