bug#64141: 27.1 newly has clunky type-break y-or-n-p

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Samuel Wales]

i upgraded to 27.1 and found that type-break is asking me about breaks
as follows:

when it asks in minibuffer

- point is in another window, requiring focus, changing point, and
tending to type-break
- home or page up key puts point at beg of message, where it will not
accept a reply
- it is generally more high-maintenane to get it to do a type-break or
to get it to postpone it
- it seems to arrive in the wrong place or at the wrong instant more
[this is subjective and please ignore if nonsensical]

apologies for bad bug report.  perhaps it will trigger a memory of a
code change or changelog/diff will make it obvious.

-- 
The Kafka Pandemic

A blog about science, health, human rights, and misopathy:
https://thekafkapandemic.blogspot.com

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Eli Zaretskii]

> From: Samuel Wales <samologist@gmail.com>
> Date: Sat, 17 Jun 2023 18:17:00 -0700
> 
> i upgraded to 27.1 and found that type-break is asking me about breaks
> as follows:

Which version of Emacs did you have installed before the upgrade?
Was type-break working differently/better in that older version?

> when it asks in minibuffer
> 
> - point is in another window, requiring focus, changing point, and
> tending to type-break
> - home or page up key puts point at beg of message, where it will not
> accept a reply
> - it is generally more high-maintenane to get it to do a type-break or
> to get it to postpone it
> - it seems to arrive in the wrong place or at the wrong instant more
> [this is subjective and please ignore if nonsensical]
> 
> apologies for bad bug report.  perhaps it will trigger a memory of a
> code change or changelog/diff will make it obvious.

I don't see any significant changes in type-break.el since Emacs 26.1.
But maybe I missed something, so please try to provide a more detailed
recipe to reproduce the problems, preferably starting from "emacs -Q",
so that we could understand the problems and fix them if necessary.

Thanks.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Samuel Wales]

hi

On 6/18/23, Eli Zaretskii <eliz@gnu.org> wrote:
>> From: Samuel Wales <samologist@gmail.com>
>> Date: Sat, 17 Jun 2023 18:17:00 -0700
>>
>> i upgraded to 27.1 and found that type-break is asking me about breaks
>> as follows:
>
> Which version of Emacs did you have installed before the upgrade?
> Was type-break working differently/better in that older version?

it was working better [than 27.1] before 26.1 for certain.  but idk if
it was broken in 26.1.

> I don't see any significant changes in type-break.el since Emacs 26.1.
> But maybe I missed something, so please try to provide a more detailed
> recipe to reproduce the problems, preferably starting from "emacs -Q",
> so that we could understand the problems and fix them if necessary.

it occurs frequently wihtout my brain making connections with the
conditions under which it occurs, but if i can i will try paying
attention to the conditions.  i know it occurs when i get an appt
notification and thus more than one window.

for some reason i keep hitting home, which is set to move point to
bob, but if at prompt [by chance?], then that just puts point at its
bol and then y and n will not work, which makes type-break confuse me
and require other-window, eol, etc.

if i can i will try setting the vars to short delays and -Q.  [i am
limited in computer use.]

-- 
The Kafka Pandemic

A blog about science, health, human rights, and misopathy:
https://thekafkapandemic.blogspot.com

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Samuel Wales]

i used 26.1 only briefly, soon upgrading to 27.1 because i had access to both.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Samuel Wales]

just now in normal usage, the y-or-n-p popped up in the middle of an isearch.

point was in the middle of hte isearch still.  it was not trivial  to
answer the y-or-n-p (point not there) and fix state.

On 6/18/23, Samuel Wales <samologist@gmail.com> wrote:
> i used 26.1 only briefly, soon upgrading to 27.1 because i had access to
> both.
>


-- 
The Kafka Pandemic

A blog about science, health, human rights, and misopathy:
https://thekafkapandemic.blogspot.com

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Juri Linkov]

> just now in normal usage, the y-or-n-p popped up in the middle of an isearch.
>
> point was in the middle of hte isearch still.  it was not trivial  to
> answer the y-or-n-p (point not there) and fix state.

