X-Authentication-Warning:	delorie.com: mail set sender to djgpp-workers-bounces using -f
Date:	Wed, 12 Jan 2005 22:43:19 +0200
From:	"Eli Zaretskii" <eliz AT gnu DOT org>
Sender:	halo1 AT zahav DOT net DOT il
To:	Paul Eggert <eggert AT CS DOT UCLA DOT EDU>
Message-ID:	<01c4f8e7$Blat.v2.2.2$7e7754a0@zahav.net.il>
X-Mailer:	emacs 21.3.50 (via feedmail 8 I) and Blat ver 2.2.2
CC:	djgpp-workers AT delorie DOT com, bug-bison AT gnu DOT org
In-reply-to:	<87is62o3io.fsf@penguin.cs.ucla.edu> (message from Paul Eggert on
	Wed, 12 Jan 2005 11:23:27 -0800)
Subject:	Re: ANNOUNCE: port of alpha bison-2.0
References:	<41E44BA6 DOT 16849 DOT 47FF9F7 AT localhost>
	<01c4f871$Blat.v2.2.2$89afa8c0 AT zahav DOT net DOT il>
	<87is62o3io DOT fsf AT penguin DOT cs DOT ucla DOT edu>
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

> Cc: djgpp-workers AT delorie DOT com, bug-bison AT gnu DOT org
> From: Paul Eggert <eggert AT CS DOT UCLA DOT EDU>
> Date: Wed, 12 Jan 2005 11:23:27 -0800
> 
> "Eli Zaretskii" <eliz AT gnu DOT org> writes:
> 
> > Could someone of the Bison maintainers please tell why Juan's patches
> > were rejected?
> 
> I don't recall the details, but I do recall a lot of proposed changes
> to subpipe.c that didn't seem appropriate at the time.  Perhaps this
> could be revisited by sending the small patches that you mentioned to
> bison-patches.

Thanks.

However, if it's true that Bison maintainers decided to drop support
for the DJGPP port, I'd think twice before starting to work on such
patches.

And what about the minor changes to other places, like output.c and
system.h?  Can these be accepted now?  I attach them below for your
convenience.


diff -apruNU5 bison-2.0.orig/src/output.c bison-2.0/src/output.c
--- bison-2.0.orig/src/output.c	2004-12-17 19:56:44.000000000 +0000
+++ bison-2.0/src/output.c	2005-01-10 02:46:34.000000000 +0000
@@ -579,10 +579,15 @@ output_skeleton (void)
   fputs ("m4_divert_push(0)dnl\n", out);
   xfclose (out);
 
   /* Read and process m4's output.  */
   timevar_push (TV_M4);
+#if !HAVE_WORKING_VFORK && !HAVE_WORKING_FORK
+  /* On non posix systems that lack vfork or vork
+     m4 must be explicitly spwaned to process bison output. */
+  run_m4 (filter_fd);
+#endif
   in = fdopen (filter_fd[1], "r");
   if (! in)
     error (EXIT_FAILURE, get_errno (), "fdopen");
   scan_skel (in);
   xfclose (in);
diff -apruNU5 bison-2.0.orig/src/system.h bison-2.0/src/system.h
--- bison-2.0.orig/src/system.h	2004-12-09 05:41:24.000000000 +0000
+++ bison-2.0/src/system.h	2005-01-10 02:46:34.000000000 +0000
@@ -174,13 +174,20 @@ do {								\
   /* VMS. */
 # define TAB_EXT	"_tab"
 # define OUTPUT_EXT	".output"
 #else /* ! VMS */
 # ifdef MSDOS
-   /* MS DOS. */
-#  define TAB_EXT	"_tab"
-#  define OUTPUT_EXT	".out"
+#  ifdef __DJGPP__
+    /* DJGPP. */
+#   include <unistd.h>
+#   define TAB_EXT	((pathconf(NULL, _PC_NAME_MAX) > 12) ? ".tab" : "_tab")
+#   define OUTPUT_EXT	((pathconf(NULL, _PC_NAME_MAX) > 12) ? ".output" : ".out")
+#  else /* ! __DJGPP__ */
+    /* MS DOS. */
+#   define TAB_EXT	"_tab"
+#   define OUTPUT_EXT	".out"
+#  endif /* ! __DJGPP__ */
 # else /* ! MSDOS */
   /* Standard. */
 #  define TAB_EXT	".tab"
 #  define OUTPUT_EXT	".output"
 # endif /* ! MSDOS */

- Raw text -