bug#15799: 24.3.50; dired-hide-details-mode: Provide default key binding

bug#15799: 24.3.50; dired-hide-details-mode: Provide default key binding

[Jambunathan K]

All my dired buffers, hide the details by default (See below).  I
propose "@" as a key for toggling the details.

I suggest 

    (defun dired-mode-hook ()
      (dired-hide-details-mode 1)
      (local-set-key "@" 'dired-hide-details-mode))


In GNU Emacs 24.3.50.5 (i686-pc-linux-gnu, GTK+ Version 2.20.1)
 of 2013-11-04 on debian-6.05
Bzr revision: 114933 monnier@iro.umontreal.ca-20131104042959-sy10svhxghg2z9ei
Windowing system distributor `The X.Org Foundation', version 11.0.10707000
System Description:	Debian GNU/Linux 6.0.5 (squeeze)

bug#15799: 24.3.50; dired-hide-details-mode: Provide default key binding

[Stefan Monnier]

severity 15799 wishlist
thanks

> All my dired buffers, hide the details by default (See below).  I
> propose "@" as a key for toggling the details.

I wouldn't mind, but since I don't myself use dired much, I'd like to
hear what actual users think about it.  Also, wouldn't it make sense to
try and bring hide-details "closer" to the dired-omit facilities?


        Stefan

bug#15799: 24.3.50; dired-hide-details-mode: Provide default key binding

[Drew Adams]

> > All my dired buffers, hide the details by default (See below).  I
> > propose "@" as a key for toggling the details.

See bug #14940, which provides a patch with a much better solution.

> I wouldn't mind, but since I don't myself use dired much, I'd like
> to hear what actual users think about it.

See bug #14940.  See also bug #14914.

See also Dired+
http://www.emacswiki.org/emacs-en/download/dired%2b.el

> Also, wouldn't it make sense to try and bring hide-details
> "closer" to the dired-omit facilities?

Definitely not.  No relation.  Completely different uses.

bug#15799: 24.3.50; dired-hide-details-mode: Provide default key binding

[Stefan Monnier]

> See bug #14940.  See also bug #14914.

Thanks.  For #14914, there's no doubt for me that we should use
a "[✓] Hide details" menu entry.

>> Also, wouldn't it make sense to try and bring hide-details
>> "closer" to the dired-omit facilities?
> Definitely not.  No relation.  Completely different uses.

Of course, they're different, but I assume the @ was chosen because it's
also used for things like outline-minor-mode, i.e. for other features
that hide parts of the buffer.

So, maybe @ should be a prefix key for "hiding stuff", in which we can
place key bindings for "omit" and for "details".


        Stefan

bug#14914: bug#15799: 24.3.50; dired-hide-details-mode: Provide default key binding

[Drew Adams]

> Thanks.  For #14914, there's no doubt for me that we should use
> a "[✓] Hide details" menu entry.

Sounds good to me (except it should be title case: `Hide Details').

> >> Also, wouldn't it make sense to try and bring hide-details
> >> "closer" to the dired-omit facilities?
> >
> > Definitely not.  No relation.  Completely different uses.
> 
> Of course, they're different, but I assume the @ was chosen because
> it's also used for things like outline-minor-mode, i.e. for other
> features that hide parts of the buffer.

`@' was proposed by Jambunathan, not I.  I have no special objection
to `@' instead of `(', however.

But I'm not convinced that it is good to confuse hiding stuff in
Dired with outline-minor-mode hiding.

> So, maybe @ should be a prefix key for "hiding stuff", in which we
> can place key bindings for "omit" and for "details".

I don't object to that.  As long as the two, hiding columns and
hiding rows, are kept clear and separate, no problem.

But there is really more to it than that.  There are at least these
hide/show commands in Dired:

$    (`dired-hide-subdir')
M-$  (`dired-hide-all')
(    (`dired-hide-details-mode')
M-o  (`dired-omit-mode')

It is good that each of these is quick, but yes, we could put them
all on a prefix key.  In that case, `M-o', `M-$', and `$' are also
good candidates for the prefix key.

We could have, for example:

M-o s  (`dired-hide-subdir')       - `s' for subdi
M-o S  (`dired-hide-all')          - `S' for subdirs
M-o c  (`dired-hide-details-mode') - `c' for columns
M-o r  (`dired-omit-mode')         - `r' for rows

No Shift key needed for them, except for the one that boosts `s'.

FWIW, Dired+ adds these commands, which could also be added to the
list, perhaps with these bindings:

M-o m  (`diredp-omit-marked')      - `m' for marked
M-o u  (`diredp-omit-unmarked')    - `u' for unmarked

Here are their (trivial) definitions:

(defun diredp-omit-marked ()
  "Omit lines of marked files.  Return the number of lines omitted."
  (interactive)
  (let ((old-modified-p  (buffer-modified-p))
        count)
    (when (interactive-p) (message "Omitting marked lines..."))
    (setq count  (dired-do-kill-lines nil "Omitted %d line%s."))
    (set-buffer-modified-p old-modified-p) ; So no `%*' appear in mode-line.
    count))

(defun diredp-omit-unmarked ()
  "Omit lines of unmarked files.  Return the number of lines omitted."
  (interactive)
  (let ((old-modified-p  (buffer-modified-p))
        count)
    (dired-toggle-marks)
    (message "Omitting unmarked lines...")
    (setq count  (diredp-omit-marked))
    (dired-toggle-marks)                ; Marks all except `.', `..'
    (set-buffer-modified-p old-modified-p) ; So no `%*' appear in mode-line.
    count))

bug#14914: bug#15799: 24.3.50; dired-hide-details-mode: Provide default key binding

[Jambunathan K]

>> Of course, they're different, but I assume the @ was chosen because
>> it's also used for things like outline-minor-mode, i.e. for other
>> features that hide parts of the buffer.
>
> `@' was proposed by Jambunathan, not I.  I have no special objection
> to `@' instead of `(', however.

I do object to '('.  

Partly because I am tempted to think ')' will undo whatever it's other
pair did.

> But there is really more to it than that.  There are at least these
> hide/show commands in Dired:
>
> $    (`dired-hide-subdir')
> M-$  (`dired-hide-all')
> (    (`dired-hide-details-mode')
> M-o  (`dired-omit-mode')
>
> It is good that each of these is quick, but yes, we could put them
> all on a prefix key.  In that case, `M-o', `M-$', and `$' are also
> good candidates for the prefix key.

$ is indeed better than @.  Remember `$' is used in conjunction with
selective display.  

> We could have, for example:
>
> M-o s  (`dired-hide-subdir')       - `s' for subdi
> M-o S  (`dired-hide-all')          - `S' for subdirs
> M-o c  (`dired-hide-details-mode') - `c' for columns
> M-o r  (`dired-omit-mode')         - `r' for rows

I object to M-o.  The verb `Omit' is generally not used.  It has bad
connotations.  (My mom will spank me if I omitted something)

Let's omit omit and stick with show or hide.

bug#15799: 24.3.50; dired-hide-details-mode: Provide default key binding

[Jambunathan K]

OP here.  Closed.

bug#14914: bug#15799: 24.3.50; dired-hide-details-mode: Provide default key binding

[Stefan Kangas]

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

>> Thanks.  For #14914, there's no doubt for me that we should use
>> a "[✓] Hide details" menu entry.
>
> Sounds good to me (except it should be title case: `Hide Details').

This is now the case on current master, so I'm closing this bug.