Q151073: Frequently Asked Questions for Fortran PowerStation 32
Article: Q151073
Product(s): Microsoft Fortran Compiler
Version(s): 4.0
Operating System(s):
Keyword(s): kbsetup kbCompiler kbFL32 kbFortranPS kbLangFortrankbfaq
Last Modified: 21-OCT-2000
-------------------------------------------------------------------------------
The information in this article applies to:
- Microsoft FORTRAN PowerStation 32 for Windows NT and Windows 95, version 4.0
-------------------------------------------------------------------------------
SUMMARY
=======
The following is a list of Frequently Asked Questions (FAQs) about Microsoft
Fortran PowerStation.
MORE INFORMATION
================
Q. Can I install Fortran PowerStation on a network?
A. Yes, you can. Instructions for installing and administering Fortran
PowerStation on a network are located in the file \FPSSETUP\NETWORK.BAT. Network
installation is only possible with the Fortran PowerStation CD.
Q. When I uninstall the product using the Remove All option, why does the
FPSSetup directory remain in the Fortran PowerStation program group in Windows
NT 3.51?
A. Uninstall does not remove the FPSSetup directory on Windows NT 3.51. This is
the way the setup works. The SETUP.EXE file itself is deleted properly, but not
the \FPSSETUP directory.
Q. Can I create a change or create a new project type based on an existing
project?
A. If you have a project of one type, such as a console application, and you want
to change it to another type, such as QuickWin, you must start over. The same
applies if you have a project that you want to use as the basis for another
similar project of a different type.
Q. How do the keywords get colored in Microsoft Developer Studio?
A. By design, the keywords are colored without any knowledge of context.
Microsoft Developer Studio has no compiler parser. For example, if you have a
variable with the name "end", it will be colored because there is a valid
statement with the same name.
Q. Is syntax code coloring available for user-defined types?
A. No, the code coloring is not available for user-defined data types.
Q. Can I save the Watch or Variables windows in the debugger?
A. In the past you were able to save the various windows that the debugger used
as files. You can no longer do this with the Watch and Variables windows.
However, you can copy the content to the Clipboard and paste it in a file.
Q. How can I debug assumed-size arrays?
A. When you display an assumed-size array, it expands to display the first
element of the left-most index. If, for example, you have an array A(3), you
only see A(1) in the array expansion in the Watch, Quickwatch, and Variables
windows. If, on the other hand, you have a multidimensional array such as
A(2,2,2) when it is expanded, you see A(1,1,1), A(2,1,1), A(1,2,1), A(2,2,1).
Assumed-size arrays have many restrictions on them in source code. If you want
to look at the rest of the array, you need to use the Memory window.
Q. If you have a DO loop and set a breakpoint on the DO line, the breakpoint only
stops the first time through when debugging. For example, if I said DO N=1,100,
I would expect to stop on the line 100 times, yet I only stop there once. Why
does this occur?
A. A debug source line consists of zero or more assembly lines. When you set a
breakpoint on a source line, a breakpoint is set on the first instruction in a
group of assembly instructions. A simple DO loop consists of about a dozen
assembly lines. The first half sets up the variables and the last half
decrements the variables and checks for loop termination. When you set a
breakpoint on the source line, you are actually setting a breakpoint on the
initialization code. If you step through this in disassembly mode you can follow
what is happening. If you want to set a breakpoint that stops at each loop
iteration, you can do one of the following:
- Set the breakpoint on the first source line inside the DO loop.
- or -
- Select Disassembly from the View menu after the debugger has been started,
and set the breakpoint on the loop code.
Q. If the exception for floating divide-by-zero is set at "Stop if not handled"
in the Debug\Exceptions menu, the debugger does not stop even if divide-by-zero
is not handled. It does stop if "Stop always" is selected. Why?
A. The floating point divide-by-zero, and, in general, all floating point errors,
are handled by the system with the response of terminating whatever is causing
the error unless told otherwise. The debugger stops only if "Stop always" is
selected in Debug\Exceptions menu. This behavior is by design.
Q. If I type the name of a common block into the Watch window, I get an "Invalid
debug info" error. If I put slashes (/) around it, I get a syntax error. Why?
A. This is a known problem. You can watch common block local variables in the
Variables window. Common block variables that are not local cannot be watched.
Q. When I try to use PRINT * within InitialMenu, I get an error message. For
example: "An application error has occurred and an application error log is
being generated. Exception: access violation (0xc0000005), Address: 0x004100d8"
is returned and the program crashes.
INCLUDE 'FLIB.FI'
INCLUDE 'FLIB.FD'
PRINT*, 'Test'
END
LOGICAL*4 FUNCTION InitialMenu
INCLUDE 'FLIB.FD'
print*, 'Test'
InitialMenu = .TRUE.
