Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!watmath!clyde!caip!brl-adm!brl-smoke!gwyn
From: gwyn@brl-smoke.ARPA (Doug Gwyn )
Newsgroups: net.unix
Subject: Re: improved IEEE 1003.1 mkdir, rmdir for UNIX System V
Message-ID: <2010@brl-smoke.ARPA>
Date: Tue, 8-Jul-86 00:09:34 EDT
Article-I.D.: brl-smok.2010
Posted: Tue Jul  8 00:09:34 1986
Date-Received: Wed, 9-Jul-86 01:44:35 EDT
References: <1981@brl-smoke.ARPA> <438@meccts.UUCP>
Reply-To: gwyn@brl.arpa (Doug Gwyn (VLD/VMB) <gwyn>)
Organization: Ballistic Research Lab (BRL), APG, MD.
Lines: 19

In article <438@meccts.UUCP> ahby@meccts.UUCP (Shane P. McCarron) writes:
-In article <1981@brl-smoke.ARPA> gwyn@BRL.ARPA writes:
->This function is compatible with the definition in the
->IEEE 1003.1
-Since this standard is not finished, and has only been published in a
-Trial Use Document, is this compatible with the trial use, or with the
-in progress document being developed by the working committee.

Trial use, but I expect this interface won't change much.
Note that it is not a precisely conforming implementation,
since it cheats on some of the error conditions.  If
/bin/mkdir and /bin/rmdir return distinct exit status for
different failure modes, this could be improved.

Also note that, as Chris Torek pointed out, it would be
better to move the SIGCLD reset to before the fork().

I have a similar rename() implementation if anyone needs it
and can't figure out how to modify rmdir.c to get it.