Reading type-break.el raised some questions: please confirm does the same
problem occur after using the default value of 'type-break-query-function'
that is 'yes-or-no-p', and not 'y-or-n-p'?

It looks like type-break uses 'run-at-time' and then tries to
call 'y-or-n-p' in the middle of an isearch.  I tried both

  (run-with-timer 5 5 #'yes-or-no-p "OK? ")
  (run-with-timer 5 5 #'y-or-n-p "OK? ")

and activated isearch.  Then the typed answer was added
to the search string.  Emacs 28 has an option
'y-or-n-p-use-read-key' that you can set for type-break
that pops up the query at random times.  But the problem
still exists for 'yes-or-no-p'.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Eli Zaretskii]

> From: Juri Linkov <juri@linkov.net>
> Cc: Eli Zaretskii <eliz@gnu.org>,  64141@debbugs.gnu.org
> Date: Tue, 20 Jun 2023 09:57:51 +0300
> 
> > just now in normal usage, the y-or-n-p popped up in the middle of an isearch.
> >
> > point was in the middle of hte isearch still.  it was not trivial  to
> > answer the y-or-n-p (point not there) and fix state.
> 
> Reading type-break.el raised some questions: please confirm does the same
> problem occur after using the default value of 'type-break-query-function'
> that is 'yes-or-no-p', and not 'y-or-n-p'?
> 
> It looks like type-break uses 'run-at-time' and then tries to
> call 'y-or-n-p' in the middle of an isearch.  I tried both
> 
>   (run-with-timer 5 5 #'yes-or-no-p "OK? ")
>   (run-with-timer 5 5 #'y-or-n-p "OK? ")
> 
> and activated isearch.  Then the typed answer was added
> to the search string.  Emacs 28 has an option
> 'y-or-n-p-use-read-key' that you can set for type-break
> that pops up the query at random times.  But the problem
> still exists for 'yes-or-no-p'.

Is this a new problem?  That is, did we have this working OK in some
previous version of Emacs?  If so, which version had it right?

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Juri Linkov]

>>   (run-with-timer 5 5 #'yes-or-no-p "OK? ")
>>
>> and activated isearch.  Then the typed answer was added
>> to the search string.  Emacs 28 has an option
>> 'y-or-n-p-use-read-key' that you can set for type-break
>> that pops up the query at random times.  But the problem
>> still exists for 'yes-or-no-p'.
>
> Is this a new problem?  That is, did we have this working OK in some
> previous version of Emacs?  If so, which version had it right?

This never worked in isearch:

  (run-with-timer 5 5 #'yes-or-no-p "OK? ")

The patch to support this case was proposed in bug#62032.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Samuel Wales]

the clunkiness that i think of that has always been there is this,
which might or might not be related:

  - if you are using the mouse when the query occurs, the query might
be in the minibuffer.  it should be in a dialog box.
  - if you are using the kb when the query occurs, the query might be
a dialog box.  it should be in the minibuffer.

the reason is that i am almost always either in mouse mode or kb mode.
switching causes impact on my system as i have to lift my arms,  or
drop one and possibly lift it again, which can be significantly
problematic in my case for unrelated reasons.

for isearch, i do not recall previous instances of it in v<=26.1.  i
was only on 26.1 briefly.  27.1 was a first that i recall.

-- 
The Kafka Pandemic

A blog about science, health, human rights, and misopathy:
https://thekafkapandemic.blogspot.com

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Juri Linkov]

> the clunkiness that i think of that has always been there is this,
> which might or might not be related:
>
>   - if you are using the mouse when the query occurs, the query might
> be in the minibuffer.  it should be in a dialog box.
>   - if you are using the kb when the query occurs, the query might be
> a dialog box.  it should be in the minibuffer.
>
> the reason is that i am almost always either in mouse mode or kb mode.
> switching causes impact on my system as i have to lift my arms,  or
> drop one and possibly lift it again, which can be significantly
> problematic in my case for unrelated reasons.
>
> for isearch, i do not recall previous instances of it in v<=26.1.  i
> was only on 26.1 briefly.  27.1 was a first that i recall.

