bug#37447: 26.3; Commands on <delete> and <backspace>

bug#37447: 26.3; Commands on <delete> and <backspace>

[Sebastian Urban]

Just a few thoughts about deleting and killing commands.

We have:
+ "Delete"
   - <deletechar> [<delete>] (delete-forward-char)
   - <C-delete> (kill-word)
   - M-DEL [<M-delete>] (backward-kill-word)

? Perhaps C-<delete> should do (backward-kill-word) and <M-delete>
   should (kill-word).  Then <M-delete> would be like 'M-d', which
   seems to be correct way.

+ "Backspace"
   - DEL [<backspace>] (delete-backward-char)
   - <C-backspace> (backward-kill-word)
   - M-DEL [M-<backspace>] (backward-kill-word)

? <C-backspace> as another (backward-kill-word) on the same button is
   useless(?), maybe change it to (backward-kill-sentence), because
   it's less typing than 'C-x DEL'?  And then (perhaps) assign 'C-x
   DEL' to new function "backward kill line" (kill-line with 0
   argument?).  OR let 'C-x DEL' backward kill sentence, and assign
   backward kill line to 'C-DEL'.

? Also perhaps rebind 'C-S-backspace' to 'C-M-backspace' for
   (kill-whole-line)?


S. U.

GNU Emacs 26.3 (build 1, x86_64-w64-mingw32) of 2019-08-29

bug#37447: 26.3; Commands on <delete> and <backspace>

[Eli Zaretskii]

> From: Sebastian Urban <mrsebastianurban@gmail.com>
> Date: Wed, 18 Sep 2019 13:45:57 +0200
> 
> We have:
> + "Delete"
>    - <deletechar> [<delete>] (delete-forward-char)
>    - <C-delete> (kill-word)
>    - M-DEL [<M-delete>] (backward-kill-word)
> 
> ? Perhaps C-<delete> should do (backward-kill-word) and <M-delete>
>    should (kill-word).  Then <M-delete> would be like 'M-d', which
>    seems to be correct way.
> 
> + "Backspace"
>    - DEL [<backspace>] (delete-backward-char)
>    - <C-backspace> (backward-kill-word)
>    - M-DEL [M-<backspace>] (backward-kill-word)
> 
> ? <C-backspace> as another (backward-kill-word) on the same button is
>    useless(?), maybe change it to (backward-kill-sentence), because
>    it's less typing than 'C-x DEL'?  And then (perhaps) assign 'C-x
>    DEL' to new function "backward kill line" (kill-line with 0
>    argument?).  OR let 'C-x DEL' backward kill sentence, and assign
>    backward kill line to 'C-DEL'.
> 
> ? Also perhaps rebind 'C-S-backspace' to 'C-M-backspace' for
>    (kill-whole-line)?

These key bindings are too old to change them in the default
configuration.  You can, of course, change them in your personal
configuration.

Thanks.

bug#37447: 26.3; Commands on <delete> and <backspace>

[Sebastian Urban]

> These key bindings are too old to change them in the default
> configuration.

So inconsistency with 'M-d' and 'M-<delete>' will stay and
'C-<backspace>' & 'M-<backspace>' doing the same thing will also stay
because of history?  I could ignore 'C-S-<backspace>', but previous
two cases should be "updated".

But if "No." means "No." - close the bug, please.  Unless we are
waiting for more opinions.

bug#37447: 26.3; Commands on <delete> and <backspace>

[Juri Linkov]

>> These key bindings are too old to change them in the default
>> configuration.
>
> So inconsistency with 'M-d' and 'M-<delete>' will stay and
> 'C-<backspace>' & 'M-<backspace>' doing the same thing will also stay
> because of history?  I could ignore 'C-S-<backspace>', but previous
> two cases should be "updated".
>
> But if "No." means "No." - close the bug, please.  Unless we are
> waiting for more opinions.

