GNU bug report logs - #80036
ispell-region cannot be suspended

Previous Next

To reply to this bug, email your comments to 80036 AT debbugs.gnu.org.

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

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

From: Sébastien Hinderer <sebastien.hinderer <at> ens-lyon.org>
To: bug-gnu-emacs <at> gnu.org
Subject: ispell-region cannot be suspended
Date: Fri, 19 Dec 2025 10:36:21 +0100

Dear Emacs Maintainers,

I am using GNU Emacs version 30.2.

To reproduce, run LANG=C emacs -Q and then type:
Hel<SPC>lo
(That is, Hello with a space between the two letters l.)
Then type M-x ispell-message
The spell-checker should complain about the misspelled word Hel.
Now type C-g

Observed behaviour: "Message send aborted" is displayed in the
minibuffer.

Expected behaviour: the minibuffer should display:

Spell-checking suspended; use C-u M-$ to resume"

Notes:

1. Although the minibuffer does not display the text I would expect,
typing C-u M-$ _does_ seem to resume the spell checking.

2. If one runs:

LANG=C emacs -Q

And then types:

Hel<SPC>lo <M-SPC> C-a M-x ispell-region

the spell checker will stumblein thesame way as previously but, if
interrupted with C-g, it will display in the minibuffer the message
explaining how to resume spell checking.

Message #8 received at 80036 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Sébastien Hinderer <sebastien.hinderer <at> ens-lyon.org>
Cc: 80036 <at> debbugs.gnu.org
Subject: Re: bug#80036: ispell-region cannot be suspended
Date: Sat, 20 Dec 2025 09:33:32 +0200

> Date: Fri, 19 Dec 2025 10:36:21 +0100
> From: Sébastien Hinderer <sebastien.hinderer <at> ens-lyon.org>
> 
> Dear Emacs Maintainers,
> 
> I am using GNU Emacs version 30.2.
> 
> To reproduce, run LANG=C emacs -Q and then type:
> Hel<SPC>lo
> (That is, Hello with a space between the two letters l.)
> Then type M-x ispell-message
> The spell-checker should complain about the misspelled word Hel.
> Now type C-g
> 
> Observed behaviour: "Message send aborted" is displayed in the
> minibuffer.
> 
> Expected behaviour: the minibuffer should display:
> 
> Spell-checking suspended; use C-u M-$ to resume"

If you look at the *Messages* buffer, you will see that the message
you expected was in fact shown, but it was immediately overwritten by
"Message send aborted".

ispell-message is mainly meant to be used in a hook of sending a
message, which is why it is important to tell the user that the
message will not be sent.  I can suggest the patch below to avoid
losing the information about the possible resumption of
spell-checking:

diff --git a/lisp/textmodes/ispell.el b/lisp/textmodes/ispell.el
index 6476a16..4fcc6c6 100644
--- a/lisp/textmodes/ispell.el
+++ b/lisp/textmodes/ispell.el
@@ -3250,7 +3250,11 @@ ispell-region
 	    (if (and ispell-checking-message (numberp ispell-quit))
 		(progn
 		  (setq ispell-quit nil)
-		  (error "Message send aborted")))
+                  (error "%s"
+                         (concat "Message sending aborted (use "
+                                 (substitute-command-keys
+		                  "\\[universal-argument] \\[ispell-word]")
+                                 " to resume spell-checking)"))))
 	    (if (not recheckp) (setq ispell-quit nil)))
 	(if (not recheckp) (set-marker ispell-region-end nil))
 	;; Only save if successful exit.

Message #11 received at 80036 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Sébastien Hinderer <sebastien.hinderer <at> ens-lyon.org>
Cc: 80036 <at> debbugs.gnu.org
Subject: Re: bug#80036: ispell-region cannot be suspended
Date: Sat, 20 Dec 2025 11:46:07 +0200

> Date: Sat, 20 Dec 2025 10:15:08 +0100
> From: Sébastien Hinderer <sebastien.hinderer <at> ens-lyon.org>
> Cc: 80036 <at> debbugs.gnu.org
> 
> Ah I see, sorry I didn't think about thecking the messages window.
> 
> However, as far as I understand, ispell-mesage *is* an interactive
> function. It is thus not wrong to call it directly, independently of
> whether one is in the process of sending a message or not. And
> this is actually my use-case for it since I use Emacs as an
> external editor, my MUA being mutt.
> 
> Would it perhaps make sense that the exposed ispell-message does not
> care about sending at all, ormaybe has an argument to let it know
> whether one is in the context of sending a message or not?

To distinguish between the case of this function running in a
message-send-hook and the case it was invoked would need deeper
changes.  Patches for doing that will be most welcome.

Message #14 received at 80036 <at> debbugs.gnu.org (full text, mbox):

From: Sébastien Hinderer <sebastien.hinderer <at> ens-lyon.org>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 80036 <at> debbugs.gnu.org
Subject: Re: bug#80036: ispell-region cannot be suspended
Date: Sat, 20 Dec 2025 10:15:08 +0100

Ah I see, sorry I didn't think about thecking the messages window.

However, as far as I understand, ispell-mesage *is* an interactive
function. It is thus not wrong to call it directly, independently of
whether one is in the process of sending a message or not. And
this is actually my use-case for it since I use Emacs as an
external editor, my MUA being mutt.

Would it perhaps make sense that the exposed ispell-message does not
care about sending at all, ormaybe has an argument to let it know
whether one is in the context of sending a message or not?

Message #17 received at 80036 <at> debbugs.gnu.org (full text, mbox):

From: Sébastien Hinderer <sebastien.hinderer <at> ens-lyon.org>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 80036 <at> debbugs.gnu.org
Subject: Re: bug#80036: ispell-region cannot be suspended
Date: Sat, 20 Dec 2025 10:51:28 +0100

Eli Zaretskii (2025/12/20 11:46 +0200):
> To distinguish between the case of this function running in a
> message-send-hook and the case it was invoked would need deeper
> changes.

Yes, I agree that to be done cleanly this will require a bit of
rewriting / refactoring.

My Emacs-Lisp-fu is not big enough for me to address such a challenge
right now but I will definitely keep that in mind as one thing to work
on when I become expert enough, if it has not been addressed before.

Previous Next

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