GNU bug report logs - #75933
29; Disabling smerge-command-prefix results in error

Previous Next

To reply to this bug, email your comments to 75933 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: Jonas Bernoulli <jonas <at> bernoul.li>
To: bug-gnu-emacs <at> gnu.org
Subject: 29; Disabling smerge-command-prefix results in error
Date: Wed, 29 Jan 2025 20:28:32 +0100

Hello,

In Emacs 28 the prefix key was defined like this:

  (defcustom smerge-command-prefix "\C-c^"
    "Prefix for `smerge-mode' commands."
    :type '(choice (const :tag "ESC"   "\e")
                   (const :tag "C-c ^" "\C-c^" )
                   (const :tag "none"  "")
                   string))

  (easy-mmode-defmap smerge-mode-map
    `((,smerge-command-prefix . ,smerge-basic-map))
    "Keymap for `smerge-mode'.")

In Emacs 29.1 (after 2ae3b66fa80 and some fixup commits), the former
remained as it was in 28, but the latter was replaced with:

  (defvar-keymap smerge-mode-map
    (key-description smerge-command-prefix) smerge-basic-map)

Setting the option to "", to add no prefix command, results in:

  keymap--check: "" is not a valid key definition; see `key-valid-p'

A possible fix could be to do it old-school:

  (defvar smerge-mode-map
    (let ((map (make-sparse-keymap))
          (key (key-description smerge-command-prefix)))
      (when (key-valid-p key)
        (keymap-set map key smerge-basic-map))
      map))

Best regards,
Jonas

Message #8 received at 75933 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Jonas Bernoulli <jonas <at> bernoul.li>,
 Stefan Monnier <monnier <at> iro.umontreal.ca>
Cc: 75933 <at> debbugs.gnu.org
Subject: Re: bug#75933: 29; Disabling smerge-command-prefix results in error
Date: Wed, 29 Jan 2025 21:53:29 +0200

> Date: Wed, 29 Jan 2025 20:28:32 +0100
> From:  Jonas Bernoulli via "Bug reports for GNU Emacs,
>  the Swiss army knife of text editors" <bug-gnu-emacs <at> gnu.org>
> 
> Hello,
> 
> In Emacs 28 the prefix key was defined like this:
> 
>   (defcustom smerge-command-prefix "\C-c^"
>     "Prefix for `smerge-mode' commands."
>     :type '(choice (const :tag "ESC"   "\e")
>                    (const :tag "C-c ^" "\C-c^" )
>                    (const :tag "none"  "")
>                    string))
> 
>   (easy-mmode-defmap smerge-mode-map
>     `((,smerge-command-prefix . ,smerge-basic-map))
>     "Keymap for `smerge-mode'.")
> 
> In Emacs 29.1 (after 2ae3b66fa80 and some fixup commits), the former
> remained as it was in 28, but the latter was replaced with:
> 
>   (defvar-keymap smerge-mode-map
>     (key-description smerge-command-prefix) smerge-basic-map)
> 
> Setting the option to "", to add no prefix command, results in:
> 
>   keymap--check: "" is not a valid key definition; see `key-valid-p'
> 
> A possible fix could be to do it old-school:
> 
>   (defvar smerge-mode-map
>     (let ((map (make-sparse-keymap))
>           (key (key-description smerge-command-prefix)))
>       (when (key-valid-p key)
>         (keymap-set map key smerge-basic-map))
>       map))

Thanks.

Stefan, can you suggest a solution?

Message #11 received at 75933 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Monnier <monnier <at> iro.umontreal.ca>
To: Jonas Bernoulli <jonas <at> bernoul.li>
Cc: 75933 <at> debbugs.gnu.org
Subject: Re: bug#75933: 29; Disabling smerge-command-prefix results in error
Date: Wed, 05 Feb 2025 18:08:26 -0500

>   (easy-mmode-defmap smerge-mode-map
>     `((,smerge-command-prefix . ,smerge-basic-map))
>     "Keymap for `smerge-mode'.")
>
> In Emacs 29.1 (after 2ae3b66fa80 and some fixup commits), the former
> remained as it was in 28, but the latter was replaced with:
>
>   (defvar-keymap smerge-mode-map
>     (key-description smerge-command-prefix) smerge-basic-map)
>
> Setting the option to "", to add no prefix command, results in:
>
>   keymap--check: "" is not a valid key definition; see `key-valid-p'

Emacs Emacs<29 setting the option to "" resulted in keeping
`smerge-mode-map` empty, AFAICT.  Is that what you wanted or did you
expect all the bindings of `smerge-basic-map` to be made available
without a prefix?

To my mind the value "" would be closer to meaning that all the bindings
of `smerge-basic-map` should be made available without a prefix, and if
we want to allow the bindings to be completely absent, it might make
more sense to use a nil value.


