GNU bug report logs - #79827
Add option to display Quail guidance on one line

Previous Next

To reply to this bug, email your comments to 79827 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: Philip Kaludercic <philipk <at> posteo.net>
To: bug-gnu-emacs <at> gnu.org
Subject: Add option to display Quail guidance on one line
Date: Thu, 13 Nov 2025 19:29:53 +0000

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

The background to this change is that certain input methods, such as
Agda use a guidance string that resizes the minibuffer.  I find this
visually annoying, since resizing the minibuffer can cause the point to
move and the viewport of a buffer to change, so I'd like to have the
option to disable the behaviour.

Of course, we could also consider to just hard-code the change without
an user option, if we don't see any reason to keep the newline, but that
would be a more controversial decision.

I have separately proposed a patch to agda2-mode that would fix the
issue on older versions of Emacs, but the approach is finicky and it
would be better to upstream it: https://github.com/agda/agda/pull/8177.

[0001-Add-option-to-display-Quail-guidance-on-one-line.patch (text/x-patch, attachment)]

Message #8 received at 79827 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Philip Kaludercic <philipk <at> posteo.net>
Cc: 79827 <at> debbugs.gnu.org
Subject: Re: bug#79827: Add option to display Quail guidance on one line
Date: Thu, 13 Nov 2025 22:10:49 +0200

> From: Philip Kaludercic <philipk <at> posteo.net>
> Date: Thu, 13 Nov 2025 19:29:53 +0000
> 
> The background to this change is that certain input methods, such as
> Agda use a guidance string that resizes the minibuffer.  I find this
> visually annoying, since resizing the minibuffer can cause the point to
> move and the viewport of a buffer to change, so I'd like to have the
> option to disable the behaviour.

This will only help if the guidance string is short enough, so is it
really a good idea to provide a user option that works only sometimes?

Btw, did you customize the value of resize-mini-windows, or is the
behavior you describe frequent with the default value of 'grow-only'?

Message #11 received at 79827 <at> debbugs.gnu.org (full text, mbox):

From: Philip Kaludercic <philipk <at> posteo.net>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 79827 <at> debbugs.gnu.org
Subject: Re: bug#79827: Add option to display Quail guidance on one line
Date: Fri, 14 Nov 2025 22:31:50 +0000

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Philip Kaludercic <philipk <at> posteo.net>
>> Date: Thu, 13 Nov 2025 19:29:53 +0000
>> 
>> The background to this change is that certain input methods, such as
>> Agda use a guidance string that resizes the minibuffer.  I find this
>> visually annoying, since resizing the minibuffer can cause the point to
>> move and the viewport of a buffer to change, so I'd like to have the
>> option to disable the behaviour.
>
> This will only help if the guidance string is short enough, so is it
> really a good idea to provide a user option that works only sometimes?

You mean that it would check if the entire input and the guidance string
fits in (window-width)?  What do we do when the line exceeds the width
due to user input -- or did I misunderstand what you are insinuating? 

> Btw, did you customize the value of resize-mini-windows, or is the
> behavior you describe frequent with the default value of 'grow-only'?

I did in fact change `resize-mini-windows' as you expected.  But that
shouldn't change the fact that having a newline increases the size of
the minibuffer, right?

Message #14 received at 79827 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Philip Kaludercic <philipk <at> posteo.net>
Cc: 79827 <at> debbugs.gnu.org
Subject: Re: bug#79827: Add option to display Quail guidance on one line
Date: Sat, 15 Nov 2025 09:39:52 +0200

> From: Philip Kaludercic <philipk <at> posteo.net>
> Cc: 79827 <at> debbugs.gnu.org
> Date: Fri, 14 Nov 2025 22:31:50 +0000
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> >> From: Philip Kaludercic <philipk <at> posteo.net>
> >> Date: Thu, 13 Nov 2025 19:29:53 +0000
> >> 
> >> The background to this change is that certain input methods, such as
> >> Agda use a guidance string that resizes the minibuffer.  I find this
> >> visually annoying, since resizing the minibuffer can cause the point to
> >> move and the viewport of a buffer to change, so I'd like to have the
> >> option to disable the behaviour.
> >
> > This will only help if the guidance string is short enough, so is it
> > really a good idea to provide a user option that works only sometimes?
> 
> You mean that it would check if the entire input and the guidance string
> fits in (window-width)?  What do we do when the line exceeds the width
> due to user input -- or did I misunderstand what you are insinuating? 