Imagine many users using the same keys for years, and now
they will start doing the opposite - a recipe for disaster.
But if a safer solution is possible, then I'd recommend
binding one of C- or M- prefixes to sexp killing commands
because their default prefix C-M- can't be used: C-M-DEL
runs system shutdown, and C-M-<backspace> kills the X server.
There is a replacement C-M-k for kill-sexp, but no key
for backward-kill-sexp.

bug#37447: 26.3; Commands on <delete> and <backspace>

[Drew Adams]

> But if a safer solution is possible, then I'd recommend
> binding one of C- or M- prefixes to sexp killing commands
> because their default prefix C-M- can't be used: C-M-DEL
> runs system shutdown, and C-M-<backspace> kills the X server.
> There is a replacement C-M-k for kill-sexp, but no key
> for backward-kill-sexp.

Please don't use just `C-' or just `M-' for sexp
commands.  Please use the conventional `C-M-'.

FWIW, `C-M-k' for `kill-sexp', and `C-M-<backspace>
for `backward-kill-sexp', work on at least some
platforms, such as MS Windows. 

And for uses of Meta that are problematic for some
platforms, users can always use ESC.  E.g. `ESC C-k'
instead of `C-M-k' or `ESC C-<delete>' for
`C-M-<delete>'.  Admittedly not as handy, but usable
(and users with that problem can always rebind them).

(BTW, did you mean `C-M-<delete>' instead of `C-M-DEL'?
The latter doesn't seem to be bound by default.)

bug#37447: 26.3; Commands on <delete> and <backspace>

[Noam Postavsky]

Drew Adams <drew.adams@oracle.com> writes:

>> But if a safer solution is possible, then I'd recommend
>> binding one of C- or M- prefixes to sexp killing commands
>> because their default prefix C-M- can't be used: C-M-DEL
>> runs system shutdown, and C-M-<backspace> kills the X server.

> FWIW, `C-M-k' for `kill-sexp', and `C-M-<backspace>
> for `backward-kill-sexp', work on at least some
> platforms, such as MS Windows. 

FWIW, recent (since around 2007~2010 from what I see on the web) Xorg
disables C-M-<backspace> as the kill X server by default.

bug#37447: 26.3; Commands on <delete> and <backspace>

[Juri Linkov]

>>> But if a safer solution is possible, then I'd recommend
>>> binding one of C- or M- prefixes to sexp killing commands
>>> because their default prefix C-M- can't be used: C-M-DEL
>>> runs system shutdown, and C-M-<backspace> kills the X server.
>
>> FWIW, `C-M-k' for `kill-sexp', and `C-M-<backspace>
>> for `backward-kill-sexp', work on at least some
>> platforms, such as MS Windows.

Currently I use `M-- C-M-k' as a replacement of `backward-kill-sexp'.

> FWIW, recent (since around 2007~2010 from what I see on the web) Xorg
> disables C-M-<backspace> as the kill X server by default.

I can't confirm since I'm afraid of typing these keys :)

bug#37447: 26.3; Commands on <delete> and <backspace>

[Dmitry Gutov]

On 20.09.2019 1:13, Juri Linkov wrote:

>> FWIW, recent (since around 2007~2010 from what I see on the web) Xorg
>> disables C-M-<backspace> as the kill X server by default.
> 
> I can't confirm since I'm afraid of typing these keys :)

I can confirm, and I tried and failed to use this combination to kill 
Xorg a couple of years ago. Just once :)

bug#37447: 26.3; Commands on <delete> and <backspace>

[Sebastian Urban]

> Imagine many users using the same keys for years, and now
> they will start doing the opposite - a recipe for disaster.

And that's why I'm not proposing changing C-b/n/p/f to C-h/j/k/l.  But
these two cases (without C-S-<backspace>) I pointed out doesn't seem
very critical, moreover they look like a bug (with 'M-<delete>'
deleting backward instead of forward as 'M-d' does - 'C-d' does
similar thing do <delete>, so 'M-...' should also be similar) or like
a leftover from the past ('C-<backspace>') that could be reused for
something useful.