        Stefan

Message #14 received at 75933 <at> debbugs.gnu.org (full text, mbox):

From: Jonas Bernoulli <jonas <at> bernoul.li>
To: Stefan Monnier <monnier <at> iro.umontreal.ca>
Cc: 75933 <at> debbugs.gnu.org
Subject: Re: bug#75933: 29; Disabling smerge-command-prefix results in error
Date: Thu, 06 Feb 2025 14:50:04 +0100

Stefan Monnier <monnier <at> iro.umontreal.ca> writes:

>>   (easy-mmode-defmap smerge-mode-map
>>     `((,smerge-command-prefix . ,smerge-basic-map))
>>     "Keymap for `smerge-mode'.")
>>
>> In Emacs 29.1 (after 2ae3b66fa80 and some fixup commits), the former
>> remained as it was in 28, but the latter was replaced with:
>>
>>   (defvar-keymap smerge-mode-map
>>     (key-description smerge-command-prefix) smerge-basic-map)
>>
>> Setting the option to "", to add no prefix command, results in:
>>
>>   keymap--check: "" is not a valid key definition; see `key-valid-p'
>
> Emacs Emacs<29 setting the option to "" resulted in keeping
> `smerge-mode-map` empty, AFAICT.  Is that what you wanted or did you
> expect all the bindings of `smerge-basic-map` to be made available
> without a prefix?

Well *I* am perfectly happy with the defaults.  But there's at least one
Magit user who never uses Smerge and thus doesn't want keymaps cluttered
with its bindings.

> To my mind the value "" would be closer to meaning that all the bindings
> of `smerge-basic-map` should be made available without a prefix, and if
> we want to allow the bindings to be completely absent, it might make
> more sense to use a nil value.

That does make sense to me too, but I don't think that users who have
set this to "", to not even use a top-level binding for Smerge commands,
would be happy if the meaning were changed to add *multiple* top-level
bindings, that's worse for them than the default of one prefix keymap.
So it might be better to use two new symbols, such as 'top-level and
'none, and make "" a backward compatibility alias for 'none.

Message #17 received at 75933 <at> debbugs.gnu.org (full text, mbox):

From: Eli Zaretskii <eliz <at> gnu.org>
To: Jonas Bernoulli <jonas <at> bernoul.li>
Cc: 75933 <at> debbugs.gnu.org, monnier <at> iro.umontreal.ca
Subject: Re: bug#75933: 29; Disabling smerge-command-prefix results in error
Date: Sat, 22 Feb 2025 11:16:14 +0200

Ping! How should we proceed with this issue?

> Cc: 75933 <at> debbugs.gnu.org
> Date: Thu, 06 Feb 2025 14:50:04 +0100
> From:  Jonas Bernoulli via "Bug reports for GNU Emacs,
>  the Swiss army knife of text editors" <bug-gnu-emacs <at> gnu.org>
> 
> Stefan Monnier <monnier <at> iro.umontreal.ca> writes:
> 
> >>   (easy-mmode-defmap smerge-mode-map
> >>     `((,smerge-command-prefix . ,smerge-basic-map))
> >>     "Keymap for `smerge-mode'.")
> >>
> >> In Emacs 29.1 (after 2ae3b66fa80 and some fixup commits), the former
> >> remained as it was in 28, but the latter was replaced with:
> >>
> >>   (defvar-keymap smerge-mode-map
> >>     (key-description smerge-command-prefix) smerge-basic-map)
> >>
> >> Setting the option to "", to add no prefix command, results in:
> >>
> >>   keymap--check: "" is not a valid key definition; see `key-valid-p'
> >
> > Emacs Emacs<29 setting the option to "" resulted in keeping
> > `smerge-mode-map` empty, AFAICT.  Is that what you wanted or did you
> > expect all the bindings of `smerge-basic-map` to be made available
> > without a prefix?
> 
> Well *I* am perfectly happy with the defaults.  But there's at least one
> Magit user who never uses Smerge and thus doesn't want keymaps cluttered
> with its bindings.
> 
> > To my mind the value "" would be closer to meaning that all the bindings
> > of `smerge-basic-map` should be made available without a prefix, and if
> > we want to allow the bindings to be completely absent, it might make
> > more sense to use a nil value.
> 
> That does make sense to me too, but I don't think that users who have
> set this to "", to not even use a top-level binding for Smerge commands,
> would be happy if the meaning were changed to add *multiple* top-level
> bindings, that's worse for them than the default of one prefix keymap.
> So it might be better to use two new symbols, such as 'top-level and
> 'none, and make "" a backward compatibility alias for 'none.
> 
> 
> 
> 

Message #20 received at 75933 <at> debbugs.gnu.org (full text, mbox):

From: Stefan Monnier <monnier <at> iro.umontreal.ca>
To: Eli Zaretskii <eliz <at> gnu.org>
Cc: Jonas Bernoulli <jonas <at> bernoul.li>, 75933 <at> debbugs.gnu.org
Subject: Re: bug#75933: 29; Disabling smerge-command-prefix results in error
Date: Sat, 22 Feb 2025 11:23:41 -0500

> Ping! How should we proceed with this issue?

