Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site dayton.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxn!ihnp4!stolaf!mmm!dicomed!dayton!sjm
From: sjm@dayton.UUCP (Steven J. McDowall)
Newsgroups: net.unix-wizards
Subject: Re: SYS V unlink() call
Message-ID: <686@dayton.UUCP>
Date: Tue, 18-Feb-86 09:40:36 EST
Article-I.D.: dayton.686
Posted: Tue Feb 18 09:40:36 1986
Date-Received: Fri, 21-Feb-86 05:20:04 EST
References: <683@dayton.UUCP> <4980@alice.uUCp>
Reply-To: sjm@dayton.UUCP (Steven J. McDowall)
Organization: Dayton Hudson Dept. Stores Mpls, MN
Lines: 30
Summary: 

In article <4980@alice.uUCp> ark@alice.UucP (Andrew Koenig) writes:
>
>Why can't you believe this is what is really meant?
>

Ok ok.. As the original author:

1)  I believe! I believe!
2)  It seemed to me (still does) that this restriction is
    a hinder to developing major systems that uses directories
    for structuring information. 
3)  Yes, I know* that I can do a system() call to issue either
    a mkdir or rmdir.. My point is that, if a program can
    perform the that function (via system) then it would seem
    logical for the OS to provide the service w/o the overhead
    of invoking the system call. (system() is pretty expensive, 
    isn't it?)

    Anyway, thanks for the answers.. It does* make sense that
    you can't arbitrarily delete directories, thought I would
    have thought that if unlink() (I do like BSD's rmdir, mkdir)
    would allow us to remove directories, it would take care of
    all the checking for us..Same of course if we could
    do a mkdir()....


-- 
Steven J. McDowall	
Dayton-Hudson Dept. Store. Co.		UUCP: ihnp4!rosevax!dayton!sjm
700 on the Mall				ATT:  1 612 375 2816
Mpls, Mn. 55408