Also, there are "obsolete" functions that are removed after few
versions of Emacs, so why not do the same thing to key bindings?  They
don't change that often, so it won't be a revolution, right?  Keeping
old key bindings forever(?), especially when user can change them, is
rather strange - confusion among new users, less keys for new/other
commands, etc.  I think Emacs should have one default key per command,
older should be marked as "obsolete" and removed after 1 or 2 major
versions.

But, this is from perspective of a beginner, so...

bug#37447: 26.3; Commands on <delete> and <backspace>

[Eli Zaretskii]

> From: Sebastian Urban <mrsebastianurban@gmail.com>
> Date: Fri, 20 Sep 2019 10:17:38 +0200
> Cc: 37447@debbugs.gnu.org
> 
> Also, there are "obsolete" functions that are removed after few
> versions of Emacs, so why not do the same thing to key bindings?

Because they are not "obsolete" by any measure.  They are highly
popular key bindings that I, FWIW, am using every day.

And btw, I can never understand these arguments about changing what
Emacs does by default.  You can easily rebind the keys yourself, so
why force your personal preferences on everyone else?

> Keeping old key bindings forever(?), especially when user can change
> them, is rather strange

It's not strange, it's called stability.

bug#37447: 26.3; Commands on <delete> and <backspace>

[Sebastian Urban]

> (...) They are highly popular key bindings that I, FWIW, am using
> every day.

Even 'M-<delete>' that does something opposite to 'M-d', when <delete>
and 'C-d' do the same thing?  Doesn't that bother you?  For me it's
inconsistency, but maybe there is a reason for this...

> And btw, I can never understand these arguments about changing what
> Emacs does by default.  You can easily rebind the keys yourself, so
> why force your personal preferences on everyone else?

If I would want to change 'C-a' to 'C-b' (or 'C-S-<backspace>' as
I tried, but not any more) then you are right, but not in this case.

The above example is not a change just because I like it (I don't even
use it), but rather spotting inconsistency and pointing it out.

As for 'C-<backspace>' also "no", because I was trying to reuse
something that is already there ('M-<backspace>').  But fine, someone
may use 'C-...', so I could leave it alone as well.

bug#37447: 26.3; Commands on <delete> and <backspace>

[Eli Zaretskii]

> From: Sebastian Urban <mrsebastianurban@gmail.com>
> Cc: 37447@debbugs.gnu.org
> Date: Fri, 20 Sep 2019 11:17:29 +0200
> 
> Even 'M-<delete>' that does something opposite to 'M-d', when <delete>
> and 'C-d' do the same thing?  Doesn't that bother you?

What does "C-h c M-<delete>" display?  Does that text answer your
questions, when you contrast it with what "C-h c" shows for <delete>
and C-<delete>?

bug#37447: 26.3; Commands on <delete> and <backspace>

[Sebastian Urban]

 > What does "C-h c M-<delete>" display?  Does that text answer your
 > questions, when you contrast it with what "C-h c" shows for <delete>
 > and C-<delete>?

C-h c M-<delete>:
     M-DEL (translated from <M-delete>) runs the command
     backward-kill-word

C-h c <delete>:
     <deletechar> (translated from <delete>) runs the command
     delete-forward-char

C-h c C-<delete>:
     <C-delete> runs the command kill-word

I'm not sure how this could help me or how I'm suppose to understand
this.

