FIX: ClassWizard Floating-Point Parsing Errors

WINDOWS   | WINDOWS NT

The information in this article applies to:

  - Microsoft Visual C++ for Windows, versions 1.0 and 1.5
  - Microsoft Visual C++ 32-bit Edition, versions 1.0 and 2.0

SYMPTOMS

ClassWizard cannot parse floating-point numbers written in scientific floating-point notation, such as 1.e+006. If a number formatted in this manner is in the lines of code that ClassWizard must parse, the following error is generated:

   Parsing error: Expected ")".
   Input Line: "DDV_MinMaxFloat(pDX, m_Var, 0., 1.e+006);"

1. Add an edit control to a dialog box in App Studio.

2. Add a float or double member variable to class of the previously added edit control.

3. In the range validation fields, add a floating-point number that fits one of the following criteria, for member variable of type float:

   a. It has seven digits or more to the left of the decimal point.

   b. It has one or more zeros to the right of the decimal point, between

         the decimal point and the first nonzero number.
   

   Or the range validation fields contain a floating-pointer number that fits one of the following criteria, for a member variable of type double:

   a. It has 16 digits or more to the left of the decimal point.

   b. It has one or more zeros to the right of the decimal point, between

         the decimal point and the first nonzero number.
   

RESOLUTION

The only work around is to modify the code by hand and convert it back to the original notation. Be aware, however, that if you edit the variables for that class again, the numbers will be converted back to scientific notation.

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++ 2.1.