Date: Mon, 2 Feb 1998 19:42:24 -0500 (EST)
Message-Id: <199802030042.TAA22344@delorie.com>
From: DJ Delorie <dj AT delorie DOT com>
To: andrewc AT rosemail DOT rose DOT hp DOT com
CC: djgpp-workers AT delorie DOT com
In-reply-to: <199802030001.AA211414084@typhoon.rose.hp.com> (message from
	Andrew Crabtree on Mon, 02 Feb 1998 16:01:24 PST)
Subject: Re: iostream concern
Precedence: bulk


> Right - I was just suggesting that since the standard (C9x draft anyway), 
> specifies that size_t is to be defined in stddef.h instead of stdio.h,
> that if we fix that there won't be a conflict.  I don't think we should
> actually use the gcc stddef, but it would be nice if it didn't
> crash things when bootstrapping the compiler.  

If you follow the spec to the letter, you can't use <stdio.h> without
including <stddef.h> first.  I put a lot of effort into making the
standard headers work without dependencies on each other without
violating the spec (much).  This is one such case.  They are designed
to work together.  They are not designed to work with any permutation
of replacement from another package.

DJGPP is pretty old now.  What's in gnu's stddef.h that djgpp's been
missing all these years?  If it's so important that it can't wait
until the next djgpp release, why hasn't anyone noticed the lack in
the last 8-10 years?