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// Low-level UTF handling functions. Most code will want to use the functions
9// in utf_string_conversions.h
10
11#include <stddef.h>
12#include <stdint.h>
13
14#include <string>
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 ||
30         (code_point >= 0xE000u && code_point < 0xFDD0u) ||
31         (code_point > 0xFDEFu && code_point <= 0x10FFFFu &&
32          (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.
44bool ReadUnicodeCharacter(const char* src,
45                          int32_t src_len,
46                          int32_t* char_index,
47                          uint32_t* code_point_out);
48
49// Reads a UTF-16 character. The usage is the same as the 8-bit version above.
50bool ReadUnicodeCharacter(const char16_t* src,
51                          int32_t src_len,
52                          int32_t* char_index,
53                          uint32_t* code_point);
54
55// WriteUnicodeCharacter -------------------------------------------------------
56
57// Appends a UTF-8 character to the given 8-bit string.  Returns the number of
58// bytes written.
59size_t WriteUnicodeCharacter(uint32_t code_point, std::string* output);
60
61// Appends the given code point as a UTF-16 character to the given 16-bit
62// string.  Returns the number of 16-bit values written.
63size_t WriteUnicodeCharacter(uint32_t code_point, std::u16string* output);
64
65// Generalized Unicode converter -----------------------------------------------
66
67// Guesses the length of the output in UTF-8 in bytes, clears that output
68// string, and reserves that amount of space.  We assume that the input
69// character types are unsigned, which will be true for UTF-16 and -32 on our
70// systems.
71template <typename CHAR>
72void PrepareForUTF8Output(const CHAR* src, size_t src_len, std::string* output);
73
74// Prepares an output buffer (containing either UTF-16 or -32 data) given some
75// UTF-8 input that will be converted to it.  See PrepareForUTF8Output().
76template <typename STRING>
77void PrepareForUTF16Or32Output(const char* src, size_t src_len, STRING* output);
78
79}  // namespace base
80
81#endif  // BASE_STRINGS_UTF_STRING_CONVERSION_UTILS_H_
82