Message-ID: <382EC055.8726D5BF@softhome.net>
Date: Sun, 14 Nov 1999 15:59:49 +0200
From: Laurynas Biveinis <lauras AT softhome DOT net>
X-Mailer: Mozilla 4.7 [en] (Win98; I)
X-Accept-Language: lt,en
MIME-Version: 1.0
To: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
CC: DJGPP Workers <djgpp-workers AT delorie DOT com>
Subject: Re: First alpha symlink patch
References: <Pine DOT SUN DOT 3 DOT 91 DOT 991114160607 DOT 20280A-100000 AT is>
Content-Type: text/plain; charset=iso-8859-4
Content-Transfer-Encoding: 7bit
Reply-To: djgpp-workers AT delorie DOT com

Eli Zaretskii wrote:
> I don't like static global variables: they are evil in recursive
> situations.

Are they harmful in this particular case ?

> What I had in mind was a situation when someon zips a DJGPP directory
> tree and then moves it to another machine, or uses some recursive copy
> program to do similar things.  If that program or that archiver don't
> copy or don't put the symlink files, you will have a subtly broken
> installation in the new place.

<Sigh> Yes... I tried - programs do not see symlinks. I really want to 
have a rare attribute combination, but if that causes problems, it will be 
reconsidered. I wanted that first symlink file test would be fast and sort 
out 95% of non symlink cases, and only remaining 5% or so files would be 
opened, readed first few charactes,  compared with '!<symlink>'... If I
change attribute combination to any other not involving ``system'' or
``hidden'', perfomance will drop. I think I will remove attribute checking
at all, because it virtually won't help.

Laurynas Biveinis