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// This should only be used by the various UTF string conversion files. 9 10#include <stddef.h> 11#include <stdint.h> 12 13#include "base/base_export.h" 14#include "base/strings/string16.h" 15 16namespace base { 17 18inline bool IsValidCodepoint(uint32_t code_point) { 19 // Excludes the surrogate code points ([0xD800, 0xDFFF]) and 20 // codepoints larger than 0x10FFFF (the highest codepoint allowed). 21 // Non-characters and unassigned codepoints are allowed. 22 return code_point < 0xD800u || 23 (code_point >= 0xE000u && code_point <= 0x10FFFFu); 24} 25 26inline bool IsValidCharacter(uint32_t code_point) { 27 // Excludes non-characters (U+FDD0..U+FDEF, and all codepoints ending in 28 // 0xFFFE or 0xFFFF) from the set of valid code points. 29 return code_point < 0xD800u || (code_point >= 0xE000u && 30 code_point < 0xFDD0u) || (code_point > 0xFDEFu && 31 code_point <= 0x10FFFFu && (code_point & 0xFFFEu) != 0xFFFEu); 32} 33 34// ReadUnicodeCharacter -------------------------------------------------------- 35 36// Reads a UTF-8 stream, placing the next code point into the given output 37// |*code_point|. |src| represents the entire string to read, and |*char_index| 38// is the character offset within the string to start reading at. |*char_index| 39// will be updated to index the last character read, such that incrementing it 40// (as in a for loop) will take the reader to the next character. 41// 42// Returns true on success. On false, |*code_point| will be invalid. 43BASE_EXPORT bool ReadUnicodeCharacter(const char* src, 44 int32_t src_len, 45 int32_t* char_index, 46 uint32_t* code_point_out); 47 48// Reads a UTF-16 character. The usage is the same as the 8-bit version above. 49BASE_EXPORT bool ReadUnicodeCharacter(const char16* src, 50 int32_t src_len, 51 int32_t* char_index, 52 uint32_t* code_point); 53 54#if defined(WCHAR_T_IS_UTF32) 55// Reads UTF-32 character. The usage is the same as the 8-bit version above. 56BASE_EXPORT bool ReadUnicodeCharacter(const wchar_t* src, 57 int32_t src_len, 58 int32_t* char_index, 59 uint32_t* code_point); 60#endif // defined(WCHAR_T_IS_UTF32) 61 62// WriteUnicodeCharacter ------------------------------------------------------- 63 64// Appends a UTF-8 character to the given 8-bit string. Returns the number of 65// bytes written. 66BASE_EXPORT size_t WriteUnicodeCharacter(uint32_t code_point, 67 std::string* output); 68 69// Appends the given code point as a UTF-16 character to the given 16-bit 70// string. Returns the number of 16-bit values written. 71BASE_EXPORT size_t WriteUnicodeCharacter(uint32_t code_point, string16* output); 72 73#if defined(WCHAR_T_IS_UTF32) 74// Appends the given UTF-32 character to the given 32-bit string. Returns the 75// number of 32-bit values written. 76inline size_t WriteUnicodeCharacter(uint32_t code_point, std::wstring* output) { 77 // This is the easy case, just append the character. 78 output->push_back(code_point); 79 return 1; 80} 81#endif // defined(WCHAR_T_IS_UTF32) 82 83// Generalized Unicode converter ----------------------------------------------- 84 85// Guesses the length of the output in UTF-8 in bytes, clears that output 86// string, and reserves that amount of space. We assume that the input 87// character types are unsigned, which will be true for UTF-16 and -32 on our 88// systems. 89template<typename CHAR> 90void PrepareForUTF8Output(const CHAR* src, size_t src_len, std::string* output); 91 92// Prepares an output buffer (containing either UTF-16 or -32 data) given some 93// UTF-8 input that will be converted to it. See PrepareForUTF8Output(). 94template<typename STRING> 95void PrepareForUTF16Or32Output(const char* src, size_t src_len, STRING* output); 96 97} // namespace base 98 99#endif // BASE_STRINGS_UTF_STRING_CONVERSION_UTILS_H_ 100