END
A. The problem is that the connection between QuickWin and the I/O system isn't
there yet. Writing to an external file does work. InitialMenu should not attempt
to alter any QuickWin windows through I/O, graphics routines, OPENs, and
others.
Q. Can an erroneous mouse callback interface be flagged?
A. No. The compiler cannot detect errors caused by mismatches in the number or
types of the arguments to callback routines.
Q. Does the PEEKCHARQQ work in standard graphics or in a QuickWin application?
A. No. This is a known limitation. PEEKCHARQQ does not work in standard graphics
or in QuickWin applications.
Q. GETGTEXTEXTENT sometimes returns values greater than I expect. Why?
A. When the argument to GETGTEXTEXTENT is a variable, any trailing spaces are
included in the calculation.
Q. Is there a run-time function that returns a size of a type? Some of the Win32
API's need this information.
A. No. But you can get the information using the LOC function, as follows:
type mytype
integer a
real b
end type
type (mytype) dummyarray(2)
integer size
size= loc(dummyarray(2)) - loc(dummyarray(1))
print*, size
Q. Why does the following code compile and execute under Fortran PowerStation 1.0
but generates a compiler error message when compiled with Fortran PowerStation
4.0?
The error occurs when a function is declared with a varname*length, but a regular
variable is okay when it is declared that way.
INTEGER I*4 !This is okay
END
INTEGER FUNCTION X()*4 !This causes a compilation error
END
A. This is a known problem in Fortran PowerStation version 4.0. Avoid the
varname*length expression with the function declaration.
Q. Why does the RETURN statement cause a compiler error in a main Avoid program?
The following code compiles cleanly under Fortran PowerStation 1.0 but generates
a compilation error under PowerStation 4.0. This error occurs because of the use
of the RETURN statement in the main program.
i=1
RETURN !This line produces a compilation error
END
A. This is a known problem in Fortran PowerStation 4.0. To avoid the "Avoid the
error" message, remove the RETURN statement from the main Avoid program.
Q. Can I reference a program unit that contains a BLOCK DATA statement with an
EXTERNAL statement? For example:
block data MyFunc
common /g/ i
data i /1/
end
A. Fortran does not generate an external for the block data name for the previous
example "_MyFunc" in the .obj file. This feature might be useful when sharing
the DLL data. See the Readme section in Books Online for more information.
Q. I have a directory drive with a name that is just one letter long and the
Command fl32 /? doesn't work. Why?
A. If you have a single-letter directory on your drive, then the fl32 driver
program picks it up and thinks that it is a flag. For example:
Command line warning D2009: unrecognized flag /F ignored.
Command line warning D2009: unrecognized flag /L ignored.
Command line error D2003: missing source filename
If you have a root directory with a name that is one character long, then "fl32
/?" gives you an error like those above instead of giving you the Help page. Use
"fl32 -?" as a workaround.
Q. How can I get help regarding Windows API's?
A. If you are using the CD-ROM version of Fortran PowerStation 4.0, you can gain
access to Help for Windows API's by selecting Open Information Title from the
Help menu in Microsoft Developer Studio, and then choosing the Win32 Software
Development Kit title. You can also click the InfoViewer toolbar in Microsoft
Developer Studio to open this title.
Additional query words: 4.00
======================================================================
Keywords : kbsetup kbCompiler kbFL32 kbFortranPS kbLangFortran kbfaq
Technology : kbAudDeveloper kbFortranSearch kbZNotKeyword2 kbFORTRANPower32400NT
Version : :4.0
=============================================================================
THE INFORMATION PROVIDED IN THE MICROSOFT KNOWLEDGE BASE IS PROVIDED "AS IS" WITHOUT WARRANTY OF ANY KIND. MICROSOFT DISCLAIMS ALL WARRANTIES, EITHER EXPRESS OR IMPLIED, INCLUDING THE WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT SHALL MICROSOFT CORPORATION OR ITS SUPPLIERS BE LIABLE FOR ANY DAMAGES WHATSOEVER INCLUDING DIRECT, INDIRECT, INCIDENTAL, CONSEQUENTIAL, LOSS OF BUSINESS PROFITS OR SPECIAL DAMAGES, EVEN IF MICROSOFT CORPORATION OR ITS SUPPLIERS HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. SOME STATES DO NOT ALLOW THE EXCLUSION OR LIMITATION OF LIABILITY FOR CONSEQUENTIAL OR INCIDENTAL DAMAGES SO THE FOREGOING LIMITATION MAY NOT APPLY.
Copyright Microsoft Corporation 1986-2002.