Indeed, this is how it was supposed to work.
But there is a regression in 29.0.

When tried

  (run-with-timer 5 5 #'yes-or-no-p "OK? ")

it pops up a dialog box in 26.3, 27.2, 28.2,
but uses the minibuffer in 29.0.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Juri Linkov]

>> the clunkiness that i think of that has always been there is this,
>> which might or might not be related:
>>
>>   - if you are using the mouse when the query occurs, the query might
>> be in the minibuffer.  it should be in a dialog box.
>>   - if you are using the kb when the query occurs, the query might be
>> a dialog box.  it should be in the minibuffer.
>>
>> the reason is that i am almost always either in mouse mode or kb mode.
>> switching causes impact on my system as i have to lift my arms,  or
>> drop one and possibly lift it again, which can be significantly
>> problematic in my case for unrelated reasons.
>>
>> for isearch, i do not recall previous instances of it in v<=26.1.  i
>> was only on 26.1 briefly.  27.1 was a first that i recall.
>
> Indeed, this is how it was supposed to work.
> But there is a regression in 29.0.
>
> When tried
>
>   (run-with-timer 5 5 #'yes-or-no-p "OK? ")
>
> it pops up a dialog box in 26.3, 27.2, 28.2,
> but uses the minibuffer in 29.0.

Sorry, this is not a regression.  It still pops up the dialog box
when you are using a mouse.  So I see no more problems.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Eli Zaretskii]

> From: Juri Linkov <juri@linkov.net>
> Cc: 64141@debbugs.gnu.org,  Eli Zaretskii <eliz@gnu.org>
> Date: Wed, 21 Jun 2023 09:49:05 +0300
> 
> >> the clunkiness that i think of that has always been there is this,
> >> which might or might not be related:
> >>
> >>   - if you are using the mouse when the query occurs, the query might
> >> be in the minibuffer.  it should be in a dialog box.
> >>   - if you are using the kb when the query occurs, the query might be
> >> a dialog box.  it should be in the minibuffer.
> >>
> >> the reason is that i am almost always either in mouse mode or kb mode.
> >> switching causes impact on my system as i have to lift my arms,  or
> >> drop one and possibly lift it again, which can be significantly
> >> problematic in my case for unrelated reasons.
> >>
> >> for isearch, i do not recall previous instances of it in v<=26.1.  i
> >> was only on 26.1 briefly.  27.1 was a first that i recall.
> >
> > Indeed, this is how it was supposed to work.
> > But there is a regression in 29.0.
> >
> > When tried
> >
> >   (run-with-timer 5 5 #'yes-or-no-p "OK? ")
> >
> > it pops up a dialog box in 26.3, 27.2, 28.2,
> > but uses the minibuffer in 29.0.
> 
> Sorry, this is not a regression.  It still pops up the dialog box
> when you are using a mouse.  So I see no more problems.

Sorry, I'm confused: could you summarize which problems are there, and
whether any of them are regressions?

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Eli Zaretskii]

> From: Samuel Wales <samologist@gmail.com>
> Date: Tue, 20 Jun 2023 15:44:45 -0700
> Cc: Juri Linkov <juri@linkov.net>, 64141@debbugs.gnu.org
> 
> the clunkiness that i think of that has always been there is this,
> which might or might not be related:
> 
>   - if you are using the mouse when the query occurs, the query might
> be in the minibuffer.  it should be in a dialog box.
>   - if you are using the kb when the query occurs, the query might be
> a dialog box.  it should be in the minibuffer.

