bug#73954: 30.0.91; Don't hard-couple `dired-hide-details-hide-absolute-location' with `dired-hide-details-mode'

bug#73954: 30.0.91; Don't hard-couple `dired-hide-details-hide-absolute-location' with `dired-hide-details-mode'

[Drew Adams via Bug reports for GNU Emacs, the Swiss army knife of text editors]

Starting with Emacs 30 Dired directory header lines apparently show
relative names when `dired-hide-details-mode' is on, if option
`dired-hide-details-hide-absolute-location' is non-nil.

(The only Windows v30 pretest snapshot I found doesn't have this option,
so I'm going only by the description from bug #72272.)

The behavior of being able to show only a relative name is a good one,
but it shouldn't be hard-coupled with `dired-hide-details-mode'.  Users
should be able to choose:

 nil                         Always show absolute name.
                             (This needs to stay the default.) 
 if-dired-hide-details-mode  Show relative name only when
                             `dired-hide-details-mode' is on.
 t                           Always show relative name.
 FUNCTION                    Show relative name only when invoking
                             FUNCTION with no args returns non-nil.

The most important behaviors are for nil, t, and
`if-hiding-detail-columns'.  FUNCTION is a nice-to-have.

(An alternative would be to use t for showing relative name when
`dired-hide-details-mode' is on and `always' for showing relative names
always.  That keeps the current pre-release behavior for t.  I don't
think it's preferable, but it's a possibility.)


In GNU Emacs 30.0.91 (build 2, x86_64-w64-mingw32) of 2024-09-11 built
 on AVALON
Windowing system distributor 'Microsoft Corp.', version 10.0.19045
System Description: Microsoft Windows 10 Pro (v10.0.2009.19045.5011)

Configured using:
 'configure --with-modules --without-dbus --with-native-compilation=aot
 --without-compress-install --with-tree-sitter
 --enable-checking=yes,glyphs 'CFLAGS=-O0 -g3''

Configured features:
ACL GIF GMP GNUTLS HARFBUZZ JPEG LCMS2 LIBXML2 MODULES NATIVE_COMP
NOTIFY W32NOTIFY PDUMPER PNG RSVG SOUND SQLITE3 THREADS TIFF
TOOLKIT_SCROLL_BARS TREE_SITTER WEBP XPM ZLIB

(NATIVE_COMP present but libgccjit not available)

Important settings:
  value of $LANG: ENU
  locale-coding-system: cp1252

bug#73954: 30.0.91; Don't hard-couple `dired-hide-details-hide-absolute-location' with `dired-hide-details-mode'

[Eli Zaretskii]

severity 73954 wishlist
thanks

> Date: Tue, 22 Oct 2024 18:34:25 +0000
> From:  Drew Adams via "Bug reports for GNU Emacs,
>  the Swiss army knife of text editors" <bug-gnu-emacs@gnu.org>
> 
> Starting with Emacs 30 Dired directory header lines apparently show
> relative names when `dired-hide-details-mode' is on, if option
> `dired-hide-details-hide-absolute-location' is non-nil.
> 
> (The only Windows v30 pretest snapshot I found doesn't have this option,
> so I'm going only by the description from bug #72272.)

You are not seeing it in Emacs 30 for a good reason: it was added for
Emacs 31.

> The behavior of being able to show only a relative name is a good one,
> but it shouldn't be hard-coupled with `dired-hide-details-mode'.  Users
> should be able to choose:

"Should" why? so we have yet another hard-to-remember and never-used
combination of options?

We just added this feature to the development branch.  Let's first
collect some experience with it and see if it is popular enough to
extend it in any way.  At this point, it is as possible that the only
person to use it is the person who contributed it as it is possible
that it will be the hottest new Emacs feature, so let's wait and see
which is it.

bug#73954: 30.0.91; Don't hard-couple `dired-hide-details-hide-absolute-location' with `dired-hide-details-mode'

[Drew Adams via Bug reports for GNU Emacs, the Swiss army knife of text editors]

> We just added this feature to the development branch.  Let's first
> collect some experience with it and see if it is popular enough to
> extend it in any way.  At this point, it is as possible that the only
> person to use it is the person who contributed it as it is possible
> that it will be the hottest new Emacs feature, so let's wait and see
> which is it.

If you put it out there as it is, it'll be used as it is.

What's the connection between wanting relative or absolute
names for directory headers and wanting to hide columns
other than the file name?  There's only a connection if
you happen to want to do them together.

bug#73954: 30.0.91; Don't hard-couple `dired-hide-details-hide-absolute-location' with `dired-hide-details-mode'

[Eli Zaretskii]

> From: Drew Adams <drew.adams@oracle.com>
> CC: "73954@debbugs.gnu.org" <73954@debbugs.gnu.org>
> Date: Tue, 22 Oct 2024 19:09:57 +0000
> 
> > We just added this feature to the development branch.  Let's first
> > collect some experience with it and see if it is popular enough to
> > extend it in any way.  At this point, it is as possible that the only
> > person to use it is the person who contributed it as it is possible
> > that it will be the hottest new Emacs feature, so let's wait and see
> > which is it.
> 
> If you put it out there as it is, it'll be used as it is.
> 
> What's the connection between wanting relative or absolute
> names for directory headers and wanting to hide columns
> other than the file name?

The connection is the "hide" part.  People who are likely to want to
hide the leading directories are those who use
dired-hide-details-mode.  To break the connection, we'd need evidence,
not theoretical possibilities.

bug#73954: 30.0.91; Don't hard-couple `dired-hide-details-hide-absolute-location' with `dired-hide-details-mode'

[Drew Adams via Bug reports for GNU Emacs, the Swiss army knife of text editors]

> > > We just added this feature to the development branch.  Let's first
> > > collect some experience with it and see if it is popular enough to
> > > extend it in any way.  At this point, it is as possible that the
> only
> > > person to use it is the person who contributed it as it is possible
> > > that it will be the hottest new Emacs feature, so let's wait and see
> > > which is it.
> >
> > If you put it out there as it is, it'll be used as it is.
> >
> > What's the connection between wanting relative or absolute
> > names for directory headers and wanting to hide columns
> > other than the file name?
> 
> The connection is the "hide" part.  People who are likely to want to
> hide the leading directories are those who use
> dired-hide-details-mode.  To break the connection, we'd need evidence,
> not theoretical possibilities.

That's the foot in the door, if you allow it.

There are several things in Dired that can be
hidden in one way or another.  So far, they're
NOT all enabled willy nilly whenever you turn
on `dired-hide-details-mode' - fortunately.

You can hide subdirs (`$', `dired-kill-subdir',
`M-$'), lines (`dired-kill-line'), whole trees
(`dired-kill-tree'), uninteresting files
(`dired-omit-mode'), image thumbnails
(`C-t C-t'), etc.  And in Dired enhancement
libraries there are more hiding possibilities. 

Instead of requiring evidence to "break the
connection" - which is a brand new, _proposed_
connection, you ought to require evidence for
why this particular proposed hiding should be
connected to `dired-hide-details-mode'.

Just putting "hide" in the name of something
doesn't, logically, connect it to something
else that has "hide" in its name.  I'm quite
surprised to hear such a rationale from you,
frankly.