M-<delete> doesn't run command, but is translated to 'M-DEL' that does
run the command.  M-<backspace> uses the same translation.  Couldn't
it be:
- <M-delete> runs the command kill-word
- M-DEL (translated from <C-delete>) runs the command
   backward-kill-word

Or couldn't 'C-<delete>' run command backward-kill-word directly?

bug#37447: 26.3; Commands on <delete> and <backspace>

[Eli Zaretskii]

> Cc: 37447@debbugs.gnu.org
> From: Sebastian Urban <mrsebastianurban@gmail.com>
> Date: Fri, 20 Sep 2019 12:32:19 +0200
> 
>  > What does "C-h c M-<delete>" display?  Does that text answer your
>  > questions, when you contrast it with what "C-h c" shows for <delete>
>  > and C-<delete>?
> 
> C-h c M-<delete>:
>      M-DEL (translated from <M-delete>) runs the command
>      backward-kill-word
> 
> C-h c <delete>:
>      <deletechar> (translated from <delete>) runs the command
>      delete-forward-char
> 
> C-h c C-<delete>:
>      <C-delete> runs the command kill-word
> 
> I'm not sure how this could help me or how I'm suppose to understand
> this.

The DEL part should be your hint.  DEL is a key that deletes
_backward_.

> M-<delete> doesn't run command, but is translated to 'M-DEL' that does
> run the command.  M-<backspace> uses the same translation.  Couldn't
> it be:
> - <M-delete> runs the command kill-word
> - M-DEL (translated from <C-delete>) runs the command
>    backward-kill-word

See above.

bug#37447: 26.3; Commands on <delete> and <backspace>

[Sebastian Urban]

 > The DEL part should be your hint.  DEL is a key that deletes
 > _backward_.

Ok, now I'm confused even more...  I know DEL deletes backward, but in
the manual it is related to <backspace> key.  Now, I'm writing about
"Delete" key (button) that is below "Insert" and next to "End", not
about "Backspace" key.

So, to write it as simple as possible, on keyboard 'C-d' deletes char
forward, so does 'Delete' button.  'M-d' kills word forward, so I'm
expecting 'M-Delete' will do the same, but instead it kills backward
because it's translated to 'M-DEL'...

bug#29430: 26.0.90; C-M-backspace is undefined

[Devon Sean McCullough]

C-M-DEL and ESC C-DEL should both run the same command.

<C-M-backspace> is undefined

ESC <C-backspace> (translated from <escape> <C-backspace>) runs the
command backward-kill-sexp (found in global-map), which is an
interactive compiled Lisp function in ‘lisp.el’.
It is bound to ESC <C-backspace>, ESC <C-delete>.

		Peace
			--Devon

P.S. Perhaps a missing translation?

M-DEL (translated from <M-backspace>) runs the command
backward-kill-word (found in global-map), which is an interactive
compiled Lisp function in ‘simple.el’.
It is bound to <C-backspace>, M-DEL.

In GNU Emacs 26.0.90 (build 1, x86_64-apple-darwin13.4.0, NS appkit-1265.21 Version 10.9.5 (Build 13F1911))
 of 2017-10-11 built on builder10-9.local
Windowing system distributor 'Apple', version 10.3.1404
Recent messages:
For information about GNU Emacs and the GNU system, type C-h C-a.
Making completion list... [2 times]
backward-kill-sexp is on ESC <C-backspace>, ESC <C-delete>
Mark set
<C-M-backspace> is undefined
kill-region: Buffer is read-only: #<buffer *GNU Emacs*> [3 times]
Configured using:
 'configure --with-ns '--enable-locallisppath=/Library/Application
 Support/Emacs/${version}/site-lisp:/Library/Application
 Support/Emacs/site-lisp' --with-modules'

Configured features:
NOTIFY ACL GNUTLS LIBXML2 ZLIB TOOLKIT_SCROLL_BARS NS MODULES

Important settings:
  value of $LANG: en_US.UTF-8
  locale-coding-system: utf-8-unix

Major mode: Fundamental

Minor modes in effect:
  tooltip-mode: t
  global-eldoc-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
  blink-cursor-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  buffer-read-only: t
  line-number-mode: t
  transient-mark-mode: t

Load-path shadows:
None found.

