bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

1. Start emacs (vanilla install) with -Q option in command line
2. Select `Specific Option' from File Menu > Options > Customize Emacs
3. Specify isearch-wrap-pause option when prompted with `Customize variable:'
4. Click on the Value Menu button in the Customize Option buffer
5. Select either `No pause before wrapping' or `No pause and no flashing' value
6. Click on the "Apply" button in the same buffer
7. Perform an applicable forward or reverse search
8. Search wrap fails

In GNU Emacs 28.1 (build 1, x86_64-pc-linux-gnu, GTK+ Version 3.24.24,
cairo version 1.16.0)
of 2022-06-03 built on payasam
Windowing system distributor 'The X.Org Foundation', version 11.0.12011000
System Description: Debian GNU/Linux 11 (bullseye)


Configured features:

ACL CAIRO DBUS FREETYPE GIF GLIB GMP GNUTLS GPM GSETTINGS HARFBUZZ JPEG
JSON LCMS2 LIBOTF LIBSELINUX LIBSYSTEMD LIBXML2 M17N_FLT MODULES NOTIFY
INOTIFY PDUMPER PNG RSVG SECCOMP SOUND THREADS TIFF TOOLKIT_SCROLL_BARS
X11 XDBE XIM XPM GTK3 ZLIB

Important settings:

value of $LANG: en_IN
value of $XMODIFIERS: @im=ibus
locale-coding-system: iso-latin-1-unix


Major mode: Custom


Minor modes in effect:

tooltip-mode: t
global-eldoc-mode: t
show-paren-mode: t
electric-indent-mode: t
mouse-wheel-mode: t
tool-bar-mode: t
menu-bar-mode: t
file-name-shadow-mode: t
global-font-lock-mode: t
font-lock-mode: t
blink-cursor-mode: t
auto-composition-mode: t
auto-encryption-mode: t
auto-compression-mode: t
line-number-mode: t
indent-tabs-mode: t
transient-mark-mode: t

Load-path shadows:
None found.


Features:

(shadow sort mail-extr emacsbug message rmc puny dired dired-loaddefs
rfc822 mml mml-sec epa derived epg rfc6068 epg-config gnus-util rmail
rmail-loaddefs auth-source cl-seq eieio eieio-core cl-macs
eieio-loaddefs password-cache json map text-property-search mm-decode
mm-bodies mm-encode mail-parse rfc2231 mailabbrev gmm-utils mailheader
sendmail rfc2047 rfc2045 ietf-drums mm-util mail-prsvr mail-utils
misearch multi-isearch seq byte-opt gv bytecomp byte-compile cconv
thingatpt help-fns radix-tree help-mode cus-edit pp cus-start cus-load
wid-edit time-date subr-x cl-loaddefs cl-lib iso-transl tooltip eldoc
paren electric uniquify ediff-hook vc-hooks lisp-float-type elisp-mode
mwheel term/x-win x-win term/common-win x-dnd tool-bar dnd fontset image
regexp-opt fringe tabulated-list replace newcomment text-mode lisp-mode
prog-mode register page tab-bar menu-bar rfn-eshadow isearch easymenu
timer select scroll-bar mouse jit-lock font-lock syntax font-core
term/tty-colors frame minibuffer cl-generic cham georgian utf-8-lang
misc-lang vietnamese tibetan thai tai-viet lao korean japanese eucjp-ms
cp51932 hebrew greek romanian slovak czech european ethiopic indian
cyrillic chinese composite emoji-zwj charscript charprop case-table
epa-hook jka-cmpr-hook help simple abbrev obarray cl-preloaded nadvice
button loaddefs faces cus-face macroexp files window text-properties
overlay sha1 md5 base64 format env code-pages mule custom widget
hashtable-print-readable backquote threads dbusbind inotify lcms2
dynamic-setting system-font-setting font-render-setting cairo
move-toolbar gtk x-toolkit x multi-tty make-network-process emacs)


Memory information:

((conses 16 73697 9000)
(symbols 48 8461 1)
(strings 32 23365 1869)
(string-bytes 1 725489)
(vectors 16 15533)
(vector-slots 8 201941 7551)
(floats 8 29 38)
(intervals 56 229 0)
(buffers 992 12))

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Eli Zaretskii]