What I mean (but do not "insinuate") is that if the total length of
the prompt and the guidance is greater than window-width, the
mini-window will be resized anyway.  If the user customizes this
option in the hope that it will prevent resizing, they will be
disappointed at best when the prompt+guidance force the resize anyway.

So this option sounds like a half-measure, perhaps only good enough in
the specific case of the input method you are using, but not in
others.

> > Btw, did you customize the value of resize-mini-windows, or is the
> > behavior you describe frequent with the default value of 'grow-only'?
> 
> I did in fact change `resize-mini-windows' as you expected.  But that
> shouldn't change the fact that having a newline increases the size of
> the minibuffer, right?

If you customize it to nil, it will never resize, no (at the price of
showing only part of the min-window's contents).  Which value did you
customize it to?

Message #17 received at 79827 <at> debbugs.gnu.org (full text, mbox):

From: Philip Kaludercic <philipk <at> posteo.net>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 79827 <at> debbugs.gnu.org
Subject: Re: bug#79827: Add option to display Quail guidance on one line
Date: Sat, 15 Nov 2025 09:51:50 +0000

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Philip Kaludercic <philipk <at> posteo.net>
>> Cc: 79827 <at> debbugs.gnu.org
>> Date: Fri, 14 Nov 2025 22:31:50 +0000
>> 
>> Eli Zaretskii <eliz <at> gnu.org> writes:
>> 
>> >> From: Philip Kaludercic <philipk <at> posteo.net>
>> >> Date: Thu, 13 Nov 2025 19:29:53 +0000
>> >> 
>> >> The background to this change is that certain input methods, such as
>> >> Agda use a guidance string that resizes the minibuffer.  I find this
>> >> visually annoying, since resizing the minibuffer can cause the point to
>> >> move and the viewport of a buffer to change, so I'd like to have the
>> >> option to disable the behaviour.
>> >
>> > This will only help if the guidance string is short enough, so is it
>> > really a good idea to provide a user option that works only sometimes?
>> 
>> You mean that it would check if the entire input and the guidance string
>> fits in (window-width)?  What do we do when the line exceeds the width
>> due to user input -- or did I misunderstand what you are insinuating? 
>
> What I mean (but do not "insinuate") is that if the total length of
> the prompt and the guidance is greater than window-width, the
> mini-window will be resized anyway.  If the user customizes this
> option in the hope that it will prevent resizing, they will be
> disappointed at best when the prompt+guidance force the resize anyway.

OK, thanks for the explanation.  And yes, that is an issue we could
improve on by clarifying that the user option reduces the cases where
resizing the mini-window has to occur.  I could also add a third value
to the user option that truncates the guidance string?

> So this option sounds like a half-measure, perhaps only good enough in
> the specific case of the input method you are using, but not in
> others.
>
>> > Btw, did you customize the value of resize-mini-windows, or is the
>> > behavior you describe frequent with the default value of 'grow-only'?
>> 
>> I did in fact change `resize-mini-windows' as you expected.  But that
>> shouldn't change the fact that having a newline increases the size of
>> the minibuffer, right?
>
> If you customize it to nil, it will never resize, no (at the price of
> showing only part of the min-window's contents).  Which value did you
> customize it to?

I set it to 'grow-only'.

Message #20 received at 79827 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Philip Kaludercic <philipk <at> posteo.net>
Cc: 79827 <at> debbugs.gnu.org
Subject: Re: bug#79827: Add option to display Quail guidance on one line
Date: Sat, 15 Nov 2025 13:35:55 +0200

> From: Philip Kaludercic <philipk <at> posteo.net>
> Cc: 79827 <at> debbugs.gnu.org
> Date: Sat, 15 Nov 2025 09:51:50 +0000
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> >> You mean that it would check if the entire input and the guidance string
> >> fits in (window-width)?  What do we do when the line exceeds the width
> >> due to user input -- or did I misunderstand what you are insinuating? 
> >
> > What I mean (but do not "insinuate") is that if the total length of
> > the prompt and the guidance is greater than window-width, the
> > mini-window will be resized anyway.  If the user customizes this
> > option in the hope that it will prevent resizing, they will be
> > disappointed at best when the prompt+guidance force the resize anyway.
> 
> OK, thanks for the explanation.  And yes, that is an issue we could
> improve on by clarifying that the user option reduces the cases where
> resizing the mini-window has to occur.  I could also add a third value
> to the user option that truncates the guidance string?

