GNU bug report logs - #33847
27.0.50; emacsclient does not find server socket

Previous Next

Package: emacs;

Reported by: Ulrich Mueller <ulm <at> gentoo.org>

Date: Sun, 23 Dec 2018 09:49:01 UTC

Severity: normal

Tags: patch

Merged with 41707

Found in version 27.0.50

Fixed in version 28.1

Done: Lars Ingebrigtsen <larsi <at> gnus.org>

Bug is archived. No further changes may be made.

Full log


Message #126 received at 33847 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Lars Ingebrigtsen <larsi <at> gnus.org>
Cc: teika <at> gmx.com, eggert <at> cs.ucla.edu, 33847 <at> debbugs.gnu.org, ulm <at> gentoo.org
Subject: Re: bug#33847: 27.0.50; emacsclient does not find server socket
Date: Thu, 22 Jul 2021 20:30:20 +0300
> From: Lars Ingebrigtsen <larsi <at> gnus.org>
> Cc: eggert <at> cs.ucla.edu,  teika <at> gmx.com,  33847 <at> debbugs.gnu.org,  ulm <at> gentoo.org
> Date: Thu, 22 Jul 2021 19:05:11 +0200
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> > Looks like you didn't attach the right patch?
> 
> Yup; 2nd attempt now.

Thanks.

It looks like the w32-related aspect of this is easier to resolve than
I originally thought, because almost all of the changes are in the
SOCKETS_IN_FILE_SYSTEM part, which is not compiled for the MS-Windows
build.  The only potential issues I spotted are:

> --- a/lib-src/emacsclient.c
> +++ b/lib-src/emacsclient.c
> @@ -80,6 +80,7 @@ Copyright (C) 1986-1987, 1994, 1999-2021 Free Software Foundation, Inc.
>  #include <sys/stat.h>
>  #include <unistd.h>
>  
> +#include <acl.h>  <<<<<<<<<<<<<<<<<<<<<<<<<<<
>  #include <filename.h>
>  #include <intprops.h>
>  #include <min-max.h>

We cannot unconditionally include acl.h on MS-Windows, so this has to
be #ifdef'ed away.

> @@ -91,6 +92,10 @@ Copyright (C) 1986-1987, 1994, 1999-2021 Free Software Foundation, Inc.
>  # pragma GCC diagnostic ignored "-Wformat-truncation=2"
>  #endif
>  
> +#ifndef O_PATH
> +# define O_PATH O_SEARCH <<<<<<<<<<<<<<<<<<<<<<<<<
> +#endif

This is probably harmless, but I'd like to #ifdef it away as well.

> diff --git a/lib/gnulib.mk.in b/lib/gnulib.mk.in
> index 07736f9b8b..0b9aaf6d9e 100644
> --- a/lib/gnulib.mk.in
> +++ b/lib/gnulib.mk.in
> @@ -98,6 +98,7 @@
>  #  fcntl \
>  #  fcntl-h \
>  #  fdopendir \
> +#  file-has-acl \
>  #  filemode \
>  #  filename \
>  #  filevercmp \
> @@ -1788,6 +1789,16 @@ EXTRA_libgnu_a_SOURCES += fdopendir.c
>  endif
>  ## end   gnulib module fdopendir
>  
> +## begin gnulib module file-has-acl
> +ifeq (,$(OMIT_GNULIB_MODULE_file-has-acl))
> +
> +libgnu_a_SOURCES += file-has-acl.c
> +
> +EXTRA_DIST += acl-internal.h
> +
> +endif
> +## end   gnulib module file-has-acl
> +
>  ## begin gnulib module filemode
>  ifeq (,$(OMIT_GNULIB_MODULE_filemode))
>  

We need to add stuff to nt/gnulib-cfg.mk to omit building the
file-has-acl.c file on MS-Windows.  (I can do that after the rest is
installed.)

The only other worry is whether we indeed have ways to test this
thoroughly enough on Posix systems.




This bug report was last modified 3 years and 225 days ago.

Previous Next


GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.