GetClipboardData (CF_UNICODETEXT);

Question

Tell me please, why I get this issues:

• if clipboard contains unicode chars (e.q. russian) I get only first selected word. First word before "space" character.

• if clipboard not contains unicode chars (english only) I get first character of the selected text.

Get selected text:

CStringA getClipboard()
{
     CStringA strData;

     if (OpenClipboard(NULL)){

         HANDLE hClipboardData = GetClipboardData(CF_UNICODETEXT);
         char *pchData = (char*)GlobalLock(hClipboardData);
         strData = pchData;
         GlobalUnlock(hClipboardData);
         CloseClipboard();

    }
    return strData;
}

Set text:

bool setClipboard(CStringA textToclipboard)
{
    bool success = true;

    if (OpenClipboard(NULL)){

        EmptyClipboard();
        HGLOBAL hClipboardData;
        size_t size = (textToclipboard.GetLength()+1) * sizeof(TCHAR);
        hClipboardData = GlobalAlloc(NULL, size);
        TCHAR* pchData = (TCHAR*)GlobalLock(hClipboardData);
        memcpy(pchData, LPCTSTR(textToclipboard.GetString()), size);
        SetClipboardData(CF_UNICODETEXT, hClipboardData);
        GlobalUnlock(hClipboardData);
        CloseClipboard();
    }

    return success;

}

Simply get and set clipboard contents.

CStringA str = getClipboard();
setClipboard(str);

Answer 1

CF_UNICODETEXT uses UTF-16. On Windows, wchar_t data elements are used for UTF-16, but your code is using char instead. CStringA is not compatible with UTF-16. You are mismatching the data in both functions, that is why you do not get the results you are expecting.

One solution is to use CStringW instead of CStringA:

CStringW getClipboard()
{
    CStringW strData;

    if (OpenClipboard(NULL))
    {
        HANDLE hClipboardData = GetClipboardData(CF_UNICODETEXT);
        if (hClipboardData)
        {
            WCHAR *pchData = (WCHAR*) GlobalLock(hClipboardData);
            if (pchData)
            {
                strData = pchData;
                GlobalUnlock(hClipboardData);
            }
        }
        CloseClipboard();
    }
    return strData;
}

bool setClipboard(CStringW textToclipboard)
{
    bool success = true;

    if (OpenClipboard(NULL))
    {
        EmptyClipboard();
        size_t size = (textToclipboard.GetLength()+1) * sizeof(WCHAR);
        HGLOBAL hClipboardData = GlobalAlloc(NULL, size);
        if (hClipboardData)
        {
            WCHAR* pchData = (WCHAR*) GlobalLock(hClipboardData);
            if (pchData)
            {
                memcpy(pchData, (WCHAR*) textToclipboard.GetString(), size);
                GlobalUnlock(hClipboardData);
                SetClipboardData(CF_UNICODETEXT, hClipboardData);
            }
        }
        CloseClipboard();
    }
    return success;
}

If you need to stick with CStringA, then either:

1. use CF_TEXT instead of CF_UNICODETEXT and let the clipboard handle conversions between Ansi and Unicode for you:

   CStringA getClipboard()
   {
       CStringA strData;
   
       if (OpenClipboard(NULL))
       {
           HANDLE hClipboardData = GetClipboardData(CF_TEXT);
           if (hClipboardData)
           {
               CHAR *pchData = (CHAR*) GlobalLock(hClipboardData);
               if (pchData)
               {
                   strData = pchData;
                   GlobalUnlock(hClipboardData);
               }
           }
           CloseClipboard();
       }
       return strData;
   }
   
   bool setClipboard(CStringA textToclipboard)
   {
       bool success = true;
   
       if (OpenClipboard(NULL))
       {
           EmptyClipboard();
           size_t size = (textToclipboard.GetLength()+1) * sizeof(CHAR);
           HGLOBAL hClipboardData = GlobalAlloc(NULL, size);
           if (hClipboardData)
           {
               CHAR* pchData = (CHAR*) GlobalLock(hClipboardData);
               if (pchData)
               {
                   memcpy(pchData, (CHAR*) textToclipboard.GetString(), size);
                   GlobalUnlock(hClipboardData);
                   SetClipboardData(CF_TEXT, hClipboardData);
               }
           }
           CloseClipboard();
       }
       return success;
   }
   

