From owner-freebsd-current  Fri May  9 09:05:18 1997
Return-Path: <owner-current>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id JAA12141
          for current-outgoing; Fri, 9 May 1997 09:05:18 -0700 (PDT)
Received: from bitbox.follo.net (bitbox.follo.net [194.198.43.36])
          by hub.freebsd.org (8.8.5/8.8.5) with ESMTP id JAA12120
          for <freebsd-current@FreeBSD.ORG>; Fri, 9 May 1997 09:04:58 -0700 (PDT)
Received: (from eivind@localhost) by bitbox.follo.net (8.7.6/8.7.3) id SAA02595; Fri, 9 May 1997 18:04:14 +0200 (MET DST)
Date: Fri, 9 May 1997 18:04:14 +0200 (MET DST)
Message-Id: <199705091604.SAA02595@bitbox.follo.net>
From: Eivind Eklund <perhaps@yes.no>
To: Warner Losh <imp@village.org>
CC: bde@zeta.org.au, sos@sos.freebsd.dk, freebsd-current@FreeBSD.ORG,
        roberto@keltia.freenix.fr
In-reply-to: Warner Losh's message of Fri, 09 May 1997 09:40:47 -0600
Subject: Re: #include problem breaks current 
References: <199705081756.TAA06245@bitbox.follo.net>
	<199705072230.IAA21803@godzilla.zeta.org.au>
	<E0wPrmp-0000Os-00@rover.village.org>
Sender: owner-current@FreeBSD.ORG
X-Loop: FreeBSD.org
Precedence: bulk

> 
> In message <199705081756.TAA06245@bitbox.follo.net> Eivind Eklund writes:
> : I'll try to go through all the include files and at least document the
> : cases where they depend on each other.
> : Then we can discuss what to do about each case afterwards.
> 
> Right now I'd say the vast majority of cases are documented.  For
> example, see the connect man page.  It clearly shows that you have to
> include sys/types.h before sys/socket.h.  fts clearly shows that
> sys/types.h and sys/stat.h are requried to be included before fts.h.

Yeah.  I found some bad cases, but not many.  And I was thinking of a
collective documentation, to let us have the data to make a decision.

> : Sigh.  All this fuzz "just to close a PR"...
> 
> Not all PRs should be closed.  Many are clearly wrong.

IMHO, all PRs should be closed.  Not all PRs should be committed or
result in any action; but all should be closed.  If no corrective
action is necessary, close with "PR is wrong - closed."

Otherwise, we'll just end up with an infinite set of bad PRs - and
that is a Bad Idea.

Eivind.