Or maybe advise to customize resize-mini-windows?

> >> I did in fact change `resize-mini-windows' as you expected.  But that
> >> shouldn't change the fact that having a newline increases the size of
> >> the minibuffer, right?
> >
> > If you customize it to nil, it will never resize, no (at the price of
> > showing only part of the min-window's contents).  Which value did you
> > customize it to?
> 
> I set it to 'grow-only'.

That's the default value.

Message #23 received at 79827 <at> debbugs.gnu.org (full text, mbox):

From: Philip Kaludercic <philipk <at> posteo.net>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: 79827 <at> debbugs.gnu.org
Subject: Re: bug#79827: Add option to display Quail guidance on one line
Date: Sat, 15 Nov 2025 20:56:59 +0000

Eli Zaretskii <eliz <at> gnu.org> writes:

>> From: Philip Kaludercic <philipk <at> posteo.net>
>> Cc: 79827 <at> debbugs.gnu.org
>> Date: Sat, 15 Nov 2025 09:51:50 +0000
>> 
>> Eli Zaretskii <eliz <at> gnu.org> writes:
>> 
>> >> You mean that it would check if the entire input and the guidance string
>> >> fits in (window-width)?  What do we do when the line exceeds the width
>> >> due to user input -- or did I misunderstand what you are insinuating? 
>> >
>> > What I mean (but do not "insinuate") is that if the total length of
>> > the prompt and the guidance is greater than window-width, the
>> > mini-window will be resized anyway.  If the user customizes this
>> > option in the hope that it will prevent resizing, they will be
>> > disappointed at best when the prompt+guidance force the resize anyway.
>> 
>> OK, thanks for the explanation.  And yes, that is an issue we could
>> improve on by clarifying that the user option reduces the cases where
>> resizing the mini-window has to occur.  I could also add a third value
>> to the user option that truncates the guidance string?
>
> Or maybe advise to customize resize-mini-windows?

I am afraid I don't understand what you mean... Advise whom, where?

>> >> I did in fact change `resize-mini-windows' as you expected.  But that
>> >> shouldn't change the fact that having a newline increases the size of
>> >> the minibuffer, right?
>> >
>> > If you customize it to nil, it will never resize, no (at the price of
>> > showing only part of the min-window's contents).  Which value did you
>> > customize it to?
>> 
>> I set it to 'grow-only'.
>
> That's the default value.

My bad, I misread the *Help* buffer.  It actually says:

  Its value is t
  Original value was ‘grow-only’

Message #26 received at 79827 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Philip Kaludercic <philipk <at> posteo.net>
Cc: 79827 <at> debbugs.gnu.org
Subject: Re: bug#79827: Add option to display Quail guidance on one line
Date: Sun, 16 Nov 2025 08:04:45 +0200

> From: Philip Kaludercic <philipk <at> posteo.net>
> Cc: 79827 <at> debbugs.gnu.org
> Date: Sat, 15 Nov 2025 20:56:59 +0000
> 
> Eli Zaretskii <eliz <at> gnu.org> writes:
> 
> >> OK, thanks for the explanation.  And yes, that is an issue we could
> >> improve on by clarifying that the user option reduces the cases where
> >> resizing the mini-window has to occur.  I could also add a third value
> >> to the user option that truncates the guidance string?
> >
> > Or maybe advise to customize resize-mini-windows?
> 
> I am afraid I don't understand what you mean... Advise whom, where?

Advise our users in the relevant documentation.

> >> >> I did in fact change `resize-mini-windows' as you expected.  But that
> >> >> shouldn't change the fact that having a newline increases the size of
> >> >> the minibuffer, right?
> >> >
> >> > If you customize it to nil, it will never resize, no (at the price of
> >> > showing only part of the min-window's contents).  Which value did you
> >> > customize it to?
> >> 
> >> I set it to 'grow-only'.
> >
> > That's the default value.
> 
> My bad, I misread the *Help* buffer.  It actually says:
> 
>   Its value is t
>   Original value was ‘grow-only’

Ah, okay.  The value t will exacerbate the resizing annoyance.  So
could you perhaps leave this at its default value and see if the
current code still annoys you so much?  Then try the value of nil as
well.

Thanks.

Previous Next

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