bug#39558: 26.3; icomplete and display *Completions* in child frame

bug#39558: 26.3; icomplete and display *Completions* in child frame

[Andrii Kolomoiets]

1. emacs -Q
2. evaluate
(customize-set-variable
 'display-buffer-alist
 '(("\\*Completions\\*" display-buffer-in-child-frame)))
3. M-x icomplete-mode
4. M-x f<TAB>

Error in post-command-hook (icomplete-post-command-hook): (wrong-type-argument number-or-marker-p nil)

In GNU Emacs 26.3 (build 1, x86_64-apple-darwin18.2.0, NS appkit-1671.20 Version 10.14.3 (Build 18D109))
 of 2019-09-02 built on builder10-14.porkrind.org
Windowing system distributor 'Apple', version 10.3.1894

Recent messages:
(("\\*Completions\\*" display-buffer-in-child-frame))
Icomplete mode enabled
Making completion list...
In #<buffer  *Minibuf-1*> (w=#<window 5 on *Completions*>): nil
Error in post-command-hook (icomplete-post-command-hook): (wrong-type-argument number-or-marker-p nil)

bug#39558: 26.3; icomplete and display *Completions* in child frame

[Eli Zaretskii]

> From: Andrii Kolomoiets <andreyk.mad@gmail.com>
> Date: Tue, 11 Feb 2020 08:55:48 +0200
> 
> 1. emacs -Q
> 2. evaluate
> (customize-set-variable
>  'display-buffer-alist
>  '(("\\*Completions\\*" display-buffer-in-child-frame)))
> 3. M-x icomplete-mode
> 4. M-x f<TAB>
> 
> Error in post-command-hook (icomplete-post-command-hook): (wrong-type-argument number-or-marker-p nil)

I cannot reproduce this, so maybe this is Darwin-specific.

Thanks.

bug#39558: 26.3; icomplete and display *Completions* in child frame

[Drew Adams]

> > 1. emacs -Q
> > 2. evaluate
> > (customize-set-variable
> >  'display-buffer-alist
> >  '(("\\*Completions\\*" display-buffer-in-child-frame)))
> > 3. M-x icomplete-mode
> > 4. M-x f<TAB>
> >
> > Error in post-command-hook (icomplete-post-command-hook): (wrong-type-
> argument number-or-marker-p nil)
> 
> I cannot reproduce this, so maybe this is Darwin-specific.

FWIW:

This has been reported multiple times.  I
reported it years ago.  It happened now &
then (I don't use icomplete anymore, so I
haven't tracked it).  I tried to track it
down but never succeeded.  Discussed this
with Stefan a few times.

I use MS Windows, so if what I saw is the
same as what's been reported by others a
few times since then, I doubt that it is
Darwin-specific.

`icomplete-exhibit' was where this happened.
As I say, this is longstanding, and no doubt
Icomplete has undergone multiple changes since
I first reported this.  The fact that the same,
or seemingly the same, thing is still happening
suggests that whatever changes have been made
to Icomplete haven't affected this.

bug#39558: 26.3; icomplete and display *Completions* in child frame

[Andrii Kolomoiets]

Eli Zaretskii <eliz@gnu.org> writes:

>> 1. emacs -Q
>> 2. evaluate
>> (customize-set-variable
>>  'display-buffer-alist
>>  '(("\\*Completions\\*" display-buffer-in-child-frame)))
>> 3. M-x icomplete-mode
>> 4. M-x f<TAB>
>> 
>> Error in post-command-hook (icomplete-post-command-hook): (wrong-type-argument number-or-marker-p nil)
>
> I cannot reproduce this, so maybe this is Darwin-specific.

You are right.

There are this code in lisp/term/ns-win.el:581:
(add-hook 'after-make-frame-functions 'select-frame)

Evaluating it allow you to catch that error in
icomplete-post-command-hook under X as well.

To avoid original issue under Darwin:

(setq after-make-frame-functions nil)
(customize-set-variable
 'display-buffer-alist
 '(("\\*Completions\\*" display-buffer-in-child-frame
    (child-frame-parameters . ((minibuffer . nil))))))

(The `(minibuffer . nil)' frame parameter is required if we want to leave
focus on original frame. IDK why `(no-focus-on-map . t)' doesn't help
here.)

bug#39558: 26.3; icomplete and display *Completions* in child frame

[Lars Ingebrigtsen]

Andrii Kolomoiets <andreyk.mad@gmail.com> writes:

>> I cannot reproduce this, so maybe this is Darwin-specific.
>
> You are right.
>
> There are this code in lisp/term/ns-win.el:581:
> (add-hook 'after-make-frame-functions 'select-frame)
>
> Evaluating it allow you to catch that error in
> icomplete-post-command-hook under X as well.
>
> To avoid original issue under Darwin:
>
> (setq after-make-frame-functions nil)
> (customize-set-variable
>  'display-buffer-alist
>  '(("\\*Completions\\*" display-buffer-in-child-frame
>     (child-frame-parameters . ((minibuffer . nil))))))
>
> (The `(minibuffer . nil)' frame parameter is required if we want to leave
> focus on original frame. IDK why `(no-focus-on-map . t)' doesn't help
> here.)

Andrii, are you still seeing this issue in more recent versions of
Emacs?  If so, perhaps Alan has some comments here (added to the CCs).

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

bug#39558: 26.3; icomplete and display *Completions* in child frame

[Alan Third]

On Wed, Aug 11, 2021 at 02:34:37PM +0200, Lars Ingebrigtsen wrote:
> Andrii Kolomoiets <andreyk.mad@gmail.com> writes:
> 
> >> I cannot reproduce this, so maybe this is Darwin-specific.
> >
> > You are right.
> >
> > There are this code in lisp/term/ns-win.el:581:
> > (add-hook 'after-make-frame-functions 'select-frame)
> >
> > Evaluating it allow you to catch that error in
> > icomplete-post-command-hook under X as well.
> >
> > To avoid original issue under Darwin:
> >
> > (setq after-make-frame-functions nil)
> > (customize-set-variable
> >  'display-buffer-alist
> >  '(("\\*Completions\\*" display-buffer-in-child-frame
> >     (child-frame-parameters . ((minibuffer . nil))))))
> >
> > (The `(minibuffer . nil)' frame parameter is required if we want to leave
> > focus on original frame. IDK why `(no-focus-on-map . t)' doesn't help
> > here.)
> 
> Andrii, are you still seeing this issue in more recent versions of
> Emacs?  If so, perhaps Alan has some comments here (added to the CCs).

I don't know if it's any help but the after-make-frame-functions
setting is related to bug#3397.

I've had a look and haven't been able to work out what we need to do
to fix that one.

-- 
Alan Third

bug#39558: 26.3; icomplete and display *Completions* in child frame

[Andrii Kolomoiets]

Lars Ingebrigtsen <larsi@gnus.org> writes:

> Andrii Kolomoiets <andreyk.mad@gmail.com> writes:
>
>>> I cannot reproduce this, so maybe this is Darwin-specific.
>>
>> You are right.
>>
>> There are this code in lisp/term/ns-win.el:581:
>> (add-hook 'after-make-frame-functions 'select-frame)
>>
>> Evaluating it allow you to catch that error in
>> icomplete-post-command-hook under X as well.
>
> Andrii, are you still seeing this issue in more recent versions of
> Emacs?  If so, perhaps Alan has some comments here (added to the CCs).

Yes.  I can reproduce this in recent Emacs version.

-- 
Andrii