base/strings/utf_string_conversion_utils.h

   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 "base/base_export.h"
  11 #include "base/strings/string16.h"
  12
  13 namespace base {
  14
  15 inline bool IsValidCodepoint(uint32 code_point) {
  16   // Excludes the surrogate code points ([0xD800, 0xDFFF]) and
  17   // codepoints larger than 0x10FFFF (the highest codepoint allowed).
  18   // Non-characters and unassigned codepoints are allowed.
  19   return code_point < 0xD800u ||
  20          (code_point >= 0xE000u && code_point <= 0x10FFFFu);
  21 }
  22
  23 inline bool IsValidCharacter(uint32 code_point) {
  24   // Excludes non-characters (U+FDD0..U+FDEF, and all codepoints ending in
  25   // 0xFFFE or 0xFFFF) from the set of valid code points.
  26   return code_point < 0xD800u || (code_point >= 0xE000u &&
  27       code_point < 0xFDD0u) || (code_point > 0xFDEFu &&
  28       code_point <= 0x10FFFFu && (code_point & 0xFFFEu) != 0xFFFEu);
  29 }
  30
  31 // ReadUnicodeCharacter --------------------------------------------------------
  32
  33 // Reads a UTF-8 stream, placing the next code point into the given output
  34 // |*code_point|. |src| represents the entire string to read, and |*char_index|
  35 // is the character offset within the string to start reading at. |*char_index|
  36 // will be updated to index the last character read, such that incrementing it
  37 // (as in a for loop) will take the reader to the next character.
  38 //
  39 // Returns true on success. On false, |*code_point| will be invalid.
  40 BASE_EXPORT bool ReadUnicodeCharacter(const char* src,
  41                                       int32 src_len,
  42                                       int32* char_index,
  43                                       uint32* code_point_out);
  44
  45 // Reads a UTF-16 character. The usage is the same as the 8-bit version above.
  46 BASE_EXPORT bool ReadUnicodeCharacter(const char16* src,
  47                                       int32 src_len,
  48                                       int32* char_index,
  49                                       uint32* code_point);
  50
  51 #if defined(WCHAR_T_IS_UTF32)
  52 // Reads UTF-32 character. The usage is the same as the 8-bit version above.
  53 BASE_EXPORT bool ReadUnicodeCharacter(const wchar_t* src,
  54                                       int32 src_len,
  55                                       int32* char_index,
  56                                       uint32* code_point);
  57 #endif  // defined(WCHAR_T_IS_UTF32)
  58
  59 // WriteUnicodeCharacter -------------------------------------------------------
  60
  61 // Appends a UTF-8 character to the given 8-bit string.  Returns the number of
  62 // bytes written.
  63 // TODO(brettw) Bug 79631: This function should not be exposed.
  64 BASE_EXPORT size_t WriteUnicodeCharacter(uint32 code_point,
  65                                          std::string* output);
  66
  67 // Appends the given code point as a UTF-16 character to the given 16-bit
  68 // string.  Returns the number of 16-bit values written.
  69 BASE_EXPORT size_t WriteUnicodeCharacter(uint32 code_point, string16* output);
  70
  71 #if defined(WCHAR_T_IS_UTF32)
  72 // Appends the given UTF-32 character to the given 32-bit string.  Returns the
  73 // number of 32-bit values written.
  74 inline size_t WriteUnicodeCharacter(uint32 code_point, std::wstring* output) {
  75   // This is the easy case, just append the character.
  76   output->push_back(code_point);
  77   return 1;
  78 }
  79 #endif  // defined(WCHAR_T_IS_UTF32)
  80
  81 // Generalized Unicode converter -----------------------------------------------
  82
  83 // Guesses the length of the output in UTF-8 in bytes, clears that output
  84 // string, and reserves that amount of space.  We assume that the input
  85 // character types are unsigned, which will be true for UTF-16 and -32 on our
  86 // systems.
  87 template<typename CHAR>
  88 void PrepareForUTF8Output(const CHAR* src, size_t src_len, std::string* output);
  89
  90 // Prepares an output buffer (containing either UTF-16 or -32 data) given some
  91 // UTF-8 input that will be converted to it.  See PrepareForUTF8Output().
  92 template<typename STRING>
  93 void PrepareForUTF16Or32Output(const char* src, size_t src_len, STRING* output);
  94
  95 }  // namespace base
  96
  97 #endif  // BASE_STRINGS_UTF_STRING_CONVERSION_UTILS_H_