Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site harvard.ARPA
Path: utzoo!linus!philabs!cmcl2!seismo!harvard!draves
From: draves@harvard.ARPA (Richard Draves)
Newsgroups: net.bugs.4bsd
Subject: limit on C declarations
Message-ID: <452@harvard.ARPA>
Date: Sun, 10-Mar-85 02:58:24 EST
Article-I.D.: harvard.452
Posted: Sun Mar 10 02:58:24 1985
Date-Received: Tue, 12-Mar-85 20:19:44 EST
Distribution: net
Organization: Aiken Computation Laboratory, Harvard
Lines: 19

There seems to be some sort of limit to how complicated a
C type may become (in our 4.2 cc):

	main() {
	    int a, **** **** **** **b; /* no error yet */
	    a =    **** **** **** **b; /* illegal indirection */
	}

I can understand the implementors' decision to put such a limit on
types, presumably to eliminate the need for some expensive data structure.
However, this should be documented in the man page (it isn't in ours),
and overflow should be reported as such, while compiling the declaration.

Rich
-- 

	"a picture in the head is a gory murder in an art gallery"

					-- Stephen Kosslyn