bug#60741: 29.0.60; 'M-c' shadowed in the minibuffer

bug#60741: 29.0.60; 'M-c' shadowed in the minibuffer

[Juri Linkov]

etc/NEWS:

  +++
  ** 'M-c' in 'read-regexp' now toggles case folding.

'M-c' is bound globally to 'capitalize-word', but this keybinding
is shadowed in the minibuffer, thus making it impossible to
capitalize words.

Isearch had this problem for a long time but now it's already solved with:

  (put 'isearch-toggle-case-fold :advertised-binding "\M-sc")

where 'M-s c' is used to toggle search case-sensitivity.

So to avoid the problem the same could be done in 'read-regexp':

```
diff --git a/lisp/replace.el b/lisp/replace.el
index 2f063bbf66b..9471435bda3 100644
--- a/lisp/replace.el
+++ b/lisp/replace.el
@@ -824,7 +830,7 @@ read-regexp-suggestions
 
 (defvar-keymap read-regexp-map
   :parent minibuffer-local-map
-  "M-c" #'read-regexp-toggle-case-folding)
+  "M-s c" #'read-regexp-toggle-case-folding)
 
 (defvar read-regexp--case-fold nil)
 
```

bug#60741: 29.0.60; 'M-c' shadowed in the minibuffer

[Juri Linkov]

close 60741 29.0.60
quit

> 'M-c' is bound globally to 'capitalize-word', but this keybinding
> is shadowed in the minibuffer, thus making it impossible to
> capitalize words.
>
> Isearch had this problem for a long time but now it's already solved with:
>
>   (put 'isearch-toggle-case-fold :advertised-binding "\M-sc")
>
> where 'M-s c' is used to toggle search case-sensitivity.
>
> So to avoid the problem the same could be done in 'read-regexp':

Pushed to make compatible with the search map currently
discussed in bug#60815 where it will be unified in master
among other M-s keys under the general setting search-key.

bug#60741: 29.0.60; 'M-c' shadowed in the minibuffer

[Drew Adams]

> > 'M-c' is bound globally to 'capitalize-word', but this keybinding
> > is shadowed in the minibuffer, thus making it impossible to
> > capitalize words.

Is this something new in Emacs 29?  I don't see it
in previous versions (I don't have 29).  What is
`M-c' bound to now in the minibuffer keymaps?

Why should we lose this longstanding, useful key
binding?  Users should be able to use most normal
editing keys in the minibuffer - it's first and
foremost an editing buffer.

bug#60741: 29.0.60; 'M-c' shadowed in the minibuffer

[Juri Linkov]

>> > 'M-c' is bound globally to 'capitalize-word', but this keybinding
>> > is shadowed in the minibuffer, thus making it impossible to
>> > capitalize words.
>
> Is this something new in Emacs 29?

Yes, it's new.

> I don't see it in previous versions (I don't have 29).
> What is `M-c' bound to now in the minibuffer keymaps?

It was bound to `read-regexp-toggle-case-folding'.

> Why should we lose this longstanding, useful key
> binding?  Users should be able to use most normal
> editing keys in the minibuffer - it's first and
> foremost an editing buffer.

Now normal editing keys are available again.

bug#60741: 29.0.60; 'M-c' shadowed in the minibuffer

[Drew Adams]

> Now normal editing keys are available again.

Thx.

bug#60741: 29.0.60; 'M-c' shadowed in the minibuffer

[Jean Louis]

* Drew Adams <drew.adams@oracle.com> [2023-01-18 21:48]:
> > > 'M-c' is bound globally to 'capitalize-word', but this keybinding
> > > is shadowed in the minibuffer, thus making it impossible to
> > > capitalize words.
> 
> Is this something new in Emacs 29?  I don't see it
> in previous versions (I don't have 29).  What is
> `M-c' bound to now in the minibuffer keymaps?
> 
> Why should we lose this longstanding, useful key
> binding?  Users should be able to use most normal
> editing keys in the minibuffer - it's first and
> foremost an editing buffer.

I use that key all the time in the minibuffer, and can't without
it. I give you the practical example:

- PDF book in Dired is called:
  21252030 agenda for sustainable development web.pdf

- I click and wish to index the book, so initial prompt for the title
  will be:
  21252030 agenda for sustainable development web.pdf

- I delete 21252030 with M-d and use M-c to get following:
  Agenda For Sustainable Development

- then I press RET to record information

Without M-c in minibuffer I would need to re-assign key binding.

But it works, I did not know something changed.

--
Jean

Take action in Free Software Foundation campaigns:
https://www.fsf.org/campaigns

In support of Richard M. Stallman
https://stallmansupport.org/

bug#60741: 29.0.60; 'M-c' shadowed in the minibuffer

[Eli Zaretskii]

> Cc: "60741@debbugs.gnu.org" <60741@debbugs.gnu.org>,
>  Juri Linkov <juri@linkov.net>
> Date: Thu, 19 Jan 2023 08:34:05 +0300
> From: Jean Louis <bugs@gnu.support>
> 
> * Drew Adams <drew.adams@oracle.com> [2023-01-18 21:48]:
> > > > 'M-c' is bound globally to 'capitalize-word', but this keybinding
> > > > is shadowed in the minibuffer, thus making it impossible to
> > > > capitalize words.
> > 
> > Is this something new in Emacs 29?  I don't see it
> > in previous versions (I don't have 29).  What is
> > `M-c' bound to now in the minibuffer keymaps?
> > 
> > Why should we lose this longstanding, useful key
> > binding?  Users should be able to use most normal
> > editing keys in the minibuffer - it's first and
> > foremost an editing buffer.
> 
> I use that key all the time in the minibuffer, and can't without
> it. I give you the practical example:
> 
> - PDF book in Dired is called:
>   21252030 agenda for sustainable development web.pdf
> 
> - I click and wish to index the book, so initial prompt for the title
>   will be:
>   21252030 agenda for sustainable development web.pdf
> 
> - I delete 21252030 with M-d and use M-c to get following:
>   Agenda For Sustainable Development
> 
> - then I press RET to record information
> 
> Without M-c in minibuffer I would need to re-assign key binding.
> 
> But it works, I did not know something changed.

You are barking up the wrong tree: Juri has _fixed_ this, didn't
_break_ this.

Please review the changeset before posting opinions about it.