> From: Arun V <arunlalv04@gmail.com>
> Date: Wed, 13 Jul 2022 12:51:53 +0530
> 
> 1. Start emacs (vanilla install) with -Q option in command line
> 2. Select `Specific Option' from File Menu > Options > Customize Emacs
> 3. Specify isearch-wrap-pause option when prompted with `Customize variable:'
> 4. Click on the Value Menu button in the Customize Option buffer
> 5. Select either `No pause before wrapping' or `No pause and no flashing' value
> 6. Click on the "Apply" button in the same buffer
> 7. Perform an applicable forward or reverse search
> 8. Search wrap fails

I cannot reproduce this here.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Eli Zaretskii]

[Please use Reply All to keep the bug tracker on the CC list.]

> From: Arun V <arunlalv04@gmail.com>
> Date: Wed, 13 Jul 2022 21:32:08 +0530
> 
> Hi Eli,
> 
> Shall I share screenshots?
> I'm able to reproduce this in plain vanilla emacs in Windows and Linux.

There's no need to show a screenshot, I believe you.  It's just that I
cannot reproduce the problem: typing C-s over and over wraps around at
the end of the buffer and goes to the beginning.  So I wonder what is
different on your system that makes this not work.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Visuwesh]

[புதன் ஜூலை 13, 2022] Eli Zaretskii wrote:

> [Please use Reply All to keep the bug tracker on the CC list.]
>
>> From: Arun V <arunlalv04@gmail.com>
>> Date: Wed, 13 Jul 2022 21:32:08 +0530
>> 
>> Hi Eli,
>> 
>> Shall I share screenshots?
>> I'm able to reproduce this in plain vanilla emacs in Windows and Linux.
>
> There's no need to show a screenshot, I believe you.  It's just that I
> cannot reproduce the problem: typing C-s over and over wraps around at
> the end of the buffer and goes to the beginning.  So I wonder what is
> different on your system that makes this not work.

FWIW, I cannot reproduce it here as well.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

Typing C-s over and over does wrap around at the end of the buffer.
But that shouldn't be the behaviour when isearch-wrap-pause is set to
`no' or `no-ding', right?

The following is an excerpt from the documentation I'm referring to
[https://www.gnu.org/software/emacs/manual/html_node/emacs/Repeat-Isearch.html]

"You can control what happens when there are no more matches by
customizing the isearch-wrap-pause user option. If it is t (the
default), signal an error. (Repeating the search will wrap around.) If
no, issue a ding and wrap immediately after reaching the last match.
If no-ding, wrap immediately, but don’t ding. Finally, if nil, never
wrap, but just stop at the last match."

Please let me know if I'm missing anything here.


On Wed, 13 Jul 2022 at 22:29, Eli Zaretskii <eliz@gnu.org> wrote:
>
> [Please use Reply All to keep the bug tracker on the CC list.]
>
> > From: Arun V <arunlalv04@gmail.com>
> > Date: Wed, 13 Jul 2022 21:32:08 +0530
> >
> > Hi Eli,
> >
> > Shall I share screenshots?
> > I'm able to reproduce this in plain vanilla emacs in Windows and Linux.
>
> There's no need to show a screenshot, I believe you.  It's just that I
> cannot reproduce the problem: typing C-s over and over wraps around at
> the end of the buffer and goes to the beginning.  So I wonder what is
> different on your system that makes this not work.



-- 
--
Thanks and regards,
Arunlal Venugopal

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Eli Zaretskii]

> From: Arun V <arunlalv04@gmail.com>
> Date: Wed, 13 Jul 2022 22:52:49 +0530
> Cc: 56535@debbugs.gnu.org
> 
> Typing C-s over and over does wrap around at the end of the buffer.
> But that shouldn't be the behaviour when isearch-wrap-pause is set to
> `no' or `no-ding', right?
> 
> The following is an excerpt from the documentation I'm referring to
> [https://www.gnu.org/software/emacs/manual/html_node/emacs/Repeat-Isearch.html]
> 
> "You can control what happens when there are no more matches by
> customizing the isearch-wrap-pause user option. If it is t (the
> default), signal an error. (Repeating the search will wrap around.) If
> no, issue a ding and wrap immediately after reaching the last match.
> If no-ding, wrap immediately, but don’t ding. Finally, if nil, never
> wrap, but just stop at the last match."

