Date:	Mon, 11 Oct 1999 17:17:12 +0200 (IST)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender:	eliz AT is
To:	pavenis AT lanet DOT lv
cc:	Robert Hoehne <robert DOT hoehne AT gmx DOT net>, djgpp-workers AT delorie DOT com
Subject:	Re: djconfig.sh for gdb-4.18
In-Reply-To:	<B0000104726@stargate.astr.lu.lv>
Message-ID:	<Pine.SUN.3.91.991011171326.29794C-100000@is>
MIME-Version:	1.0
Reply-To:	djgpp-workers AT delorie DOT com
X-Mailing-List:	djgpp-workers AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Mon, 11 Oct 1999 pavenis AT lanet DOT lv wrote:

> +CONFIG_SHELL=`echo 'echo $0' | bash.exe -`
> +case ${CONFIG_SHELL} in *.exe) ;;
> +	*) CONFIG_SHELL=${CONFIG_SHELL}.exe;
> +	   test -f ${CONFIG_SHELL} || exit 1
> +	   ;;
> +esac;
> +		
> +export CONFIG_SHELL		

This is a good idea, but there might be one caveat.  What does "echo $0" 
print on your machine, exactly?  Specifically, does it use forward- or
back-slashes?  If it uses back-slashes, Bash (or Make) could choke on 
such a name when it is later used.

Whether the $0 is printed with backslashes or not depends on many 
installation-dependent details.  So I suggest to pass $0 through a simple 
Sed script that explicitly converts any \ into /, before assigning it to 
CONFIG_SHELL.

- Raw text -