std::ctype<CharT>:: widen, do_widen

Widening always returns a wide character, but only the characters from the basic source character set (until C++23) basic character set (since C++23) are guaranteed to have a unique, well-defined, widening transformation, which is also guaranteed to be reversible (by narrow() ). In practice, all characters whose multibyte representation is a single byte are usually widened to their wide character counterparts, and the rest of the possible single-byte values are usually mapped into the same placeholder value, typically CharT ( - 1 ) .

Widening, if successful, preserves all character classification categories known to is() .

Parameters

Return value

Example

#include <iostream>
#include <locale>
 
void try_widen(const std::ctype<wchar_t>& f, char c)
{
    wchar_t w = f.widen(c);
    std::cout << "The single-byte character " << +(unsigned char)c
              << " widens to " << +w << '\n';
}
 
int main()
{
    std::locale::global(std::locale("cs_CZ.iso88592"));
    auto& f = std::use_facet<std::ctype<wchar_t>>(std::locale());
    std::cout << std::hex << std::showbase << "In Czech ISO-8859-2 locale:\n";
    try_widen(f, 'a');
    try_widen(f, '\xdf'); // German letter ß (U+00df) in ISO-8859-2
    try_widen(f, '\xec'); // Czech letter ě (U+011b) in ISO-8859-2
 
    std::locale::global(std::locale("cs_CZ.utf8"));
    auto& f2 = std::use_facet<std::ctype<wchar_t>>(std::locale());
    std::cout << "In Czech UTF-8 locale:\n";
    try_widen(f2, 'a');
    try_widen(f2, '\xdf'); 
    try_widen(f2, '\xec'); 
}

In Czech ISO-8859-2 locale:
The single-byte character 0x61 widens to 0x61
The single-byte character 0xdf widens to 0xdf
The single-byte character 0xec widens to 0x11b
In Czech UTF-8 locale:
The single-byte character 0x61 widens to 0x61
The single-byte character 0xdf widens to 0xffffffff
The single-byte character 0xec widens to 0xffffffff

Defect reports

The following behavior-changing defect reports were applied retroactively to previously published C++ standards.

See also