Here it says "If no-ding, wrap immediately, but don’t ding."  And that
is what I see.  Where is the problem, or what am I missing?

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

For more clarity, here's an example.

Assume it's a plain vanilla emacs install with all options set to
their default values, and it has in it a buffer with the following
content (of just 3 lines).

Line one of buffer
Line two of buffer
Line three of buffer

If the point is at the beginning of the second line, and we issue a
C-r and start typing the following characters "L", "i", "n", "e", " ",
the point would be moved to the first line. If we continue to type
"t", the minibuffer will display "Failing I-search backward" and the
point will still remain on the first line. If we press C-r again, the
minibuffer will display "Wrapped I-search backward" and move the point
to the third line.

This is the default behaviour (with isearch-wrap-pause set to t). I
can see the documented behaviour when I set the value to nil i.e. no
wrap around.

But when I set the value to `no' or `no-ding', I was expecting the
point to automatically move to the third line immediately after typing
"t" as mentioned in the above use case. Atleast, this is what I could
understand from the documentation. Please correct me if I'm wrong.

On Wed, 13 Jul 2022 at 22:52, Arun V <arunlalv04@gmail.com> wrote:
>
> Typing C-s over and over does wrap around at the end of the buffer.
> But that shouldn't be the behaviour when isearch-wrap-pause is set to
> `no' or `no-ding', right?
>
> The following is an excerpt from the documentation I'm referring to
> [https://www.gnu.org/software/emacs/manual/html_node/emacs/Repeat-Isearch.html]
>
> "You can control what happens when there are no more matches by
> customizing the isearch-wrap-pause user option. If it is t (the
> default), signal an error. (Repeating the search will wrap around.) If
> no, issue a ding and wrap immediately after reaching the last match.
> If no-ding, wrap immediately, but don’t ding. Finally, if nil, never
> wrap, but just stop at the last match."
>
> Please let me know if I'm missing anything here.
>
>
> On Wed, 13 Jul 2022 at 22:29, Eli Zaretskii <eliz@gnu.org> wrote:
> >
> > [Please use Reply All to keep the bug tracker on the CC list.]
> >
> > > From: Arun V <arunlalv04@gmail.com>
> > > Date: Wed, 13 Jul 2022 21:32:08 +0530
> > >
> > > Hi Eli,
> > >
> > > Shall I share screenshots?
> > > I'm able to reproduce this in plain vanilla emacs in Windows and Linux.
> >
> > There's no need to show a screenshot, I believe you.  It's just that I
> > cannot reproduce the problem: typing C-s over and over wraps around at
> > the end of the buffer and goes to the beginning.  So I wonder what is
> > different on your system that makes this not work.
>
>
>
> --
> --
> Thanks and regards,
> Arunlal Venugopal



-- 
--
Thanks and regards,
Arunlal Venugopal

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

When I set the isearch-wrap-pause option to `no' or `no-ding', it
still displays "Failing I-search backward" on entering a non-matching
character. I still have to press C-r again to wrap around. This is the
same behaviour as having set the value of isearch-wrap-pause to t.