Maybe another way to look at it is that it's again bumping into
`define-keymap`s limitation that you can't easily "deactivate"
a binding.

We saw that back in the bug#68213: if a keymap is defined where some of
the key sequences are chosen dynamically (typically coming from a config
var), we tend to get into trouble where it can be difficult/inconvenient
to avoid duplicate key bindings (we fixed that problem by demoting that
error to a warning), and similarly there's no easy way for the user to
say "no key at all" (which was another way we could have fixed the
problem in bug#68213).

Maybe we should have an "impossible key sequence" (like we have the
`regexp-unmatchable`) and offer it in the customization interface?
And then arrange for `define-keymap` (and maybe others) to handle
it accordingly.

Not sure what that key sequence should be, but some choices that come to
mind would be:

- nil
- [] (a.k.a "")
- [impossible]


        Stefan

>> Cc: 75933 <at> debbugs.gnu.org
>> Date: Thu, 06 Feb 2025 14:50:04 +0100
>> From:  Jonas Bernoulli via "Bug reports for GNU Emacs,
>>  the Swiss army knife of text editors" <bug-gnu-emacs <at> gnu.org>
>> 
>> Stefan Monnier <monnier <at> iro.umontreal.ca> writes:
>> 
>> >>   (easy-mmode-defmap smerge-mode-map
>> >>     `((,smerge-command-prefix . ,smerge-basic-map))
>> >>     "Keymap for `smerge-mode'.")
>> >>
>> >> In Emacs 29.1 (after 2ae3b66fa80 and some fixup commits), the former
>> >> remained as it was in 28, but the latter was replaced with:
>> >>
>> >>   (defvar-keymap smerge-mode-map
>> >>     (key-description smerge-command-prefix) smerge-basic-map)
>> >>
>> >> Setting the option to "", to add no prefix command, results in:
>> >>
>> >>   keymap--check: "" is not a valid key definition; see `key-valid-p'
>> >
>> > Emacs Emacs<29 setting the option to "" resulted in keeping
>> > `smerge-mode-map` empty, AFAICT.  Is that what you wanted or did you
>> > expect all the bindings of `smerge-basic-map` to be made available
>> > without a prefix?
>> 
>> Well *I* am perfectly happy with the defaults.  But there's at least one
>> Magit user who never uses Smerge and thus doesn't want keymaps cluttered
>> with its bindings.
>> 
>> > To my mind the value "" would be closer to meaning that all the bindings
>> > of `smerge-basic-map` should be made available without a prefix, and if
>> > we want to allow the bindings to be completely absent, it might make
>> > more sense to use a nil value.
>> 
>> That does make sense to me too, but I don't think that users who have
>> set this to "", to not even use a top-level binding for Smerge commands,
>> would be happy if the meaning were changed to add *multiple* top-level
>> bindings, that's worse for them than the default of one prefix keymap.
>> So it might be better to use two new symbols, such as 'top-level and
>> 'none, and make "" a backward compatibility alias for 'none.
>> 
>> 
>> 
>> 

Message #23 received at 75933 <at> debbugs.gnu.org (full text, mbox):

From: Daniel Colascione <dancol <at> dancol.org>
To: Stefan Monnier via "Bug reports for GNU Emacs, the Swiss army knife of
 text editors" <bug-gnu-emacs <at> gnu.org>
Cc: Eli Zaretskii <eliz <at> gnu.org>, Jonas Bernoulli <jonas <at> bernoul.li>,
 75933 <at> debbugs.gnu.org, Stefan Monnier <monnier <at> iro.umontreal.ca>
Subject: Re: bug#75933: 29; Disabling smerge-command-prefix results in error
Date: Sat, 22 Feb 2025 18:10:36 -0500

Stefan Monnier via "Bug reports for GNU Emacs, the Swiss army knife of text editors" <bug-gnu-emacs <at> gnu.org> writes:

>> Ping! How should we proceed with this issue?
>
> Maybe another way to look at it is that it's again bumping into
> `define-keymap`s limitation that you can't easily "deactivate"
> a binding.
>
> We saw that back in the bug#68213: if a keymap is defined where some of
> the key sequences are chosen dynamically (typically coming from a config
> var), we tend to get into trouble where it can be difficult/inconvenient
> to avoid duplicate key bindings (we fixed that problem by demoting that
> error to a warning), and similarly there's no easy way for the user to
> say "no key at all" (which was another way we could have fixed the
> problem in bug#68213).
>
> Maybe we should have an "impossible key sequence" (like we have the
> `regexp-unmatchable`) and offer it in the customization interface?

The question is ill-formed. Why do we have keymaps with bindings based
on random defcustoms in the first place? The proper place to customize a
keymap is the keymap. 

Previous Next

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