How to turn what returns WM_CHAR into unicode?

Question

C++ / C#

There is WndProc and it has WM_CHAR processing.
Only if you translate the character code that it gives out, it will be a krakozyavra. (this is if the Russian layout)
And the code of Russian letters generally starts with 1k.

How to fix it?

Answer 1

Recardo_Recoly ,
I see.
1. It is better to use WM_UNICHAR, it also works with codes over 65535. At least WinXP is supported.
2. How to turn a code position into UTF-8, there are many options. I'll find mine now.

enum {
    SURROGATE_MIN = 0xD800,
    SURROGATE_MAX = 0xDFFF,
    SURROGATE_LO_MIN = SURROGATE_MIN,
    SURROGATE_HI_MIN = 0xDC00,
    SURROGATE_LO_MAX = SURROGATE_HI_MIN - 1,
    SURROGATE_HI_MAX = SURROGATE_MAX,
    UNICODE_MAX = 0x10FFFF,
    U8_1BYTE_MAX = 0x7F,
    U8_2BYTE_MIN = 0x80,
    U8_2BYTE_MAX = 0x7FF,
    U8_3BYTE_MIN = 0x800,
    U8_3BYTE_MAX = 0xFFFF,
    U8_4BYTE_MIN = 0x10000,
    U8_4BYTE_MAX = UNICODE_MAX,
    U16_1WORD_MAX = 0xFFFF,
    U16_2WORD_MIN = 0x10000,
    U16_2WORD_MAX = UNICODE_MAX,
};

void str::putCpNe (char*& p, unsigned long aCp)
{
    if (aCp <= U8_2BYTE_MAX) {  // 1 or 2 bytes, the most frequent case
        if (aCp <= U8_1BYTE_MAX) {  // 1 byte
            *(p++) = static_cast<char>(aCp);
        } else { // 2 bytes
            *(p++) = static_cast<char>((aCp >> 6)   | 0xC0);
            *(p++) = static_cast<char>((aCp & 0x3F) | 0x80);
        }
    } else {  // 3 or 4 bytes
        if (aCp <= U8_3BYTE_MAX) {  // 3 bytes
            *(p++) = static_cast<char>( (aCp >> 12)        | 0xE0);
            *(p++) = static_cast<char>(((aCp >> 6) & 0x3F) | 0x80);
            *(p++) = static_cast<char>( (aCp       & 0x3F) | 0x80);
        } else {    // 4 bytes
            *(p++) = static_cast<char>(((aCp >> 18) & 0x07) | 0xF0);
            *(p++) = static_cast<char>(((aCp >> 12) & 0x3F) | 0x80);
            *(p++) = static_cast<char>(((aCp >> 6)  & 0x3F) | 0x80);
            *(p++) = static_cast<char>( (aCp        & 0x3F) | 0x80);
        }
    }
}

void str::appendCp(std::string & s, unsigned long aCp)
{
    char c[5];
    char* end = c;
    putCpNe(end, aCp);
    s.append(c, end);
}

The word Cp for me means code point, Unicode code position. Ne - no error-checking, without checking code positions for correctness.
C++ also has regular functions for converting Unicode code positions to UTF-8 and UTF-16. But they are scary as hell, and half of them are deprecated in C++17 :(
And the last thing I want to say.
You are confusing two things: Unicode and its code positions, and UTF-8 and UTF-16 Unicode string formats. The ratio "one Unicode character - one char" is only in UTF-32!