FIX: Printf Gives Unexpected Results for Doubles on WinCE

• Microsoft Visual C++ for Windows CE, version 1.0

SYMPTOMS

When you use any of the printf C run-time routines (printf, _vstprintf, _stprintf, and so on) with the "%f" format specifier, you get unexpected results. The printf C run-time routines return the floating-point format specifier without the leading '%' instead of the text representation of the number. For example, the following code gives the result "f" instead of "1.23456789":

   _stprintf(pBuffer, TEXT("%f"), 1.23456789); 

CAUSE

This is a bug in the printf C run-time routines. The printf routines for Windows CE do not support floating-point numbers.

RESOLUTION

This has been fixed in version 2.0 of Visual C++ for Windows CE. To work around this problem in version 1.0, you must manually convert the floating- point number. You can use the _gcvt function to convert the floating-point number to an ANSI string, and then use the "%S" format specifier in a printf routine to convert the ANSI string to UNICODE.

For example, replace the following code:

   _stprintf(pBuffer, TEXT("%f"), 1.23456789); 

   char pTemp[256];
   _gcvt(1.23456789,9,pTemp);
   _stprintf(pBuffer, TEXT("%S"), pTemp); 

STATUS

Microsoft has confirmed this to be a bug in the Microsoft products listed at the beginning of this article. This bug has been corrected in Visual C++ for Windows CE, 2.0.

Additional query words: kbnokeyword kbwince100fix

Keywords : kbfix
Version : WINDOWS:1.0
Platform : WINDOWS
Issue type : kbbug