Newsgroups: comp.os.msdos.djgpp
From: abostick AT netcom DOT com (Alan Bostick)
Subject: New to djgpp -- curious compilation problem
Message-ID: <abostickE4BLn0.5EE@netcom.com>
Organization: Arrogant Opinions 'R' Us
Date: Mon, 20 Jan 1997 18:49:48 GMT
Lines: 66
Sender: abostick AT netcom19 DOT netcom DOT com
To: djgpp AT delorie DOT com
DJ-Gateway: from newsgroup comp.os.msdos.djgpp

I'm new to djgpp and fairly new to C++ and have run into what seems to 
me to be an odd compilation problem.

I'm trying to learn the language from Stanley Lippman's C++ PRIMER, 2nd
edition, and have downloaded Lippman's examples from his FTP site.  I
have discovered, when compiling his examples, that djgpp chokes on the
following construction:
	enum Boolean {false = 0, true};

For example, consider the following test program:


#include <iostream.h>

main()
{
        enum Boolean { false = 0, true };
        if (false)
        {
                cout << "Something is seriously wrong here!\n";
        }
        else // if !false
        {
                cout << "This is what you ought to see.\n";
        }

        return 0;
}

When I compile this with djgpp, the following happens:

C:\USR\ABOSTICK\CXX>gxx -o test.exe test.cc
test.cc: In function `int main()':
test.cc:5: parse error before `false'

C:\USR\ABOSTICK\CXX>

I'm running djgpp version 2, calling v2.7.2.1 of GCC.

When I compile the same program with GCC version 2.5.8 under SunOS 
at one of my ISP's machines, it compiles smoothly and executes without
problems.

I've also found that if I change line 5 of my test program to read
	enum Boolean { False = 0, True };
djgpp compiles it without problems.

Plainly djgpp is ascribing special meaning to the token 'false' (and
probably 'true' as well).  Is it supposed to be doing so?

I notice from some other minor compilation errors when compiling Lippman's
example code that my version of djgpp uses a later evolution of the ANSI
C++ standard than his text does.  I infer from this that it is possible
that my problem is also due to evolving standards.

Is that what's going on?  That is, does v2.7.2.1 of GCC use a later
definition of C++ than v2.5.8, one in which 'false' and 'true' are
reserved words or something similar?

Or is something else going on?  If so, what?

-- 
Alan Bostick               | To achieve harmony in bad taste is the height
mailto:abostick AT netcom DOT com | of elegance.
news:alt.grelb             |      Jean Genet
http://www.alumni.caltech.edu/~abostick