bug#73820: 31.0.50; [Feature Request]how to hide *Completions* buffer when icomplete-in-buffer is not nil

bug#73820: 31.0.50; [Feature Request]how to hide *Completions* buffer when icomplete-in-buffer is not nil

[jixiuf@qq.com via Bug reports for GNU Emacs, the Swiss army knife of text editors]

[-- Attachment #1: Type: text/plain, Size: 245 bytes --]

       
I would like to use `icomplete-mode`/`icomplete-vertical-mode` with

 (setq icomplete-in-buffer t)
 
 but I don't want to display the `*Completions*` buffer.
 
 How about adding a new value for `completion-auto-help` to achieve this?
 


[-- Attachment #2: icomplete-in-buffer --]
[-- Type: image/png, Size: 802211 bytes --]

bug#73820: 31.0.50; [Feature Request]how to hide *Completions* buffer when icomplete-in-buffer is not nil

[Juri Linkov]

> I would like to use `icomplete-mode`/`icomplete-vertical-mode` with
>
>  (setq icomplete-in-buffer t)
>
>  but I don't want to display the `*Completions*` buffer.
>
>  How about adding a new value for `completion-auto-help` to achieve this?

As suggested in https://debbugs.gnu.org/67661#114
this is currently achievable by

  (advice-add 'completion-at-point :after #'minibuffer-hide-completions)

But adding a new option is also an option.

bug#73820: 31.0.50; [Feature Request]how to hide *Completions* buffer when icomplete-in-buffer is not nil

[Sean Whitton]

Hello,

On Tue 15 Oct 2024 at 07:43pm +03, Juri Linkov wrote:

>> I would like to use `icomplete-mode`/`icomplete-vertical-mode` with
>>
>>  (setq icomplete-in-buffer t)
>>
>>  but I don't want to display the `*Completions*` buffer.
>>
>>  How about adding a new value for `completion-auto-help` to achieve this?
>
> As suggested in https://debbugs.gnu.org/67661#114
> this is currently achievable by
>
>   (advice-add 'completion-at-point :after #'minibuffer-hide-completions)
>
> But adding a new option is also an option.

Right.  This is also stated in NEWS.

-- 
Sean Whitton

bug#73820: 31.0.50; [Feature Request]how to hide *Completions* buffer when icomplete-in-buffer is not nil

[Eli Zaretskii]

> Cc: 73820@debbugs.gnu.org, "jixiuf@qq.com" <jixiuf@qq.com>
> From: Sean Whitton <spwhitton@spwhitton.name>
> Date: Wed, 16 Oct 2024 09:23:52 +0800
> 
> On Tue 15 Oct 2024 at 07:43pm +03, Juri Linkov wrote:
> 
> >> I would like to use `icomplete-mode`/`icomplete-vertical-mode` with
> >>
> >>  (setq icomplete-in-buffer t)
> >>
> >>  but I don't want to display the `*Completions*` buffer.
> >>
> >>  How about adding a new value for `completion-auto-help` to achieve this?
> >
> > As suggested in https://debbugs.gnu.org/67661#114
> > this is currently achievable by
> >
> >   (advice-add 'completion-at-point :after #'minibuffer-hide-completions)
> >
> > But adding a new option is also an option.
> 
> Right.  This is also stated in NEWS.

If this is important enough, it should be in the doc string.  So I've
now added this tip to the doc string, and I'm therefore closing this
bug.

Thanks.

bug#73820: 31.0.50; [Feature Request]how to hide *Completions* buffer when icomplete-in-buffer is not nil

[Eshel Yaron via Bug reports for GNU Emacs, the Swiss army knife of text editors]

Hi,

Eli Zaretskii <eliz@gnu.org> writes:

>> Cc: 73820@debbugs.gnu.org, "jixiuf@qq.com" <jixiuf@qq.com>
>> From: Sean Whitton <spwhitton@spwhitton.name>
>> Date: Wed, 16 Oct 2024 09:23:52 +0800
>> 
>> On Tue 15 Oct 2024 at 07:43pm +03, Juri Linkov wrote:
>> 
>> >> I would like to use `icomplete-mode`/`icomplete-vertical-mode` with
>> >>
>> >>  (setq icomplete-in-buffer t)
>> >>
>> >>  but I don't want to display the `*Completions*` buffer.
>> >>
>> >>  How about adding a new value for `completion-auto-help` to achieve this?
>> >
>> > As suggested in https://debbugs.gnu.org/67661#114
>> > this is currently achievable by
>> >
>> >   (advice-add 'completion-at-point :after #'minibuffer-hide-completions)
>> >
>> > But adding a new option is also an option.
>> 
>> Right.  This is also stated in NEWS.
>
> If this is important enough, it should be in the doc string.  So I've
> now added this tip to the doc string, and I'm therefore closing this
> bug.

Hmm, I'm not sure that that's really such a good recommendation.  Adding
this advice breaks M-: completion, for example.  I think a more refined
solution is due, if this is to be considered a supported use case...
Namely, I suggest implementing icomplete-in-buffer as a
completion-in-region-function, like some other in-buffer completion
frontends are implemented.


Best,

Eshel

bug#73820: 31.0.50; [Feature Request]how to hide *Completions* buffer when icomplete-in-buffer is not nil

[Eli Zaretskii]

> From: Eshel Yaron <me@eshelyaron.com>
> Cc: eliz@gnu.org,  jixiuf@qq.com, Sean Whitton <spwhitton@spwhitton.name>,
>  Juri Linkov <juri@linkov.net>
> Date: Wed, 16 Oct 2024 09:37:19 +0200
> 
> >> Right.  This is also stated in NEWS.
> >
> > If this is important enough, it should be in the doc string.  So I've
> > now added this tip to the doc string, and I'm therefore closing this
> > bug.
> 
> Hmm, I'm not sure that that's really such a good recommendation.  Adding
> this advice breaks M-: completion, for example.  I think a more refined
> solution is due, if this is to be considered a supported use case...
> Namely, I suggest implementing icomplete-in-buffer as a
> completion-in-region-function, like some other in-buffer completion
> frontends are implemented.

Of course, a better solution, especially if it is not based on
advising functions, are always welcome.  But I'm guessing the solution
you have in mind is not for the release branch, though?

bug#73820: 31.0.50; [Feature Request]how to hide *Completions* buffer when icomplete-in-buffer is not nil

[Eshel Yaron via Bug reports for GNU Emacs, the Swiss army knife of text editors]

Eli Zaretskii <eliz@gnu.org> writes:

>> From: Eshel Yaron <me@eshelyaron.com>
>> Cc: eliz@gnu.org,  jixiuf@qq.com, Sean Whitton <spwhitton@spwhitton.name>,
>>  Juri Linkov <juri@linkov.net>
>> Date: Wed, 16 Oct 2024 09:37:19 +0200
>> 
>> >> Right.  This is also stated in NEWS.
>> >
>> > If this is important enough, it should be in the doc string.  So I've
>> > now added this tip to the doc string, and I'm therefore closing this
>> > bug.
>> 
>> Hmm, I'm not sure that that's really such a good recommendation.  Adding
>> this advice breaks M-: completion, for example.  I think a more refined
>> solution is due, if this is to be considered a supported use case...
>> Namely, I suggest implementing icomplete-in-buffer as a
>> completion-in-region-function, like some other in-buffer completion
>> frontends are implemented.
>
> Of course, a better solution, especially if it is not based on
> advising functions, are always welcome.  But I'm guessing the solution
> you have in mind is not for the release branch, though?

Right, it'd probably be too big of a change to make it into Emacs 30.

bug#73820: 31.0.50; [Feature Request]how to hide *Completions* buffer when icomplete-in-buffer is not nil

[Sean Whitton]

Hello,

On Wed 16 Oct 2024 at 09:37am +02, Eshel Yaron wrote:

> Hmm, I'm not sure that that's really such a good recommendation.  Adding
> this advice breaks M-: completion, for example.  I think a more refined
> solution is due, if this is to be considered a supported use case...
> Namely, I suggest implementing icomplete-in-buffer as a
> completion-in-region-function, like some other in-buffer completion
> frontends are implemented.

If you were to look into that change, I would be very happy to review.

-- 
Sean Whitton