Unicode Programming Summary
To take advantage of the MFC and C run-time support for Unicode, you need to:
- Define _UNICODE.
Define the symbol _UNICODE before you build your program.
- Specify entry point.
In the Output page of the Linker folder in the project's Property Pages dialog box, set the Entry Point symbol to wWinMainCRTStartup.
- Use "portable" run-time functions and types.
Use the proper C run-time functions for Unicode string handling. You can use the wcs family of functions, but you may prefer the fully "portable" (internationally enabled) _TCHAR macros. These macros are all prefixed with _tcs; they substitute, one for one, for the str family of functions. These functions are described in detail in the Internationalization section of the Run-Time Library Reference. For more information, see Generic-Text Mappings in TCHAR.H.
Use _TCHAR and the related portable data types described in Support for Unicode.
- Handle literal strings properly.
The Visual C++ compiler interprets a literal string coded as
L"this is a literal string"
to mean a string of Unicode characters. You can use the same prefix for literal characters. Use the _T macro to code literal strings generically, so they compile as Unicode strings under Unicode or as ANSI strings (including MBCS) without Unicode. For example, instead of:
pWnd->SetWindowText( "Hello" );
pWnd->SetWindowText( _T("Hello") );
With _UNICODE defined, _T translates the literal string to the L-prefixed form; otherwise, _T translates the string without the L prefix.
Tip The _T macro is identical to the _TEXT macro.
- Be careful passing string lengths to functions.
Some functions want the number of characters in a string; others want the number of bytes. For example, if _UNICODE is defined, the following call to a CArchive object will not work (
stris a CString):
archive.Write( str, str.GetLength( ) ); // invalid
In a Unicode application, the length gives you the number of characters but not the correct number of bytes, since each character is two bytes wide. Instead, you must use:
archive.Write( str, str.GetLength( ) * sizeof( _TCHAR ) ); // valid
which specifies the correct number of bytes to write.
However, MFC member functions that are character-oriented, rather than byte-oriented, work without this extra coding:
pDC->TextOut( str, str.GetLength( ) );
CDC::TextOut takes a number of characters, not a number of bytes.
To summarize, MFC and the run-time library provide the following support for Unicode programming under Windows 2000:
- Except for database class member functions, all MFC functions are Unicode-enabled, including CString. CString also provides Unicode/ANSI conversion functions.
- The run-time library supplies Unicode versions of all string-handling functions. (The run-time library also supplies "portable" versions suitable for Unicode or for MBCS. These are the _tcs macros.)
- TCHAR.H supplies portable data types and the _T macro for translating literal strings and characters. See Generic-Text Mappings in TCHAR.H.
- The run-time library provides a wide-character version of main. Use wmain to make your application "Unicode-aware."