fork(2) download 
  1. #include <cassert>
  2. #include <cwctype>
  3. #include <cstdlib>
  4. #include <iomanip>
  5. #include <iostream>
  6. #include <locale>
  7. #include <string>
  8.  
  9. #if _WIN32 || _WIN64
  10. // Windows needs a little non-standard magic for this to work.
  11. #include <io.h>
  12. #include <fcntl.h>
  13. #include <locale.h>
  14. #endif
  15.  
  16. using std::size_t;
  17.  
  18. void init_locale(void)
  19. // Does magic so that wcout can work.
  20. {
  21. #if _WIN32 || _WIN64
  22.   // Windows needs a little non-standard magic.
  23.   constexpr char cp_utf16le[] = ".1200";
  24.   setlocale( LC_ALL, cp_utf16le );
  25.   _setmode( _fileno(stdout), _O_U16TEXT );
  26. #else
  27.   // The correct locale name may vary by OS, e.g., "en_US.utf8".
  28.   constexpr char locale_name[] = "";
  29.   std::locale::global(std::locale(locale_name));
  30.   std::wcout.imbue(std::locale());
  31. #endif
  32. }
  33.  
  34. std::u16string make_u16string( const std::wstring& ws )
  35. /* Creates a UTF-16 string from a wide-character string.  Any wide characters
  36.  * outside the allowed range of UTF-16 are mapped to the sentinel value U+FFFD,
  37.  * per the Unicode documentation. (http://w...content-available-to-author-only...e.org/faq/private_use.html
  38.  * retrieved 12 March 2017.) Unpaired surrogates in ws are also converted to
  39.  * sentinel values.  Noncharacters, however, are left intact.  As a fallback,
  40.  * if wide characters are the same size as char16_t, this does a more trivial
  41.  * construction using that implicit conversion.
  42.  */
  43. {
  44.   /* We assume that, if this test passes, a wide-character string is already
  45.    * UTF-16, or at least converts to it implicitly without needing surrogate
  46.    * pairs.
  47.    */
  48.   if ( sizeof(wchar_t) == sizeof(char16_t) ) {
  49.     return std::u16string( ws.begin(), ws.end() );
  50.   } else {
  51.     /* The conversion from UTF-32 to UTF-16 might possibly require surrogates.
  52.      * A surrogate pair suffices to represent all wide characters, because all
  53.      * characters outside the range will be mapped to the sentinel value
  54.      * U+FFFD.  Add one character for the terminating NUL.
  55.      */
  56.     const size_t max_len = 2 * ws.length() + 1;
  57.     // Our temporary UTF-16 string.
  58.     std::u16string result;
  59.  
  60.     result.reserve(max_len);
  61.  
  62.     for ( const wchar_t& wc : ws ) {
  63.       const std::wint_t chr = wc;
  64.  
  65.       if ( chr < 0 || chr > 0x10FFFF || (chr >= 0xD800 && chr <= 0xDFFF) ) {
  66.         // Invalid code point.  Replace with sentinel, per Unicode standard:
  67.         constexpr char16_t sentinel = u'\uFFFD';
  68.         result.push_back(sentinel);
  69.       } else if ( chr < 0x10000UL ) { // In the BMP.
  70.         result.push_back(static_cast<char16_t>(wc));
  71.       } else {
  72.         const char16_t leading = static_cast<char16_t>( 
  73.           ((chr-0x10000UL) / 0x400U) + 0xD800U );
  74.         const char16_t trailing = static_cast<char16_t>( 
  75.           ((chr-0x10000UL) % 0x400U) + 0xDC00U );
  76.  
  77.         result.append({leading, trailing});
  78.       } // end if
  79.     } // end for
  80.  
  81.    /* The returned string is shrunken to fit, which might not be the Right
  82.     * Thing if there is more to be added to the string.
  83.     */
  84.     result.shrink_to_fit();
  85.  
  86.     // We depend here on the compiler to optimize the move constructor.
  87.     return result;
  88.   } // end if
  89.   // Not reached.
  90. }
  91.  
  92. int main(void)
  93. {
  94.   static const std::wstring wtest(L"☪☮∈✡℩☯✝ \U0001F644");
  95.   static const std::u16string u16test(u"☪☮∈✡℩☯✝ \U0001F644");
  96.   const std::u16string converted = make_u16string(wtest);
  97.  
  98.   init_locale();
  99.  
  100.   std::wcout << L"sizeof(wchar_t) == " << sizeof(wchar_t) << L".\n";
  101.  
  102.   for( size_t i = 0; i <= u16test.length(); ++i ) {
  103.     if ( u16test[i] != converted[i] ) {
  104.       std::wcout << std::hex << std::showbase
  105.                  << std::right << std::setfill(L'0')
  106.                  << std::setw(4) << (unsigned)converted[i] << L" ≠ "
  107.                  << std::setw(4) << (unsigned)u16test[i] << L" at "
  108.                  << i << L'.' << std::endl;
  109.       return EXIT_FAILURE;
  110.     } // end if
  111.   } // end for
  112.  
  113.   std::wcout << wtest << std::endl;
  114.  
  115.   return EXIT_SUCCESS;
  116. }
Success #stdin #stdout 0s 16920KB
comments (?)
stdin 
Standard input is empty
stdout 
sizeof(wchar_t) == 4.
☪☮∈✡℩☯✝ 🙄
https://ideone.com/0ZCVqn
language:
C++14 (gcc 8.3)
created:
8 years ago
visibility:
public

Share or Embed source code

Discover > Sphere Engine API

The brand new service which powers Ideone!

Discover > IDE Widget

Widget for compiling and running the source code in a web browser!