On Wed, 13 Jul 2022 at 23:23, Arun V <arunlalv04@gmail.com> wrote:
>
> For more clarity, here's an example.
>
> Assume it's a plain vanilla emacs install with all options set to
> their default values, and it has in it a buffer with the following
> content (of just 3 lines).
>
> Line one of buffer
> Line two of buffer
> Line three of buffer
>
> If the point is at the beginning of the second line, and we issue a
> C-r and start typing the following characters "L", "i", "n", "e", " ",
> the point would be moved to the first line. If we continue to type
> "t", the minibuffer will display "Failing I-search backward" and the
> point will still remain on the first line. If we press C-r again, the
> minibuffer will display "Wrapped I-search backward" and move the point
> to the third line.
>
> This is the default behaviour (with isearch-wrap-pause set to t). I
> can see the documented behaviour when I set the value to nil i.e. no
> wrap around.
>
> But when I set the value to `no' or `no-ding', I was expecting the
> point to automatically move to the third line immediately after typing
> "t" as mentioned in the above use case. Atleast, this is what I could
> understand from the documentation. Please correct me if I'm wrong.
>
> On Wed, 13 Jul 2022 at 22:52, Arun V <arunlalv04@gmail.com> wrote:
> >
> > Typing C-s over and over does wrap around at the end of the buffer.
> > But that shouldn't be the behaviour when isearch-wrap-pause is set to
> > `no' or `no-ding', right?
> >
> > The following is an excerpt from the documentation I'm referring to
> > [https://www.gnu.org/software/emacs/manual/html_node/emacs/Repeat-Isearch.html]
> >
> > "You can control what happens when there are no more matches by
> > customizing the isearch-wrap-pause user option. If it is t (the
> > default), signal an error. (Repeating the search will wrap around.) If
> > no, issue a ding and wrap immediately after reaching the last match.
> > If no-ding, wrap immediately, but don’t ding. Finally, if nil, never
> > wrap, but just stop at the last match."
> >
> > Please let me know if I'm missing anything here.
> >
> >
> > On Wed, 13 Jul 2022 at 22:29, Eli Zaretskii <eliz@gnu.org> wrote:
> > >
> > > [Please use Reply All to keep the bug tracker on the CC list.]
> > >
> > > > From: Arun V <arunlalv04@gmail.com>
> > > > Date: Wed, 13 Jul 2022 21:32:08 +0530
> > > >
> > > > Hi Eli,
> > > >
> > > > Shall I share screenshots?
> > > > I'm able to reproduce this in plain vanilla emacs in Windows and Linux.
> > >
> > > There's no need to show a screenshot, I believe you.  It's just that I
> > > cannot reproduce the problem: typing C-s over and over wraps around at
> > > the end of the buffer and goes to the beginning.  So I wonder what is
> > > different on your system that makes this not work.
> >
> >
> >
> > --
> > --
> > Thanks and regards,
> > Arunlal Venugopal
>
>
>
> --
> --
> Thanks and regards,
> Arunlal Venugopal



-- 
--
Thanks and regards,
Arunlal Venugopal

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Eli Zaretskii]

> From: Arun V <arunlalv04@gmail.com>
> Date: Wed, 13 Jul 2022 23:23:00 +0530
> Cc: 56535@debbugs.gnu.org
> 
> But when I set the value to `no' or `no-ding', I was expecting the
> point to automatically move to the third line immediately after typing
> "t" as mentioned in the above use case. Atleast, this is what I could
> understand from the documentation. Please correct me if I'm wrong.

Ah, your expectations are wrong: Isearch only wraps silently if you
repeatedly type C-r without changing the search string.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

Ahh ok. That clarifies. Thanks for the explanation. And thanks for
your time as well. Please close this bug.

On a side note, automatically wrapping during search string input
would be a cool feature though. ;)

On Wed, 13 Jul 2022 at 23:29, Eli Zaretskii <eliz@gnu.org> wrote:
>
> > From: Arun V <arunlalv04@gmail.com>
> > Date: Wed, 13 Jul 2022 23:23:00 +0530
> > Cc: 56535@debbugs.gnu.org
> >
> > But when I set the value to `no' or `no-ding', I was expecting the
> > point to automatically move to the third line immediately after typing
> > "t" as mentioned in the above use case. Atleast, this is what I could
> > understand from the documentation. Please correct me if I'm wrong.
>
> Ah, your expectations are wrong: Isearch only wraps silently if you
> repeatedly type C-r without changing the search string.



-- 
--
Thanks and regards,
Arunlal Venugopal

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Eli Zaretskii]

> From: Arun V <arunlalv04@gmail.com>
> Date: Wed, 13 Jul 2022 23:29:04 +0530
> Cc: 56535@debbugs.gnu.org
> 
> When I set the isearch-wrap-pause option to `no' or `no-ding', it
> still displays "Failing I-search backward" on entering a non-matching
> character. I still have to press C-r again to wrap around. This is the
> same behaviour as having set the value of isearch-wrap-pause to t.

I think this is the expected behavior.  This feature only controls
wraparound of repeated Isearch.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Juri Linkov]

[-- Attachment #1: Type: text/plain, Size: 444 bytes --]

> But when I set the value to `no' or `no-ding', I was expecting the
> point to automatically move to the third line immediately after typing
> "t" as mentioned in the above use case.

