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
12 namespace base {
13
IsValidCodepoint(uint32 code_point)14 inline 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
22 // ReadUnicodeCharacter --------------------------------------------------------
23
24 // Reads a UTF-8 stream, placing the next code point into the given output
25 // |*code_point|. |src| represents the entire string to read, and |*char_index|
26 // is the character offset within the string to start reading at. |*char_index|
27 // will be updated to index the last character read, such that incrementing it
28 // (as in a for loop) will take the reader to the next character.
29 //
30 // Returns true on success. On false, |*code_point| will be invalid.
31 bool ReadUnicodeCharacter(const char* src,
32 int32 src_len,
33 int32* char_index,
34 uint32* code_point_out);
35
36 // Reads a UTF-16 character. The usage is the same as the 8-bit version above.
37 bool ReadUnicodeCharacter(const char16* src,
38 int32 src_len,
39 int32* char_index,
40 uint32* code_point);
41
42 #if defined(WCHAR_T_IS_UTF32)
43 // Reads UTF-32 character. The usage is the same as the 8-bit version above.
44 bool ReadUnicodeCharacter(const wchar_t* src,
45 int32 src_len,
46 int32* char_index,
47 uint32* code_point);
48 #endif // defined(WCHAR_T_IS_UTF32)
49
50 // WriteUnicodeCharacter -------------------------------------------------------
51
52 // Appends a UTF-8 character to the given 8-bit string. Returns the number of
53 // bytes written.
54 size_t WriteUnicodeCharacter(uint32 code_point, std::string* output);
55
56 // Appends the given code point as a UTF-16 character to the given 16-bit
57 // string. Returns the number of 16-bit values written.
58 size_t WriteUnicodeCharacter(uint32 code_point, string16* output);
59
60 #if defined(WCHAR_T_IS_UTF32)
61 // Appends the given UTF-32 character to the given 32-bit string. Returns the
62 // number of 32-bit values written.
WriteUnicodeCharacter(uint32 code_point,std::wstring * output)63 inline size_t WriteUnicodeCharacter(uint32 code_point, std::wstring* output) {
64 // This is the easy case, just append the character.
65 output->push_back(code_point);
66 return 1;
67 }
68 #endif // defined(WCHAR_T_IS_UTF32)
69
70 // Generalized Unicode converter -----------------------------------------------
71
72 // Guesses the length of the output in UTF-8 in bytes, clears that output
73 // string, and reserves that amount of space. We assume that the input
74 // character types are unsigned, which will be true for UTF-16 and -32 on our
75 // systems.
76 template<typename CHAR>
77 void PrepareForUTF8Output(const CHAR* src, size_t src_len, std::string* output);
78
79 // Prepares an output buffer (containing either UTF-16 or -32 data) given some
80 // UTF-8 input that will be converted to it. See PrepareForUTF8Output().
81 template<typename STRING>
82 void PrepareForUTF16Or32Output(const char* src, size_t src_len, STRING* output);
83
84 } // namespace base
85
86 #endif // BASE_UTF_STRING_CONVERSION_UTILS_H_
87