GNU bug report logs - #30460
27.0.50; `process-command' of a stopped process is t

Previous Next

Package: emacs;

Reported by: p.stephani2 <at> gmail.com

Date: Wed, 14 Feb 2018 22:01:02 UTC

Severity: normal

Found in version 27.0.50

Done: Philipp Stephani <p.stephani2 <at> gmail.com>

Bug is archived. No further changes may be made.

Full log


Message #8 received at 30460 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: p.stephani2 <at> gmail.com
Cc: 30460 <at> debbugs.gnu.org
Subject: Re: bug#30460: 27.0.50; `process-command' of a stopped process is t
Date: Sat, 17 Feb 2018 15:25:50 +0200
> Date: Wed, 14 Feb 2018 23:00:26 +0100
> From: p.stephani2 <at> gmail.com
> 
> 
> (process-command
>    (make-process :name "sleep" :command '("sleep" "1h") :stop t))
> 
> returns t.
> 
> This is not documented (the docstring and the manual say that this
> should only happen for a pipe, network, or serial connection), and
> breaks M-x list-processes, which assumes that the return value of
> `process-command' is a list for an ordinary process.

":stop t" currently has no effect on a "real" process running a
command, because it is not generally possible to start a process in a
stopped state.  So it's a Good Thing list-processes fails for such
bogus "processes" ;-)

We could fix this by either ignoring :stop in make-process, or by
signaling an error.  The simplest change is just delete this snipper
from Fmake_process:

  if (tem = Fplist_get (contact, QCstop), !NILP (tem))
    pset_command (XPROCESS (proc), Qt);

The correct command was already set by that time, so the above just
overwrites the correct expected value.




This bug report was last modified 6 years and 35 days ago.

Previous Next


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