BUG: ios::internal Adds No Fill Characters for int and long
ID: Q123161
|
The information in this article applies to:
-
The C/C++ Compiler (CL.EXE), included with:
-
Microsoft Visual C++ for Windows, versions 1.5, 1.51
-
Microsoft Visual C++ 32-bit Edition, versions 1.0, 2.0, 2.1, 4.0, 4.1, 4.2, 5.0
on the following platforms: x86
SYMPTOMS
Using member function ios::flags() with flag ios::internal fails to add
fill characters after the leading sign but before the value. This only
occurs for negative integers or negative long integers. For other data
types and positive integers or positive long integers, the flag works
correctly.
The following is output from the sample program given under "MORE
INFORMATION":
--- Output Starts ---
+ 1234 // Correct, positive integer with space between
+ and 1234.
+ 1234 // Correct, positive long integer.
+ 1234.000 // Correct, positive float.
+ 1234.000 // Correct, positive double.
-1234 // Incorrect, negative integer with no space
between - and 1234.
-1234 // Incorrect, negative long integer.
- 1.23e+003 // Correct, negative float in scientific notation.
- 1.23e+003 // Correct, negative double.
--- Output Ends ---
RESOLUTION
Cast the integer or long integer values to float or double values, and use
manipulator setprecision(0) and ios flag ios::fixed. See the sample code in
the "More Information" section of this article.
STATUS
Microsoft has confirmed this to be a bug in the Microsoft products listed
at the beginning of this article.
MORE INFORMATION
Sample Code to Reproduce Problem
/* Compile options needed: none
*/
#include <IOSTREAM.H>
#include <IOMANIP.H>
#pragma warning( disable : 4270 )
#define WIDTH 20
void main( void )
{
int i = 1234;
cout.flags( ios::showpos | ios::internal );
cout << setw(WIDTH) << (int) i << endl;
cout << setw(WIDTH) << (long) i << endl;
cout.precision(3);
cout.flags( ios::showpos | ios::internal | ios::fixed );
cout << setw(WIDTH) << (float) i << endl;
cout << setw(WIDTH) << (double) i << endl;
cout << setw(WIDTH) << (int)-i << endl; // Problem
cout << setw(WIDTH) << (long)-i << endl; // Problem
// The following two lines show how to work around the problem.
// cout << setprecision(0) << setw(WIDTH) << (float)-i << endl;
// cout << setprecision(0) << setw(WIDTH) << (float)-i << endl;
cout.precision(3);
cout.flags( ios::showpos | ios::internal | ios::scientific );
cout << setw(WIDTH) << (float)-i << endl;
cout << setw(WIDTH) << (double)-i << endl;
}
Additional query words:
kbVC400bug buglist1.50 buglist1.00 buglist2.00
Keywords : kbCodeGen kbVC kbVC500bug kbx86
Version : winnt:6.0
Platform : winnt
Issue type : kbbug