It was never intended to wrap around on typing a letter or yanking a text.
If such a behavior is a good idea, it could be added as a new value
to this variable.

Please try the following patch after setting isearch-wrap-pause to the
new value 'auto-repeat':


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: isearch-wrap-pause-auto-repeat.patch --]
[-- Type: text/x-diff, Size: 503 bytes --]

diff --git a/lisp/isearch.el b/lisp/isearch.el
index cedbfdc2e8..6e14003614 100644
--- a/lisp/isearch.el
+++ b/lisp/isearch.el
@@ -2838,6 +2838,9 @@ isearch-search-and-update
 			    isearch-barrier
 			    (1+ isearch-other-end)))))
       (isearch-search)
+      (when (and (eq isearch-wrap-pause 'auto-repeat)
+                 (not isearch-success))
+        (isearch-repeat (if isearch-forward 'forward 'backward)))
       ))
   (isearch-push-state)
   (if isearch-op-fun (funcall isearch-op-fun))

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

[-- Attachment #1: Type: text/plain, Size: 679 bytes --]

Hey Juri,

Regret the delayed response.

Yes, I do believe this would be a good idea.

Let me try this patch atop emacs-28.1 and get back to you.

On Thu, 14 Jul, 2022, 01:21 Juri Linkov, <juri@linkov.net> wrote:

> > But when I set the value to `no' or `no-ding', I was expecting the
> > point to automatically move to the third line immediately after typing
> > "t" as mentioned in the above use case.
>
> It was never intended to wrap around on typing a letter or yanking a text.
> If such a behavior is a good idea, it could be added as a new value
> to this variable.
>
> Please try the following patch after setting isearch-wrap-pause to the
> new value 'auto-repeat':
>
>

[-- Attachment #2: Type: text/html, Size: 1155 bytes --]

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

Thanks Juri. Your patch works.

On Sat, 16 Jul 2022 at 14:51, Arun V <arunlalv04@gmail.com> wrote:
>
> Hey Juri,
>
> Regret the delayed response.
>
> Yes, I do believe this would be a good idea.
>
> Let me try this patch atop emacs-28.1 and get back to you.
>
> On Thu, 14 Jul, 2022, 01:21 Juri Linkov, <juri@linkov.net> wrote:
>>
>> > But when I set the value to `no' or `no-ding', I was expecting the
>> > point to automatically move to the third line immediately after typing
>> > "t" as mentioned in the above use case.
>>
>> It was never intended to wrap around on typing a letter or yanking a text.
>> If such a behavior is a good idea, it could be added as a new value
>> to this variable.
>>
>> Please try the following patch after setting isearch-wrap-pause to the
>> new value 'auto-repeat':
>>


-- 
--
Thanks and regards,
Arunlal Venugopal

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Juri Linkov]

>>> Please try the following patch after setting isearch-wrap-pause to the
>>> new value 'auto-repeat':
>>>
>> Regret the delayed response.
>>
>> Yes, I do believe this would be a good idea.
>>
>> Let me try this patch atop emacs-28.1 and get back to you.
>
> Thanks Juri. Your patch works.

No problem with the delay.  Thanks for confirming that it works.

Now we have to find a better name to describe what it really does.
Probably it should have the prefix "no-" since it makes no pause
like the existing two values 'no' and 'no-ding':

  :type '(choice (const :tag "Pause before wrapping" t)
                 (const :tag "No pause before wrapping" no)
                 (const :tag "No pause and no flashing" no-ding)
                 (const :tag "Disable wrapping" nil))

This raises the question: should it ding and flash on wrapping?
Or two new separate values are needed: one to ding when wrapping
on typing a character, and another not to ding?  To ding, or not to ding?

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

Definitely ding. Lot of people do prefer visual feedback.

How about this?

