Date: Thu, 27 Aug 1998 14:26:29 +0300 (IDT)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
To: "Salvador Eduardo Tropea (SET)" <salvador AT inti DOT gov DOT ar>
cc: Deathstar <aresds AT usa DOT net>, djgpp AT delorie DOT com
Subject: Re: RHIDE Problem (Linker?)
In-Reply-To: <m0zBlWv-000S4TC@inti.gov.ar>
Message-ID: <Pine.SUN.3.91.980827142552.6326U-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Precedence: bulk


On Wed, 26 Aug 1998, Salvador Eduardo Tropea (SET) wrote:

> 1) The last line didn't have a CR+LF.

This shouldn't matter, right?  As far as I can see, the startup code
adds a trailing newline if it isn't there already.

> 2) He had a line with: "[rhide]          "
> 
> The 2 was the funniest and the most important because in this case the 
> section isn't really RHIDE!!!!

Seems like a simple change to c1loadef.c will fix this.  Could you
please submit such a change?

> Is that in the FAQ Eli?

No, I will add this.  This is the first time I ever hear about such
mutilation of DJGPP.ENV, so the problem probably doesn't happen too
much.

> Now for all the users editing *.env files: Please use RHIDE or SET Edit for 
> this task these editor will:

I think this only suggests yet again that USERS SHOULD NOT EDIT
djgpp.env.  It's too tricky, and doesn't have enough redundancy to
survive every weird mutilation out there.

Problems should be solved in the programs if possible, and DJGPP.ENV
should only be changed if you cannot reasonably fix a problem for all
the users (e.g., if a value of a variable depends on %DJDIR% that is
different on every machine).