1 // Copyright (c) 2011 The Chromium Authors. All rights reserved. 2 // Use of this source code is governed by a BSD-style license that can be 3 // found in the LICENSE file. 4 5 #ifndef BASE_STRINGS_UTF_STRING_CONVERSION_UTILS_H_ 6 #define BASE_STRINGS_UTF_STRING_CONVERSION_UTILS_H_ 7 8 #include <string> 9 10 // This should only be used by the various UTF string conversion files. 11 12 #include <stddef.h> 13 #include <stdint.h> 14 15 #include "base/base_export.h" 16 17 namespace base { 18 19 inline bool IsValidCodepoint(uint32_t code_point) { 20 // Excludes the surrogate code points ([0xD800, 0xDFFF]) and 21 // codepoints larger than 0x10FFFF (the highest codepoint allowed). 22 // Non-characters and unassigned codepoints are allowed. 23 return code_point < 0xD800u || 24 (code_point >= 0xE000u && code_point <= 0x10FFFFu); 25 } 26 27 inline bool IsValidCharacter(uint32_t code_point) { 28 // Excludes non-characters (U+FDD0..U+FDEF, and all codepoints ending in 29 // 0xFFFE or 0xFFFF) from the set of valid code points. 30 return code_point < 0xD800u || (code_point >= 0xE000u && 31 code_point < 0xFDD0u) || (code_point > 0xFDEFu && 32 code_point <= 0x10FFFFu && (code_point & 0xFFFEu) != 0xFFFEu); 33 } 34 35 // ReadUnicodeCharacter -------------------------------------------------------- 36 37 // Reads a UTF-8 stream, placing the next code point into the given output 38 // |*code_point|. |src| represents the entire string to read, and |*char_index| 39 // is the character offset within the string to start reading at. |*char_index| 40 // will be updated to index the last character read, such that incrementing it 41 // (as in a for loop) will take the reader to the next character. 42 // 43 // Returns true on success. On false, |*code_point| will be invalid. 44 bool ReadUnicodeCharacter(const char* src, 45 int32_t src_len, 46 int32_t* char_index, 47 uint32_t* code_point_out); 48 49 #if defined(WCHAR_T_IS_UTF32) 50 // Reads UTF-32 character. The usage is the same as the 8-bit version above. 51 bool ReadUnicodeCharacter(const wchar_t* src, 52 int32_t src_len, 53 int32_t* char_index, 54 uint32_t* code_point); 55 #endif // defined(WCHAR_T_IS_UTF32) 56 57 // WriteUnicodeCharacter ------------------------------------------------------- 58 59 // Appends a UTF-8 character to the given 8-bit string. Returns the number of 60 // bytes written. 61 // TODO(brettw) Bug 79631: This function should not be exposed. 62 size_t WriteUnicodeCharacter(uint32_t code_point, std::string* output); 63 64 #if defined(WCHAR_T_IS_UTF32) 65 // Appends the given UTF-32 character to the given 32-bit string. Returns the 66 // number of 32-bit values written. 67 inline size_t WriteUnicodeCharacter(uint32_t code_point, std::wstring* output) { 68 // This is the easy case, just append the character. 69 output->push_back(code_point); 70 return 1; 71 } 72 #endif // defined(WCHAR_T_IS_UTF32) 73 74 // Generalized Unicode converter ----------------------------------------------- 75 76 // Guesses the length of the output in UTF-8 in bytes, clears that output 77 // string, and reserves that amount of space. We assume that the input 78 // character types are unsigned, which will be true for UTF-16 and -32 on our 79 // systems. 80 template<typename CHAR> 81 void PrepareForUTF8Output(const CHAR* src, size_t src_len, std::string* output); 82 83 // Prepares an output buffer (containing either UTF-16 or -32 data) given some 84 // UTF-8 input that will be converted to it. See PrepareForUTF8Output(). 85 template<typename STRING> 86 void PrepareForUTF16Or32Output(const char* src, size_t src_len, STRING* output); 87 88 } // namespace base 89 90 #endif // BASE_STRINGS_UTF_STRING_CONVERSION_UTILS_H_ 91