  :type '(choice (const :tag "Pause before wrapping" t)
                 (const :tag "No pause before wrapping for
non-incremental search" nis-no)
                 (const :tag "No pause and no flashing for
non-incremental search" nis-no-ding)
                 (const :tag "Absolutely no pause before wrapping" no)
 ;;This will cover non-incremental searches as well
                 (const :tag "Absolutely no pause and no flashing"
no-ding)  ;;This will cover non-incremental searches as well
                 (const :tag "Disable wrapping" nil))

On Mon, 18 Jul 2022 at 01:05, Juri Linkov <juri@linkov.net> wrote:
>
> >>> Please try the following patch after setting isearch-wrap-pause to the
> >>> new value 'auto-repeat':
> >>>
> >> Regret the delayed response.
> >>
> >> Yes, I do believe this would be a good idea.
> >>
> >> Let me try this patch atop emacs-28.1 and get back to you.
> >
> > Thanks Juri. Your patch works.
>
> No problem with the delay.  Thanks for confirming that it works.
>
> Now we have to find a better name to describe what it really does.
> Probably it should have the prefix "no-" since it makes no pause
> like the existing two values 'no' and 'no-ding':
>
>   :type '(choice (const :tag "Pause before wrapping" t)
>                  (const :tag "No pause before wrapping" no)
>                  (const :tag "No pause and no flashing" no-ding)
>                  (const :tag "Disable wrapping" nil))
>
> This raises the question: should it ding and flash on wrapping?
> Or two new separate values are needed: one to ding when wrapping
> on typing a character, and another not to ding?  To ding, or not to ding?



-- 
--
Thanks and regards,
Arunlal Venugopal

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Juri Linkov]

>> Now we have to find a better name to describe what it really does.
>> Probably it should have the prefix "no-" since it makes no pause
>> like the existing two values 'no' and 'no-ding':
>>
>>   :type '(choice (const :tag "Pause before wrapping" t)
>>                  (const :tag "No pause before wrapping" no)
>>                  (const :tag "No pause and no flashing" no-ding)
>>                  (const :tag "Disable wrapping" nil))
>>
>> This raises the question: should it ding and flash on wrapping?
>> Or two new separate values are needed: one to ding when wrapping
>> on typing a character, and another not to ding?  To ding, or not to ding?
>
> Definitely ding. Lot of people do prefer visual feedback.
>
> How about this?
>
> :type '(choice
> (const :tag "Pause before wrapping" t)
> (const :tag "No pause before wrapping for non-incremental search" nis-no)
> (const :tag "No pause and no flashing for non-incremental search" nis-no-ding)
> ;;This will cover non-incremental searches as well
> (const :tag "Absolutely no pause before wrapping" no)
> ;;This will cover non-incremental searches as well
> (const :tag "Absolutely no pause and no flashing" no-ding)
> (const :tag "Disable wrapping" nil))

I suppose non-incremental search here means repeatable search,
i.e. the search that is being repeated with C-s C-s.
Maybe then better to emphasize this difference explicitly?
For example: 'inc-no' and 'inc-no-ding' for incremental wrapping,
and leave the existing 'no' and 'no-ding' for repeatable wrapping.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

>I suppose non-incremental search here means repeatable search,
>i.e. the search that is being repeated with C-s C-s.
>Maybe then better to emphasize this difference explicitly?
>For example: 'inc-no' and 'inc-no-ding' for incremental wrapping,
>and leave the existing 'no' and 'no-ding' for repeatable wrapping.

I too thought of it as this would preserve the original meaning of the
current values in letter and spirit.

But won't it give the notion that 'inc-no-ding' and 'no-ding' are
mutually exclusive?
Meaning, setting value as 'inc-no-ding' won't wrap for repeatable
searches, and setting 'no-ding' won't wrap for incremental searches.

Shouldn't we be conveying that one of the options includes/performs
the behaviour of all the other ones, in addition to handling the new
scenario?

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Juri Linkov]

>>I suppose non-incremental search here means repeatable search,
>>i.e. the search that is being repeated with C-s C-s.
>>Maybe then better to emphasize this difference explicitly?
>>For example: 'inc-no' and 'inc-no-ding' for incremental wrapping,
>>and leave the existing 'no' and 'no-ding' for repeatable wrapping.
>
> I too thought of it as this would preserve the original meaning of the
> current values in letter and spirit.
>
> But won't it give the notion that 'inc-no-ding' and 'no-ding' are
> mutually exclusive?
> Meaning, setting value as 'inc-no-ding' won't wrap for repeatable
> searches, and setting 'no-ding' won't wrap for incremental searches.
>
> Shouldn't we be conveying that one of the options includes/performs
> the behaviour of all the other ones, in addition to handling the new
> scenario?

