Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site umcp-cs.UUCP
Path: utzoo!linus!decvax!genrad!mit-eddie!godot!harvard!seismo!umcp-cs!chris
From: chris@umcp-cs.UUCP (Chris Torek)
Newsgroups: net.lang.c
Subject: Re: Read-only Strings
Message-ID: <636@umcp-cs.UUCP>
Date: Fri, 2-Nov-84 23:06:52 EST
Article-I.D.: umcp-cs.636
Posted: Fri Nov  2 23:06:52 1984
Date-Received: Sun, 4-Nov-84 05:01:39 EST
References: <9498@watmath.UUCP>
Organization: U of Maryland, Computer Science Dept., College Park, MD
Lines: 33

Another good(?) reason for readonly strings is that then the compiler
could use the same address for "equal" strings; that is, if I write

	printf("%d\n", x);
	.
	.
	.
	printf("%d\n", y);

the compiler might produce

		.data
	L21:	.asciz	"%d\n"
		.text
		pushl	x	# well not really but...
		pushal	L21
		calls	$2,printf
		.
		.
		.
		pushl	y
		pushal	L21
		calls	$2,printf

Many FORTRAN compilers use similar tricks when passing constants to
subroutines or functions...  I think I once used the "modify-the-constant"
trick to change the WRITE unit number, just for fun.
-- 
(This mind accidently left blank.)

In-Real-Life: Chris Torek, Univ of MD Comp Sci Dept (301) 454-7690
UUCP:	{seismo,allegra,brl-bmd}!umcp-cs!chris
CSNet:	chris@umcp-cs		ARPA:	chris@maryland