Problem When Using IMP with a Variable and a Literal (61339)

SYMPTOMS

The code example below using the IMP logical-implication operator incorrectly prints -1 when compiled with the BC.EXE environment of Microsoft Basic Professional Development System (PDS) version 7.00 or 7.10. When executed in the QBX.EXE (QuickBasic Extended) environment, the correct answer of 0 (zero) prints.

The only workaround is to change the -1 to a variable or change i% in the PRINT statement to a 0 (zero).

STATUS

Microsoft has confirmed this to be a bug in the BC.EXE environment of Microsoft Basic Professional Development System (PDS) versions 7.00 and 7.10 for MS-DOS and MS OS/2; in the BC.EXE environment of Microsoft QuickBasic versions 4.00, 4.00b, and 4.50 (buglist4.00, buglist4.00b, buglist4.50) for MS-DOS; and in the BC.EXE environment of Microsoft Basic Compiler versions 6.00 and 6.00b for MS-DOS and MS OS/2. We are researching this problem and will post new information here in the Microsoft Knowledge Base as it becomes available.

MORE INFORMATION

For your information, the following definition of IMP is taken from the Microsoft Advisor online Help system in QBX.EXE:

HELP: IMP Operator

result = numeric - expression1 IMP numeric - expression2

The logical-implication operator, IMP, compares corresponding bits in numeric-expression1 and numeric-expression2, and then sets the corresponding bit in the result according to the following table:

   Bit in first expression   Bit in second expression   Bit in result
   -----------------------   ------------------------   -------------
                1                          1                    1
                1                          0                    0
                0                          1                    1
                0                          0                    1

Code Example

   DEFINT A-Z
   i% = 0
   PRINT -1 IMP i%

If a variable is used in the place of the -1 or the i% is replaced with a 0, then this program prints the correct answer of 0.

Note that the constant -1 is stored internally (in two's complement signed binary integer format) with all 16 bits set equal to 1 (on). The constant 0, when stored in an integer variable, is stored with all 16 bits set equal to 0 (off).