It's difficult to choose self-descriptive symbols.  So we need to have
just good enough symbol names, then describe their exact meaning in the
documentation.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

> It's difficult to choose self-descriptive symbols.  So we need to have
> just good enough symbol names, then describe their exact meaning in the
> documentation.

Then I guess introducing 'inc-no' and 'inc-no-ding' for incremental
wrapping should do as it preserves the meaning of the existing values.

Just  one more question.

Why shouldn't the behaviour of 'no' and 'no-ding' be changed to handle
incremental searches as well, instead of introducing new values?

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Juri Linkov]

[-- Attachment #1: Type: text/plain, Size: 693 bytes --]

>> It's difficult to choose self-descriptive symbols.  So we need to have
>> just good enough symbol names, then describe their exact meaning in the
>> documentation.
>
> Then I guess introducing 'inc-no' and 'inc-no-ding' for incremental
> wrapping should do as it preserves the meaning of the existing values.
>
> Just  one more question.
>
> Why shouldn't the behaviour of 'no' and 'no-ding' be changed to handle
> incremental searches as well, instead of introducing new values?

Good question.  The reason was to maintain "backward-compatibility".
But do we really need to restrict the existing values only to C-s C-s,
and not to incremental commands?  In this case here is a new patch:


[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: isearch-wrap-pause-auto-repeat.patch --]
[-- Type: text/x-diff, Size: 506 bytes --]

diff --git a/lisp/isearch.el b/lisp/isearch.el
index 8f480a87d9..ad30b5e514 100644
--- a/lisp/isearch.el
+++ b/lisp/isearch.el
@@ -2838,6 +2839,9 @@ isearch-search-and-update
 			    isearch-barrier
 			    (1+ isearch-other-end)))))
       (isearch-search)
+      (when (and (memq isearch-wrap-pause '(no no-ding))
+                 (not isearch-success))
+        (isearch-repeat (if isearch-forward 'forward 'backward)))
       ))
   (isearch-push-state)
   (if isearch-op-fun (funcall isearch-op-fun))

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

> Good question.  The reason was to maintain "backward-compatibility".
> But do we really need to restrict the existing values only to C-s C-s,
> and not to incremental commands?  In this case here is a new patch:


I tested your patch and it works, Juri.

Yep, "backward-compatibility" is one reason. But, in effect this patch
will also retain part of the original behaviour.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

[-- Attachment #1: Type: text/plain, Size: 88 bytes --]

Juri,

Just curious to know if this patch will be included as part of the next
release.

[-- Attachment #2: Type: text/html, Size: 170 bytes --]

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Juri Linkov]

> Just curious to know if this patch will be included as part of the next
> release.

The problem is that I was stuck when trying to document the
new non-repeatable incremental behavior of the old values
`no' and `no-ding' of `isearch-wrap-pause'.

Currently this is documented in the Info node
(info "(emacs) Repeat Isearch"):

     You can control what happens when there are no more matches by
  customizing the ‘isearch-wrap-pause’ user option.  If it is ‘t’ (the
  default), signal an error.  (Repeating the search will wrap around.)  If
  ‘no’, issue a ‘ding’ and wrap immediately after reaching the last match.
  If ‘no-ding’, wrap immediately, but don’t ‘ding’.  Finally, if ‘nil’,
  never wrap, but just stop at the last match.

The problem is that everything in this Info node is about
repeating the search with `C-s C-s'.  I can't find a suitable place
to describe the additional incremental wrapping here.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Eli Zaretskii]

> From: Juri Linkov <juri@linkov.net>
> Cc: Eli Zaretskii <eliz@gnu.org>,  visuweshm@gmail.com,  56535@debbugs.gnu.org
> Date: Sun, 31 Jul 2022 22:58:25 +0300
> 
> Currently this is documented in the Info node
> (info "(emacs) Repeat Isearch"):
> 
>      You can control what happens when there are no more matches by
>   customizing the ‘isearch-wrap-pause’ user option.  If it is ‘t’ (the
>   default), signal an error.  (Repeating the search will wrap around.)  If
>   ‘no’, issue a ‘ding’ and wrap immediately after reaching the last match.
>   If ‘no-ding’, wrap immediately, but don’t ‘ding’.  Finally, if ‘nil’,
>   never wrap, but just stop at the last match.
> 
> The problem is that everything in this Info node is about
> repeating the search with `C-s C-s'.  I can't find a suitable place
> to describe the additional incremental wrapping here.