IIUC, Juri just said that this does work correctly, so I wonder why
you say it doesn't work correctly for you.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Juri Linkov]

>> >> the clunkiness that i think of that has always been there is this,
>> >> which might or might not be related:
>> >>
>> >>   - if you are using the mouse when the query occurs, the query might
>> >> be in the minibuffer.  it should be in a dialog box.
>> >>   - if you are using the kb when the query occurs, the query might be
>> >> a dialog box.  it should be in the minibuffer.
>> >>
>> >> the reason is that i am almost always either in mouse mode or kb mode.
>> >> switching causes impact on my system as i have to lift my arms,  or
>> >> drop one and possibly lift it again, which can be significantly
>> >> problematic in my case for unrelated reasons.
>> >>
>> >> for isearch, i do not recall previous instances of it in v<=26.1.  i
>> >> was only on 26.1 briefly.  27.1 was a first that i recall.
>> >
>> > Indeed, this is how it was supposed to work.
>> > But there is a regression in 29.0.
>> >
>> > When tried
>> >
>> >   (run-with-timer 5 5 #'yes-or-no-p "OK? ")
>> >
>> > it pops up a dialog box in 26.3, 27.2, 28.2,
>> > but uses the minibuffer in 29.0.
>>
>> Sorry, this is not a regression.  It still pops up the dialog box
>> when you are using a mouse.  So I see no more problems.
>
> Sorry, I'm confused: could you summarize which problems are there, and
> whether any of them are regressions?

There are no problems to summarize.

The "modal" minibuffer is achieved by y-or-n-p-use-read-key in 28.1.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Samuel Wales]

by modal do you mean ask as gui vs. kb depending on whether currently
using mouse?

by no problems do you mean recent emacs has none?  27.1 does, but
perhaps that is stone age.

fwiw i cannot address v>27.1.
fwiw non-modal bug existed before 26.1.


On 6/21/23, Juri Linkov <juri@linkov.net> wrote:
>>> >> the clunkiness that i think of that has always been there is this,
>>> >> which might or might not be related:
>>> >>
>>> >>   - if you are using the mouse when the query occurs, the query might
>>> >> be in the minibuffer.  it should be in a dialog box.
>>> >>   - if you are using the kb when the query occurs, the query might be
>>> >> a dialog box.  it should be in the minibuffer.
>>> >>
>>> >> the reason is that i am almost always either in mouse mode or kb
>>> >> mode.
>>> >> switching causes impact on my system as i have to lift my arms,  or
>>> >> drop one and possibly lift it again, which can be significantly
>>> >> problematic in my case for unrelated reasons.
>>> >>
>>> >> for isearch, i do not recall previous instances of it in v<=26.1.  i
>>> >> was only on 26.1 briefly.  27.1 was a first that i recall.
>>> >
>>> > Indeed, this is how it was supposed to work.
>>> > But there is a regression in 29.0.
>>> >
>>> > When tried
>>> >
>>> >   (run-with-timer 5 5 #'yes-or-no-p "OK? ")
>>> >
>>> > it pops up a dialog box in 26.3, 27.2, 28.2,
>>> > but uses the minibuffer in 29.0.
>>>
>>> Sorry, this is not a regression.  It still pops up the dialog box
>>> when you are using a mouse.  So I see no more problems.
>>
>> Sorry, I'm confused: could you summarize which problems are there, and
>> whether any of them are regressions?
>
> There are no problems to summarize.
>
> The "modal" minibuffer is achieved by y-or-n-p-use-read-key in 28.1.
>


-- 
The Kafka Pandemic

A blog about science, health, human rights, and misopathy:
https://thekafkapandemic.blogspot.com

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Juri Linkov]

> by modal do you mean ask as gui vs. kb depending on whether currently
> using mouse?

By modal I mean typing any key except y or n signals an error and does nothing.

> by no problems do you mean recent emacs has none?  27.1 does, but
> perhaps that is stone age.
>
> fwiw i cannot address v>27.1.

In v>27.1 you can set y-or-n-p-use-read-key to t that makes y-or-n-p modal.

> fwiw non-modal bug existed before 26.1.

Such non-modal bug still exists in all versions for yes-or-no-p.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Samuel Wales]

i realize this isn't useful other than the stuff i have already
provided, but i find that type-break more often than not does the
wrong thing.  cursor in non-writable spot, or in a real buffer.  or in
middle of isearch.  probably various other things.

is the idea that i would upgrade to a newer version of emacs?  i do
that by upgrading debian....

