Date: Sat, 7 Sep 2002 10:32:23 -0400
Message-Id: <200209071432.g87EWN013847@envy.delorie.com>
X-Authentication-Warning: envy.delorie.com: dj set sender to dj AT delorie DOT com using -f
From: DJ Delorie <dj AT delorie DOT com>
To: djgpp AT delorie DOT com
In-reply-to: <3d7a0313$0$8699$79c14f64@nan-newsreader-03.noos.net>
	(julian AT doomworld DOT com)
Subject: Re: No new line at end of file?
References: <3d71a27d DOT 23705885 AT news DOT nzwide DOT ihug DOT co DOT nz> <mqb8nuoav2evvgeefdoj9r2gdfa0vqmk71 AT 4ax DOT com> <3d746b3c DOT 3046168 AT news DOT nzwide DOT ihug DOT co DOT nz> <Xns927E8AC415B4Craf256com AT 213 DOT 180 DOT 128 DOT 20> <3d795645$0$24320$79c14f64 AT nan-newsreader-02 DOT noos DOT net> <200209070134 DOT g871YGP03030 AT envy DOT delorie DOT com> <001f01c25614$e8cd52a0$2ca0c6d4 AT noos DOT fr> <200209070229 DOT g872TOl03706 AT envy DOT delorie DOT com> <3d7a0313$0$8699$79c14f64 AT nan-newsreader-03 DOT noos DOT net>
Reply-To: djgpp AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


> Sorry. I thought this were a conversation, not a word-playing contest.

If you're going to assume the worst of me, I'm going to stop replying.
I'm willing to try to explain the situation, but only if you're
willing to meet me half way.

> Getting rid of the newline requirement would exactly be a "gcc-specific
> feature that *conflicts* with the ANSI standard" that's exactly what I said,
> *read*. The real thing we do not agree on is the fact this would be a RANDOM
> feature.

Not quite.  The -ansi switch is intended for those cases where a
strictly conforming ANSI program can NOT be compiled with gcc because
some gcc extension interferes with it.  For example, gcc allows the
keyword "asm" but a strictly conforming program may use "asm" as a
variable name.  Without the -ansi switch, attempts to compile this
standards-conforming program would fail.

The case where a file does not have a newline at the end is different,
because a program that does not have such a newline is NOT a
conforming program.  Whether gcc chooses to allow such a
non-conforming program or not has nothing to do with the -ansi switch.

> Do you consider BACKWARD COMPATIBILITY as  a random feature?

Relative to the operation of the -ansi switch and the ANSI/ISO
standards, yes.  Perhaps the -traditional option would be more
appropriate.  I'm not saying that backward compatibility is bad, I'm
saying that the -ansi switch is the wrong place to put it.

> But who am I to discuss this issue? You're right, I'm wrong and
> there is no argumentation possible.

That is completely wrong.  You, as a user, have every right to discuss
how gcc deals with backward compatibility.  However, it is useless to
do so here (this is the djgpp forum, not the gcc forum), and it's
useless to try to persuade folks that -ansi should control it (that's
an inappropriate option for this purpose).  Go to http://gcc.gnu.org/
and find the gcc AT gcc DOT gnu DOT org mailing list and explain the problem
you're having and see if they have some suggestions.

> Again, I'm perfectly aware this is a GCC issue but does it mean we
> can't discuss such choices?

It doesn't mean you can't discuss it.  It just means that it's
pointless to do so *here* because we can't do anything about it.

> What is the open-source community about? Closed-minded people in
> secret rooms that decide for the world?

All of the sources to both gcc and djgpp are available, and patches
are welcome as long as they don't conflict with other goals of the
projects.  And as with all open source projects, you need to send the
patch to the right mailing list.  *This* forum is the wrong place to
complain about gcc-specific features, simply because we're not the
ones doing gcc development.

> Should I take the source and make an alternate compiler that only
> modifies a couple lines or is it more clever to discuss the issue?

It would be far more clever if you discussed the issue with the people
who are actually working on the gcc sources.

> Now flame me for being insolent, for disrespecting the king, I don't
> expect anything out of this message anyway. I've seen enough
> electronic communities to recognize the one-word-answering
> chief. And I know he never listens... at least not in public.

Well, I'm the chief *here* and you've gotten far more than one word,
and obviously I'm listening and TRYING to help you by letting you know
why the software works the way it works, and where to go with your
questions to hopefully get better answers.

We've tried to answer your questions as best we can.  If you don't
like the answers, there's little else we can do for you.