Date:	Mon, 6 Dec 1999 19:08:04 +0200 (IST)
From:	Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender:	eliz AT is
To:	Juan Manuel Guerrero <ST001906 AT HRZ1 DOT HRZ DOT TU-Darmstadt DOT De>
cc:	djgpp AT delorie DOT com
Subject:	Re: Support of BISON in AUTOCONF/AUTOMAKE
In-Reply-To:	<944B391AAA@HRZ1.hrz.tu-darmstadt.de>
Message-ID:	<Pine.SUN.3.91.991206185933.21836E-100000@is>
MIME-Version:	1.0
Reply-To:	djgpp AT delorie DOT com
X-Mailing-List:	djgpp AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Mon, 6 Dec 1999, Juan Manuel Guerrero wrote:

> All this can be used as a guide line for how to properly implement
> support for MSDOS ports of BISON/BYACC.
> It should be clear that changes to autoconf AND automake will be needed.

This has come up before.  Someone even submitted the necessary changes to 
the maintainers of Autoconf and Automake, and IIRC, they were rejected, 
for some reason I cannot recall.

In any case, the patches need eventually be submitted to the package 
maintainers, again.

> IMPORTANT:
> To avoid the posibility of overwritting existing y.tab.[ch] files
> during the file name test, AC_PROG_YACC will invoke bison with
> the -b option (file prefix). Unfortunatly the DJGPP port of bison
> has a bug that will not allow to use this option.

Why should there be any existing y.tab.* files?  I think the test should 
be done in a way that makes sure no other file in the distribution has 
the same name as the one created by Bison.

> In autoconf lines like the following one are used:
>   if ! test -z "$COMSPEC$ComSpec"; then
> I would suggest:
>   if test "x$OSTYPE" = xMSDOS; then
> OSTYPE is always set  to MSDOS, at least by bash 1.14.
> and will clearly identify MSDOS code.

However, $COMSPEC is IMHO better, since it is independent of the Bash 
features.

- Raw text -