Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!clyde.concordia.ca!uunet!snorkelwacker!apple!sun-barr!decwrl!henry.jpl.nasa.gov!elroy.jpl.nasa.gov!jpl-devvax!lwall
From: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Newsgroups: comp.unix.questions
Subject: Re: Bit pattern of S_IFREG (was Re: What is wrong with this code ?)
Message-ID: <6630@jpl-devvax.JPL.NASA.GOV>
Date: 19 Dec 89 02:55:40 GMT
References: <1156@nsscb.UUCP> <1323@mitisft.Convergent.COM> <2722@auspex.UUCP> <553@mwtech.UUCP> <2736@auspex.UUCP>
Reply-To: lwall@jpl-devvax.JPL.NASA.GOV (Larry Wall)
Organization: Jet Propulsion Laboratory, Pasadena, CA
Lines: 18

In article <2736@auspex.UUCP> guy@auspex.auspex.com (Guy Harris) writes:
: >I think I have seen on some older systems, that for regular files
: >you must test for both	(st_mode & S_IFMT) == S_IFREG
: >and *additionally*	(st_mode & S_IFMT) == 0.
: >
: >I don't know if this is still true today.
: 
: If it is true on any "V7 derivative" (BSD and S3/S5 count under this
: heading), it is due to a colossal screwup on the part of the vendor.  It
: may have been true under some emulation of V7 under "last V6", or
: something like that.  *I* certainly wouldn't bother testing for
: (st_mode & S_IFMT) being 0.

However, be advised that in BSD (or at least MORE/BSD), fstat() on a socket
returns (st_mode & S_IFMT) == 0.  This one bit me in perl.  Caveat Hackor.

Larry Wall
lwall@jpl-devvax.jpl.nasa.gov