Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10 5/3/83; site gatech.UUCP
Path: utzoo!linus!security!genrad!grkermit!masscomp!clyde!akgua!gatech!arnold
From: arnold@gatech.UUCP
Newsgroups: net.lang.c
Subject: The "entry" keyword
Message-ID: <3459@gatech.UUCP>
Date: Mon, 23-Jan-84 11:59:44 EST
Article-I.D.: gatech.3459
Posted: Mon Jan 23 11:59:44 1984
Date-Received: Fri, 27-Jan-84 06:00:34 EST
Organization: Georgia Tech School of ICS, Atlanta
Lines: 38


	I quote from K&R, page 180, Reference Manual section 2.3, KEYWORDS
(*'s indicate boldface):

	The following identifers are reserved for use as keywords, and may
not be used otherwise:

	(... C reserved words, including the identifier "entry")

	The *entry* keyword is not currently implemented by any compiler but
is reserved for future use.  Some implementations also reserve the words
*fortran* and *asm*.

end quote.

	My question is this: Is "entry" still reserved?  According to the
System V documentation, it is.  BUT, the source for the System V routine
hsearch(3) uses "entry" as a structure tag (i.e. a normal identifier), and
neither that compiler nor the 4.1BSD compiler complains in the least
(neither did the System IV compiler, which is where hsearch came from
originally.  [Yes I have actually worked on a system running Unix 4.0]).

	So, howcum?  If "entry" is no longer reserved, the documentation
should say so.  If it still is, the compiler(s) should complain.  I find it
hard to believe that this has gone totally unnoticed by anybody!

	I found this when trying to port hsearch to a locally written compiler
for Pr1me computers.  That compiler follows K&R, and naturally died miserably
when it hit the "entry" used as an identifier.  (It's recursive-descent to
boot, which makes life even more fun [sigh].)

	Thanks in advance,

		Arnold Robbins
-- 
Arnold Robbins
CSNET: arnold@gatech		ARPA: arnold.gatech@CSNet-relay
UUCP:	...!{akgua,allegra,rlgvax,sb1,unmvax,ulysses,ut-sally}!gatech!arnold