utf_string_conversion_utils.h revision c407dc5cd9bdc5668497f21b26b09d988ab439de 
1// Copyright (c) 2009 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_UTF_STRING_CONVERSION_UTILS_H_
6#define BASE_UTF_STRING_CONVERSION_UTILS_H_
7
8// This should only be used by the various UTF string conversion files.
9
10#include "base/string16.h"
11
12namespace base {
13
14inline bool IsValidCodepoint(uint32 code_point) {
15  // Excludes the surrogate code points ([0xD800, 0xDFFF]) and
16  // codepoints larger than 0x10FFFF (the highest codepoint allowed).
17  // Non-characters and unassigned codepoints are allowed.
18  return code_point < 0xD800u ||
19         (code_point >= 0xE000u && code_point <= 0x10FFFFu);
20}
21
22inline bool IsValidCharacter(uint32 code_point) {
23  // Excludes non-characters (U+FDD0..U+FDEF, and all codepoints ending in
24  // 0xFFFE or 0xFFFF) from the set of valid code points.
25  return code_point < 0xD800u || (code_point >= 0xE000u &&
26      code_point < 0xFDD0u) || (code_point > 0xFDEFu &&
27      code_point <= 0x10FFFFu && (code_point & 0xFFFEu) != 0xFFFEu);
28}
29
30// ReadUnicodeCharacter --------------------------------------------------------
31
32// Reads a UTF-8 stream, placing the next code point into the given output
33// |*code_point|. |src| represents the entire string to read, and |*char_index|
34// is the character offset within the string to start reading at. |*char_index|
35// will be updated to index the last character read, such that incrementing it
36// (as in a for loop) will take the reader to the next character.
37//
38// Returns true on success. On false, |*code_point| will be invalid.
39bool ReadUnicodeCharacter(const char* src,
40                          int32 src_len,
41                          int32* char_index,
42                          uint32* code_point_out);
43
44// Reads a UTF-16 character. The usage is the same as the 8-bit version above.
45bool ReadUnicodeCharacter(const char16* src,
46                          int32 src_len,
47                          int32* char_index,
48                          uint32* code_point);
49
50#if defined(WCHAR_T_IS_UTF32)
51// Reads UTF-32 character. The usage is the same as the 8-bit version above.
52bool ReadUnicodeCharacter(const wchar_t* src,
53                          int32 src_len,
54                          int32* char_index,
55                          uint32* code_point);
56#endif  // defined(WCHAR_T_IS_UTF32)
57
58// WriteUnicodeCharacter -------------------------------------------------------
59
60// Appends a UTF-8 character to the given 8-bit string.  Returns the number of
61// bytes written.
62size_t WriteUnicodeCharacter(uint32 code_point, std::string* output);
63
64// Appends the given code point as a UTF-16 character to the given 16-bit
65// string.  Returns the number of 16-bit values written.
66size_t WriteUnicodeCharacter(uint32 code_point, string16* output);
67
68#if defined(WCHAR_T_IS_UTF32)
69// Appends the given UTF-32 character to the given 32-bit string.  Returns the
70// number of 32-bit values written.
71inline size_t WriteUnicodeCharacter(uint32 code_point, std::wstring* output) {
72  // This is the easy case, just append the character.
73  output->push_back(code_point);
74  return 1;
75}
76#endif  // defined(WCHAR_T_IS_UTF32)
77
78// Generalized Unicode converter -----------------------------------------------
79
80// Guesses the length of the output in UTF-8 in bytes, clears that output
81// string, and reserves that amount of space.  We assume that the input
82// character types are unsigned, which will be true for UTF-16 and -32 on our
83// systems.
84template<typename CHAR>
85void PrepareForUTF8Output(const CHAR* src, size_t src_len, std::string* output);
86
87// Prepares an output buffer (containing either UTF-16 or -32 data) given some
88// UTF-8 input that will be converted to it.  See PrepareForUTF8Output().
89template<typename STRING>
90void PrepareForUTF16Or32Output(const char* src, size_t src_len, STRING* output);
91
92}  // namespace base
93
94#endif  // BASE_UTF_STRING_CONVERSION_UTILS_H_
95