2. convert to/from UTF-16 manually when using CF_UNICODETEXT:

   CStringA getClipboard()
   {
       CStringW strData;
   
       if (OpenClipboard(NULL))
       {
           HANDLE hClipboardData = GetClipboardData(CF_UNICODETEXT);
           if (hClipboardData)
           {
               WCHAR *pchData = (WCHAR*) GlobalLock(hClipboardData);
               if (pchData)
               {
                   strData = pchData;
                   GlobalUnlock(hClipboardData);
               }
           }
           CloseClipboard();
       }
   
       return CStringA((WCHAR*)strData.GetString());
   }
   
   bool setClipboard(CStringA strData)
   {
       CStringW textToclipboard((CHAR*)strData.GetString());
       bool success = true;
   
       if (OpenClipboard(NULL))
       {
           EmptyClipboard();
           size_t size = (textToclipboard.GetLength()+1) * sizeof(WCHAR);
           HGLOBAL hClipboardData = GlobalAlloc(NULL, size);
           if (hClipboardData)
           {
               WCHAR* pchData = (WCHAR*) GlobalLock(hClipboardData);
               if (pchData)
               {
                   memcpy(pchData, (WCHAR*) textToclipboard.GetString(), size);
                   GlobalUnlock(hClipboardData);
                   SetClipboardData(CF_UNICODETEXT, hClipboardData);
               }
           }
           CloseClipboard();
       }
       return success;
   }
   

Another solution is to use CString instead of either CStringA or CStringW, and then use CF_TEXT or CF_UNICODETEXT depending on whether TCHAR is Ansi or Unicode:

#ifdef UNICODE
#define CF_TEXT_T CF_UNICODETEXT
#else
#define CF_TEXT_T CF_TEXT
#endif

CString getClipboard()
{
    CString strData;

    if (OpenClipboard(NULL))
    {
        HANDLE hClipboardData = GetClipboardData(CF_TEXT_T);
        if (hClipboardData)
        {
            TCHAR *pchData = (TCHAR*) GlobalLock(hClipboardData);
            if (pchData)
            {
                strData = pchData;
                GlobalUnlock(hClipboardData);
            }
        }
        CloseClipboard();
    }
    return strData;
}

bool setClipboard(CString textToclipboard)
{
    bool success = true;

    if (OpenClipboard(NULL))
    {
        EmptyClipboard();
        size_t size = (textToclipboard.GetLength()+1) * sizeof(TCHAR);
        HGLOBAL hClipboardData = GlobalAlloc(NULL, size);
        if (hClipboardData)
        {
            TCHAR* pchData = (TCHAR*) GlobalLock(hClipboardData);
            if (pchData)
            {
                memcpy(pchData, (TCHAR*) textToclipboard.GetString(), size);
                GlobalUnlock(hClipboardData);
                SetClipboardData(CF_TEXT_T, hClipboardData);
            }
        }
        CloseClipboard();
    }
    return success;
}

Answer 2

Both of them are Unicode...

But in Unicode, more that one byte represents a character. For example maybe 2 bytes is used for a character. Therefore:

When it's Russian, the string is like

                                                            '\0'
                                                            ----
0x04 0x3F 0x04 0x40 0x04 0x38 0x04 0x32 0x04 0x35 0x04 0x42 0x00 0x20
~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~
    п         р         и        в          е         т       space

It reads until space.

When it's English, the string is like

'\0'
----
0x00 0x48 0x00 0x65 0x00 0x6C 0x00 0x6C 0x00 0x6F 0x00 0x20
~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~ ~~~~~~~~~
   H          e         l        l          o       space

It reads nothing. (If you read first, it's because of order of storing bytes, LE or BE)

Note: Maybe I'm not precision in choosing words (Unicode, UTF, ...)