Message-Id: <200007041603.TAA19356@mailgw1.netvision.net.il>
Date: Tue, 04 Jul 2000 19:04:10 +0200
To: snowball3 AT bigfoot DOT com
X-Mailer: Emacs 20.6 (via feedmail 8.2.emacs20_6 I) and Blat ver 1.8.5b
From: "Eli Zaretskii" <eliz AT is DOT elta DOT co DOT il>
CC: djgpp-workers AT delorie DOT com
In-reply-to: <3961CDF5.18859.1F145B@localhost> (snowball3@bigfoot.com)
Subject: Re: DJGPP problem executing a script
References: <3961AE48 DOT 6825 DOT 87F40 AT localhost> (snowball3 AT bigfoot DOT com) <3961CDF5 DOT 18859 DOT 1F145B AT localhost>
Reply-To: djgpp-workers AT delorie DOT com
Errors-To: nobody AT delorie DOT com
X-Mailing-List: djgpp-workers AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk

> From: "Mark E." <snowball3 AT bigfoot DOT com>
> Date: Tue, 4 Jul 2000 11:43:49 -0400
> 
> > If it's __dosexec_find_on_path that's linked into Bash, it should have
> > seen forward slashes in PATH, since stock djgpp.env arranges for that.
> 
> Not in this case, because it's searching the PATH in the environment to be 
> exported to the new process/child/whatever which must use backslashes.

So PATH inside Bash does have forward slashes, but Bash forcibly
mirrors them back when it invokes external programs?  I thought it
only restored the one it found originally, when it started.

> > I think if we need to solve a local problem with scripts, it's
> > sufficient to solve it locally in script_exec.
> 
> Fair enough. Here's a patch that fixes the problem when used with Bash:

Seems okay to me.  Thanks!