bug#32969: "Confirm--do you mean to use ‘*’ as a wildcard? (y or n)" stays in minibuffer

bug#32969: "Confirm--do you mean to use ‘*’ as a wildcard? (y or n)" stays in minibuffer

[積丹尼 Dan Jacobson]

Dired has a message:
Confirm--do you mean to use ‘*’ as a wildcard? (y or n)
if one answers "n" the message just stays in the minibuffer.

It should instead say "command aborted", else the user thinks it used
'*' in its other (text replacement) way. Which it turns out it doesn't.

emacs-version "25.2.2"

bug#32969: "Confirm--do you mean to use ‘*’ as a wildcard? (y or n)" stays in minibuffer

[Tino Calancha]

積丹尼 Dan Jacobson <jidanni@jidanni.org> writes:

> Dired has a message:
> Confirm--do you mean to use ‘*’ as a wildcard? (y or n)
> if one answers "n" the message just stays in the minibuffer.
>
> It should instead say "command aborted", else the user thinks it used
> '*' in its other (text replacement) way. Which it turns out it doesn't.
Thank you for the report!
I agree.

--8<-----------------------------cut here---------------start------------->8---
commit e76bbfa62b7185cd3284d69483704366dc54e7e8
Author: Tino Calancha <tino.calancha@gmail.com>
Date:   Mon Oct 8 02:40:30 2018 +0900

    dired-do-shell-command: Notify users after abort the command
    
    * lisp/dired-aux.el (dired-do-shell-command):  Notify users that
    the command have aborted when they answer 'n' to the prompt (Bug#32969).

diff --git a/lisp/dired-aux.el b/lisp/dired-aux.el
index 1f13204b7c..a4145eb67a 100644
--- a/lisp/dired-aux.el
+++ b/lisp/dired-aux.el
@@ -763,16 +763,17 @@ dired-do-shell-command
 	            (y-or-n-p (format-message
 			       "Confirm--do you mean to use `?' as a wildcard? ")))
 	           (t))))
-    (when ok
-      (if on-each
-	  (dired-bunch-files (- 10000 (length command))
-	                     (lambda (&rest files)
-	                       (dired-run-shell-command
-                                (dired-shell-stuff-it command files t arg)))
-	                     nil file-list)
-	;; execute the shell command
-	(dired-run-shell-command
-	 (dired-shell-stuff-it command file-list nil arg)))))))
+    (cond ((not ok) (message "OK, canceled"))
+          (t
+           (if on-each
+	       (dired-bunch-files (- 10000 (length command))
+	                          (lambda (&rest files)
+	                            (dired-run-shell-command
+                                     (dired-shell-stuff-it command files t arg)))
+	                          nil file-list)
+	     ;; execute the shell command
+	     (dired-run-shell-command
+	      (dired-shell-stuff-it command file-list nil arg))))))))
 
 ;; Might use {,} for bash or csh:
 (defvar dired-mark-prefix ""

--8<-----------------------------cut here---------------end--------------->8---

bug#32969: "Confirm--do you mean to use ‘*’ as a wildcard? (y or n)" stays in minibuffer

[Eli Zaretskii]

> From: Tino Calancha <tino.calancha@gmail.com>
> Date: Mon, 08 Oct 2018 02:44:12 +0900
> Cc: 32969@debbugs.gnu.org
> 
> +    (cond ((not ok) (message "OK, canceled"))

I'd suggest to use "Command canceled".  the "OK" part might confuse
some non-native English speakers to think the command was executed.

Thanks.

bug#32969: "Confirm--do you mean to use ‘*’ as a wildcard? (y or n)" stays in minibuffer

[Tino Calancha]

On Sun, 7 Oct 2018, Eli Zaretskii wrote:

>> From: Tino Calancha <tino.calancha@gmail.com>
>> Date: Mon, 08 Oct 2018 02:44:12 +0900
>> Cc: 32969@debbugs.gnu.org
>>
>> +    (cond ((not ok) (message "OK, canceled"))
>
> I'd suggest to use "Command canceled".  the "OK" part might confuse
> some non-native English speakers to think the command was executed.
OK (I mean, I updated the patch as you mentioned :-)

BTW, emacs-26 or master?

bug#32969: "Confirm--do you mean to use ‘*’ as a wildcard? (y or n)" stays in minibuffer

[Eli Zaretskii]

> From: Tino Calancha <tino.calancha@gmail.com>
> Date: Mon, 8 Oct 2018 04:43:18 +0900 (JST)
> cc: Tino Calancha <tino.calancha@gmail.com>, jidanni@jidanni.org, 
>     32969@debbugs.gnu.org
> 
> >> +    (cond ((not ok) (message "OK, canceled"))
> >
> > I'd suggest to use "Command canceled".  the "OK" part might confuse
> > some non-native English speakers to think the command was executed.
> OK (I mean, I updated the patch as you mentioned :-)
> 
> BTW, emacs-26 or master?

I think emacs-26.  Thanks.

bug#32969: "Confirm--do you mean to use ‘*’ as a wildcard? (y or n)" stays in minibuffer

[Tino Calancha]

On Mon, 8 Oct 2018, Eli Zaretskii wrote:

>> From: Tino Calancha <tino.calancha@gmail.com>
>> Date: Mon, 8 Oct 2018 04:43:18 +0900 (JST)
>> cc: Tino Calancha <tino.calancha@gmail.com>, jidanni@jidanni.org,
>>     32969@debbugs.gnu.org
>>
>>>> +    (cond ((not ok) (message "OK, canceled"))
>>>
>>> I'd suggest to use "Command canceled".  the "OK" part might confuse
>>> some non-native English speakers to think the command was executed.
>> OK (I mean, I updated the patch as you mentioned :-)
>>
>> BTW, emacs-26 or master?
> I think emacs-26.  Thanks.
OK (I mean, 0 killed, understood, got it, entendido, wakarimashita)
I will wait some days before push it.  Thanks.

bug#32969: "Confirm--do you mean to use ‘*’ as a wildcard? (y or n)" stays in minibuffer

[Tino Calancha]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Tino Calancha <tino.calancha@gmail.com>
>> Date: Mon, 8 Oct 2018 04:43:18 +0900 (JST)
>> cc: Tino Calancha <tino.calancha@gmail.com>, jidanni@jidanni.org, 
>>     32969@debbugs.gnu.org
>> 
>> >> +    (cond ((not ok) (message "OK, canceled"))
>> >
>> > I'd suggest to use "Command canceled".  the "OK" part might confuse
>> > some non-native English speakers to think the command was executed.
>> OK (I mean, I updated the patch as you mentioned :-)
>> 
>> BTW, emacs-26 or master?
>
> I think emacs-26.  Thanks.
Implemeted in emacs-26 as commit
'dired-do-shell-command: Notify users after abort the command'
(a7ebc6bf633bd3849ccab032dad6b1fd31b1ef43)