Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.1 6/24/83; site pur-phy.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxl!ihnp4!inuxc!pur-ee!CS-Mordred!Pucc-H:Physics:crl
From: crl@pur-phy.UUCP (Charles LaBrec)
Newsgroups: net.bugs.2bsd
Subject: Re: Any one using symbolic links under 2.9?
Message-ID: <1401@pur-phy.UUCP>
Date: Fri, 24-Aug-84 02:30:51 EDT
Article-I.D.: pur-phy.1401
Posted: Fri Aug 24 02:30:51 1984
Date-Received: Sat, 25-Aug-84 06:21:00 EDT
References: <2101@usceast.UUCP>
Organization: Purdue University Physics Dept.
Lines: 20

Yes, we have an 11/24 running 2.9 with UCB_SYMLINKS defined, and it
works, but a few comments:

1) ln must be recompiled (and check to see if whoami.h or localopts.h
	is included, so as to get the above define. (same for ls, rm, 
	to name a few)

2) fsck must be fixed--there is a #define for SPECIAL (I think, it's
	been a while and I don't currently have access to the source).
	It is a big conditional and LNK is one of them.  Well, this
	makes fsck think a symlink takes no blocks, so it misses the
	block allocated to the symlink.

3) Just remembered, ls does not initialize the structure member l_linkto
	(I think) to NULL from malloc, so sometimes an ls -l prints some
	garbage as the linked-to field of a non-symlink.

Charles LaBrec
UUCP:		pur-ee!Physics:crl, purdue!Physics:crl
INTERNET:	crl @ pur-phy.UUCP