Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.3 alpha 4/3/85; site ukma.UUCP
Path: utzoo!watmath!clyde!cbosgd!ukma!sambo
From: sambo@ukma.UUCP (Inventor of micro-S)
Newsgroups: net.unix
Subject: Re: lex query: "Too many right contexts"
Message-ID: <1921@ukma.UUCP>
Date: Fri, 28-Jun-85 01:47:53 EDT
Article-I.D.: ukma.1921
Posted: Fri Jun 28 01:47:53 1985
Date-Received: Sat, 29-Jun-85 01:20:31 EDT
References: <1527@dalcs.UUCP>
Reply-To: sambo@ukma.UUCP (Inventor of micro-S)
Organization: Univ. of KY Mathematical Sciences
Lines: 25

In article <1527@dalcs.UUCP> silvert@dalcs.UUCP (Bill Silvert) writes:
>Whenever I use a right context in a long lex program I get the message:
>	(Error) Too man right contexts
>Specifically, I am working on a cross-reference program for f77,
>which has lines like:
>	WRITE	;
>to ignore reserved words . . .

In the lexical analyzer for the compiler I am writing, I used to get some
message about exceeding some limit.  I never was able to figure out how to
change that limit.  (If I tried some number close to 980, I would get this
message, and if I tried the next number, lex would crash.  I was using a
Vax 11/750, and still am.)  What I ended up doing was to treat all the keywords
as normal identifiers, and then call a function to figure out if it was actually
a keyword or an identifier.  I suspect that having lex recognize keywords
directly makes it consume large amounts of memory, making it easy to exceed
its limits.
-----------------------------------------
Samuel A. Figueroa, Dept. of CS, Univ. of KY, Lexington, KY  40506-0027
ARPA: ukma!sambo<@ANL-MCS>, or sambo%ukma.uucp@anl-mcs.arpa,
      or even anlams!ukma!sambo@ucbvax.arpa
UUCP: {ucbvax,unmvax,boulder,oddjob}!anlams!ukma!sambo,
      or cbosgd!ukma!sambo

	"Micro-S is great, if only people would start using it."