Features:
(shadow sort mail-extr emacsbug message rmc puny seq byte-opt gv
bytecomp byte-compile cconv cl-loaddefs cl-lib dired dired-loaddefs
format-spec rfc822 mml easymenu mml-sec password-cache epa derived epg
epg-config gnus-util rmail rmail-loaddefs mm-decode mm-bodies mm-encode
mail-parse rfc2231 mailabbrev gmm-utils mailheader sendmail rfc2047
rfc2045 ietf-drums mm-util mail-prsvr mail-utils thingatpt elec-pair
time-date tooltip eldoc electric uniquify ediff-hook vc-hooks
lisp-float-type mwheel term/ns-win ns-win ucs-normalize mule-util
term/common-win tool-bar dnd fontset image regexp-opt fringe
tabulated-list replace newcomment text-mode elisp-mode lisp-mode
prog-mode register page menu-bar rfn-eshadow isearch timer select
scroll-bar mouse jit-lock font-lock syntax facemenu font-core
term/tty-colors frame 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 charscript charprop case-table epa-hook jka-cmpr-hook
help simple abbrev obarray minibuffer cl-preloaded nadvice loaddefs
button faces cus-face macroexp files text-properties overlay sha1 md5
base64 format env code-pages mule custom widget hashtable-print-readable
backquote kqueue cocoa ns multi-tty make-network-process emacs)

Memory information:
((conses 16 204735 11575)
 (symbols 48 20073 1)
 (miscs 40 74 193)
 (strings 32 28896 1236)
 (string-bytes 1 754349)
 (vectors 16 35060)
 (vector-slots 8 707316 10269)
 (floats 8 48 256)
 (intervals 56 230 0)
 (buffers 992 13))

bug#29430: 26.0.90; C-M-backspace is undefined

[Noam Postavsky]

On Fri, Nov 24, 2017 at 2:12 PM, Devon Sean McCullough
<Emacs-Hacker2017@jovi.net> wrote:
> C-M-DEL and ESC C-DEL should both run the same command.
>
> <C-M-backspace> is undefined
>
> ESC <C-backspace> (translated from <escape> <C-backspace>) runs the
> command backward-kill-sexp (found in global-map), which is an
> interactive compiled Lisp function in ‘lisp.el’.
> It is bound to ESC <C-backspace>, ESC <C-delete>.
>
>                 Peace
>                         --Devon
>
> P.S. Perhaps a missing translation?

Seems to be on purpose, bindings.el:1236:

;;; These are dangerous in various situations,
;;; so let's not encourage anyone to use them.
;;;(define-key global-map [C-M-delete] 'backward-kill-sexp)
;;;(define-key global-map [C-M-backspace] 'backward-kill-sexp)

bug#37447: 26.3; Commands on <delete> and <backspace>

[Lars Ingebrigtsen]

Noam Postavsky <npostavs@users.sourceforge.net> writes:

> Seems to be on purpose, bindings.el:1236:
>
> ;;; These are dangerous in various situations,
> ;;; so let's not encourage anyone to use them.
> ;;;(define-key global-map [C-M-delete] 'backward-kill-sexp)
> ;;;(define-key global-map [C-M-backspace] 'backward-kill-sexp)

I've now reinstated the bindings, since X stopped killing itself if you
hit those keys decades ago (in all mainstream configurations).

The merged bug report discussed moving some of other related keys
around, and the conclusions was that we don't want to do that, so I'm
closing this bug report.

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

bug#37447: 26.3; Commands on <delete> and <backspace>

[Drew Adams]

> > ;;; These are dangerous in various situations,
> > ;;; so let's not encourage anyone to use them.
> > ;;;(define-key global-map [C-M-delete] 'backward-kill-sexp)
> > ;;;(define-key global-map [C-M-backspace] 'backward-kill-sexp)
> 
> I've now reinstated the bindings, since X stopped killing itself if you
> hit those keys decades ago (in all mainstream configurations).

[C-M-delete] is problematic for MS Windows, no?