Why is this node not appropriate for whatever you want to document?

Or maybe I don't have a clear idea of what you want to add there?

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Juri Linkov]

>> Currently this is documented in the Info node
>> (info "(emacs) Repeat Isearch"):
>>
>>      You can control what happens when there are no more matches by
>>   customizing the ‘isearch-wrap-pause’ user option.  If it is ‘t’ (the
>>   default), signal an error.  (Repeating the search will wrap around.)  If
>>   ‘no’, issue a ‘ding’ and wrap immediately after reaching the last match.
>>   If ‘no-ding’, wrap immediately, but don’t ‘ding’.  Finally, if ‘nil’,
>>   never wrap, but just stop at the last match.
>>
>> The problem is that everything in this Info node is about
>> repeating the search with `C-s C-s'.  I can't find a suitable place
>> to describe the additional incremental wrapping here.
>
> Why is this node not appropriate for whatever you want to document?
>
> Or maybe I don't have a clear idea of what you want to add there?

The new additional behavior of ‘no’ and ‘no-ding’ is to wrap on appending
a character to the search string.  Maybe then something like this:

diff --git a/doc/emacs/search.texi b/doc/emacs/search.texi
index 27d4db8541..7cf3a3f5a9 100644
--- a/doc/emacs/search.texi
+++ b/doc/emacs/search.texi
@@ -228,8 +228,9 @@ Repeat Isearch
 @code{t} (the default), signal an error.  (Repeating the search will
 wrap around.)  If @code{no}, issue a @code{ding} and wrap immediately
 after reaching the last match.  If @code{no-ding}, wrap immediately,
-but don't @code{ding}.  Finally, if @code{nil}, never wrap, but just
-stop at the last match.
+but don't @code{ding}.  With the values @code{ding} and @code{no-ding}
+the search will try to wrap around also on typing a character.
+Finally, if @code{nil}, never wrap, but just stop at the last match.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Eli Zaretskii]

> From: Juri Linkov <juri@linkov.net>
> Cc: arunlalv04@gmail.com,  visuweshm@gmail.com,  56535@debbugs.gnu.org
> Date: Mon, 01 Aug 2022 20:55:43 +0300
> 
> The new additional behavior of ‘no’ and ‘no-ding’ is to wrap on appending
> a character to the search string.  Maybe then something like this:
> 
> diff --git a/doc/emacs/search.texi b/doc/emacs/search.texi
> index 27d4db8541..7cf3a3f5a9 100644
> --- a/doc/emacs/search.texi
> +++ b/doc/emacs/search.texi
> @@ -228,8 +228,9 @@ Repeat Isearch
>  @code{t} (the default), signal an error.  (Repeating the search will
>  wrap around.)  If @code{no}, issue a @code{ding} and wrap immediately
>  after reaching the last match.  If @code{no-ding}, wrap immediately,
> -but don't @code{ding}.  Finally, if @code{nil}, never wrap, but just
> -stop at the last match.
> +but don't @code{ding}.  With the values @code{ding} and @code{no-ding}
> +the search will try to wrap around also on typing a character.
> +Finally, if @code{nil}, never wrap, but just stop at the last match.

LGTM, since that node already describes what happens when the search
wraps.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Juri Linkov]

close 56535 29.0.50
thanks

>> @@ -228,8 +228,9 @@ Repeat Isearch
>>  @code{t} (the default), signal an error.  (Repeating the search will
>>  wrap around.)  If @code{no}, issue a @code{ding} and wrap immediately
>>  after reaching the last match.  If @code{no-ding}, wrap immediately,
>> -but don't @code{ding}.  Finally, if @code{nil}, never wrap, but just
>> -stop at the last match.
>> +but don't @code{ding}.  With the values @code{ding} and @code{no-ding}
>> +the search will try to wrap around also on typing a character.
>> +Finally, if @code{nil}, never wrap, but just stop at the last match.
>
> LGTM, since that node already describes what happens when the search
> wraps.

So now pushed and closed.

bug#56535: 28.1; `no' and `no-ding' values for isearch-wrap-pause option (since Emacs 28.1) not working as documented

[Arun V]

Thanks Juri, Eli.