Date: Mon, 24 Nov 1997 21:44:58 -0500 (EST)
Message-Id: <199711250244.VAA29269@delorie.com>
From: DJ Delorie <dj AT delorie DOT com>
To: k3040e4 AT c210 DOT edvz DOT uni-linz DOT ac DOT at
CC: djgpp-workers AT delorie DOT com
In-reply-to: <199711242155.QAA06271@delorie.com> (message from Oberhumer
	Markus on Mon, 24 Nov 1997 22:43:42 +0100 (MET))
Subject: Re: <sys/stat.h> constants
Precedence: bulk


> > There are no "standard ones"
> 
> I'd bet you have troubles finding a S_IFDIR != 0x4000 or S_ISUID != 04000.

I have no problems finding examples.  SGI's S_ISUID is 0x0800.

> Allegro uses neither stat() nor access(). Which are the libaries
> you are concerned about ?

OK, bad example.  I still think it's a bad idea to break compatibility
"just because" when there's no good reason to do so.  Programs that
hard-code values, when macros defined by ANSI and POSIX are available,
are broken.  Changing DJGPP to support broken programs, at the risk of
breaking working programs and libraries, is unacceptable.

> BTW, I stronly suggest rebuilding most packages with 2.02 ASAP anyway, at
> least v2gnu. Call it the ultimate beta test.

Go ahead.  Unless they're broken now, and 2.02 can fix them, I see no
need to replace the simtel versions, though.