GNU bug report logs - #55620
29.0.50; Eshell doesn't reset `eshell-in-pipeline-p' in subcommands

Previous Next

To add a comment to this bug, you must first unarchive it, by sending
a message to control AT debbugs.gnu.org, with unarchive 55620 in the body.
You can then email your comments to 55620 AT debbugs.gnu.org in the normal way.

Toggle the display of automated, internal messages from the tracker.

Message #5 received at submit <at> debbugs.gnu.org (full text, mbox):

From: Jim Porter <jporterbugs <at> gmail.com>
To: bug-gnu-emacs <at> gnu.org
Subject: 29.0.50; Eshell doesn't reset `eshell-in-pipeline-p' in subcommands
Date: Tue, 24 May 2022 19:20:59 -0700

From `emacs -Q -f eshell':

  ~ $ echo $eshell-in-pipeline-p
  ;; no output
  ~ $ echo ${echo $eshell-in-pipeline-p}
  ;; no output
  ~ $ echo ${echo $eshell-in-pipeline-p} | cat
  first

(Note: in the first two cases, `eshell-in-pipeline-p' is nil, so there's 
no output.) The last case is surprising: it's true that the inner `echo' 
invocation is in a pipeline, but the pipeline is a level above it 
(`first' means that when `eshell-in-pipeline-p' was evaluated, it was 
the first command in a pipeline). Since the inner `echo' is effectively 
in a subshell, it makes more sense for the value of 
`eshell-in-pipeline-p' to be nil for the last case.

While this is a seemingly-obscure issue, there is one practical effect 
that might bite users: currently, the Eshell implementation of `cat' 
only works when *not* in a pipeline. Usually, it'll just fall back to 
using the real system `cat' program if the Eshell version doesn't work, 
but on systems without `cat' (e.g. MS Windows), you get the following error:

  ~ $ echo ${cat file.txt} | echo
  Eshell’s ‘cat’ does not work in pipelines

Patch forthcoming momentarily; just getting a bug number...

Message #8 received at 55620 <at> debbugs.gnu.org (full text, mbox):

From: Jim Porter <jporterbugs <at> gmail.com>
To: 55620 <at> debbugs.gnu.org
Subject: Re: bug#55620: [PATCH] 29.0.50; Eshell doesn't reset
 `eshell-in-pipeline-p' in subcommands
Date: Tue, 24 May 2022 20:37:01 -0700

[Message part 1 (text/plain, inline)]

(Originally mis-sent with the wrong email address.)

Attached is a patch with tests to fix this. Note that I reset 
`eshell-in-pipeline-p' in two places:

1. `eshell-subcommand-bindings': This handles any case where a 
subcommand is invoked. This can happen with the following syntaxes: 
{CMD}, ${CMD}, and $<CMD>.

2. `eshell-command-to-value': This handles cases where a command is run 
synchronously. This can happen with some of the syntaxes mentioned 
above: {CMD} and ${CMD}. However, it can also happen for Lisp forms like 
(LISP) and $(LISP). (Most Lisp forms aren't likely to consult 
`eshell-in-pipeline-p', but if you were calling an Eshell built-in 
command using Lisp syntax, you might run into this issue.)

[0001-Reset-eshell-in-pipeline-p-when-interpolating-comman.patch (text/plain, attachment)]

Message #11 received at 55620 <at> debbugs.gnu.org (full text, mbox):

From: Lars Ingebrigtsen <larsi <at> gnus.org>
To: Jim Porter <jporterbugs <at> gmail.com>
Cc: 55620 <at> debbugs.gnu.org
Subject: Re: bug#55620: 29.0.50; Eshell doesn't reset `eshell-in-pipeline-p'
 in subcommands
Date: Wed, 25 May 2022 14:13:44 +0200

Jim Porter <jporterbugs <at> gmail.com> writes:

> Attached is a patch with tests to fix this. Note that I reset
> `eshell-in-pipeline-p' in two places:

Thanks; pushed to Emacs 29.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no

Previous Next

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