On 7/11/23, Juri Linkov <juri@linkov.net> wrote:
>> by modal do you mean ask as gui vs. kb depending on whether currently
>> using mouse?
>
> By modal I mean typing any key except y or n signals an error and does
> nothing.
>
>> by no problems do you mean recent emacs has none?  27.1 does, but
>> perhaps that is stone age.
>>
>> fwiw i cannot address v>27.1.
>
> In v>27.1 you can set y-or-n-p-use-read-key to t that makes y-or-n-p modal.
>
>> fwiw non-modal bug existed before 26.1.
>
> Such non-modal bug still exists in all versions for yes-or-no-p.
>


-- 
The Kafka Pandemic

A blog about science, health, human rights, and misopathy:
https://thekafkapandemic.blogspot.com

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Juri Linkov]

> i realize this isn't useful other than the stuff i have already
> provided, but i find that type-break more often than not does the
> wrong thing.  cursor in non-writable spot, or in a real buffer.  or in
> middle of isearch.  probably various other things.
>
> is the idea that i would upgrade to a newer version of emacs?  i do
> that by upgrading debian....

Or you could take the old function `y-or-n-p' from 26.1
and add it to your config for 27.1.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Samuel Wales]

thanks

On 7/18/23, Juri Linkov <juri@linkov.net> wrote:
>> i realize this isn't useful other than the stuff i have already
>> provided, but i find that type-break more often than not does the
>> wrong thing.  cursor in non-writable spot, or in a real buffer.  or in
>> middle of isearch.  probably various other things.
>>
>> is the idea that i would upgrade to a newer version of emacs?  i do
>> that by upgrading debian....
>
> Or you could take the old function `y-or-n-p' from 26.1
> and add it to your config for 27.1.
>


-- 
The Kafka Pandemic

A blog about science, health, human rights, and misopathy:
https://thekafkapandemic.blogspot.com

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Samuel Wales]

by any chace, does emacs site have a plae where teh definition of
y-or-n-p in 26 can be found?  so tht i do not have to dl or do
dependency issues?

On 7/19/23, Samuel Wales <samologist@gmail.com> wrote:
> thanks
>
> On 7/18/23, Juri Linkov <juri@linkov.net> wrote:
>>> i realize this isn't useful other than the stuff i have already
>>> provided, but i find that type-break more often than not does the
>>> wrong thing.  cursor in non-writable spot, or in a real buffer.  or in
>>> middle of isearch.  probably various other things.
>>>
>>> is the idea that i would upgrade to a newer version of emacs?  i do
>>> that by upgrading debian....
>>
>> Or you could take the old function `y-or-n-p' from 26.1
>> and add it to your config for 27.1.
>>
>
>
> --
> The Kafka Pandemic
>
> A blog about science, health, human rights, and misopathy:
> https://thekafkapandemic.blogspot.com
>


-- 
The Kafka Pandemic

A blog about science, health, human rights, and misopathy:
https://thekafkapandemic.blogspot.com

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Juri Linkov]

> by any chace, does emacs site have a plae where teh definition of
> y-or-n-p in 26 can be found?

You can copy-paste it from here:

https://git.savannah.gnu.org/cgit/emacs.git/tree/lisp/subr.el?h=emacs-26#n2505

> so tht i do not have to dl or do dependency issues?

There should be no issues with dependency.

bug#64141: 27.1 newly has clunky type-break y-or-n-p

[Stefan Kangas]

Juri Linkov <juri@linkov.net> writes:

>> i realize this isn't useful other than the stuff i have already
>> provided, but i find that type-break more often than not does the
>> wrong thing.  cursor in non-writable spot, or in a real buffer.  or in
>> middle of isearch.  probably various other things.
>>
>> is the idea that i would upgrade to a newer version of emacs?  i do
>> that by upgrading debian....
>
> Or you could take the old function `y-or-n-p' from 26.1
> and add it to your config for 27.1.

Having read this bug thread, it seems like there is nothing more to do
here.  I'm therefore closing this bug report.

If this conclusion is incorrect and this is still an issue, please reply
to this email (use "Reply to all" in your email client) and we can
reopen the bug report.