Disappearance of hi-lock bindings from global key map.

Disappearance of hi-lock bindings from global key map.

[Alan Mackenzie]

Hello, Emacs.

The traditional key bindings for hi-lock commands, beginning C-x w, have
disappeared from the 25.1 global key map.  They have been replaced by
bindings beginning M-s h.

There is no mention in NEWS of this disappearance (although the
replacement bindings were mentioned in 24's NEWS).  Is this deliberate,
or is it an oversight?

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Disappearance of hi-lock bindings from global key map.

[Eli Zaretskii]

> Date: Fri, 5 Feb 2016 10:37:44 +0000
> From: Alan Mackenzie <acm@muc.de>
> 
> The traditional key bindings for hi-lock commands, beginning C-x w, have
> disappeared from the 25.1 global key map.  They have been replaced by
> bindings beginning M-s h.

That's not what I see here: when I turn on hi-lock-mode, the "C-x w"
bindings are shown in the output of "C-h b".

What exactly do you see that led you to the above conclusion?

Re: Disappearance of hi-lock bindings from global key map.

[Alan Mackenzie]

Hello, Eli.

On Fri, Feb 05, 2016 at 01:19:13PM +0200, Eli Zaretskii wrote:
> > Date: Fri, 5 Feb 2016 10:37:44 +0000
> > From: Alan Mackenzie <acm@muc.de>
> > 
> > The traditional key bindings for hi-lock commands, beginning C-x w, have
> > disappeared from the 25.1 global key map.  They have been replaced by
> > bindings beginning M-s h.

> That's not what I see here: when I turn on hi-lock-mode, the "C-x w"
> bindings are shown in the output of "C-h b".

> What exactly do you see that led you to the above conclusion?

emacs -Q

C-x w  (intending C-x w C-h) responds "C-x w is undefined".

On the other hand,
M-s h C-h displays the bindings in *Help*.

That led me to believe the C-x h bindings had been removed.  It would
appear there's a difference between how the C-x w bindings are set and
how the M-s h bindings are set.   

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Disappearance of hi-lock bindings from global key map.

[Kaushal Modi]

[-- Attachment #1: Type: text/plain, Size: 1452 bytes --]

I might not be understanding this correctly.... But based on the below
code, wouldn't we first need to do (require 'hi-lock) for the C-x w
bindings to work?

http://git.savannah.gnu.org/cgit/emacs.git/tree/lisp/hi-lock.el?h=emacs-25

Coincidentally, currently a parallel discussion is going on in another
emacs-devel thread if we should remove the C-x w prefix map associated to
hi-lock. Please check out that thread for the reasoning behind it.

--
Kaushal Modi
On Feb 5, 2016 6:34 AM, "Alan Mackenzie" <acm@muc.de> wrote:

> Hello, Eli.
>
> On Fri, Feb 05, 2016 at 01:19:13PM +0200, Eli Zaretskii wrote:
> > > Date: Fri, 5 Feb 2016 10:37:44 +0000
> > > From: Alan Mackenzie <acm@muc.de>
> > >
> > > The traditional key bindings for hi-lock commands, beginning C-x w,
> have
> > > disappeared from the 25.1 global key map.  They have been replaced by
> > > bindings beginning M-s h.
>
> > That's not what I see here: when I turn on hi-lock-mode, the "C-x w"
> > bindings are shown in the output of "C-h b".
>
> > What exactly do you see that led you to the above conclusion?
>
> emacs -Q
>
> C-x w  (intending C-x w C-h) responds "C-x w is undefined".
>
> On the other hand,
> M-s h C-h displays the bindings in *Help*.
>
> That led me to believe the C-x h bindings had been removed.  It would
> appear there's a difference between how the C-x w bindings are set and
> how the M-s h bindings are set.
>
> --
> Alan Mackenzie (Nuremberg, Germany).
>
>

[-- Attachment #2: Type: text/html, Size: 2059 bytes --]

Re: Disappearance of hi-lock bindings from global key map.

[Eli Zaretskii]

> Date: Fri, 5 Feb 2016 11:36:35 +0000
> Cc: emacs-devel@gnu.org
> From: Alan Mackenzie <acm@muc.de>
> 
> > > The traditional key bindings for hi-lock commands, beginning C-x w, have
> > > disappeared from the 25.1 global key map.  They have been replaced by
> > > bindings beginning M-s h.
> 
> > That's not what I see here: when I turn on hi-lock-mode, the "C-x w"
> > bindings are shown in the output of "C-h b".
> 
> > What exactly do you see that led you to the above conclusion?
> 
> emacs -Q
> 
> C-x w  (intending C-x w C-h) responds "C-x w is undefined".
> 
> On the other hand,
> M-s h C-h displays the bindings in *Help*.

I don't think you should expect the bindings of hi-lock-mode be in
effect before you turned on that mode, don't you agree?

Re: Disappearance of hi-lock bindings from global key map.

[Alan Mackenzie]

Hello, Eli.

On Fri, Feb 05, 2016 at 04:28:03PM +0200, Eli Zaretskii wrote:
> > Date: Fri, 5 Feb 2016 11:36:35 +0000
> > Cc: emacs-devel@gnu.org
> > From: Alan Mackenzie <acm@muc.de>

> > > > The traditional key bindings for hi-lock commands, beginning C-x w, have
> > > > disappeared from the 25.1 global key map.  They have been replaced by
> > > > bindings beginning M-s h.

> > > That's not what I see here: when I turn on hi-lock-mode, the "C-x w"
> > > bindings are shown in the output of "C-h b".

> > > What exactly do you see that led you to the above conclusion?

> > emacs -Q

> > C-x w  (intending C-x w C-h) responds "C-x w is undefined".

> > On the other hand,
> > M-s h C-h displays the bindings in *Help*.

> I don't think you should expect the bindings of hi-lock-mode be in
> effect before you turned on that mode, don't you agree?

I'm not sure whether I agree or not.  ;-)  But I would expect them either
to be in effect, or not to be in effect.  At the moment, half of them are
in effect before the minor mode is enabled.  This is self-evidently
confusing, at least for me.

More exactly, hi-lock-mode had changed in my mental model from being
something requiring being enabled, to a fundamental feature, always
present.  This false model is what caused my original misunderstanding.

The question remaining is why the M-s h bindings are available without
turning on the mode.  Probably this isn't an important thing to resolve.

Experimentation shows that M-s h r enables hi-lock-mode, making all the
bindings available, whereas C-x w h remains unbound until such
enablement.  This is untidy.

-- 
Alan Mackenzie (Nuremberg, Germany).

Re: Disappearance of hi-lock bindings from global key map.

[John Wiegley]

>>>>> Alan Mackenzie <acm@muc.de> writes:

> Experimentation shows that M-s h r enables hi-lock-mode, making all the
> bindings available, whereas C-x w h remains unbound until such enablement.
> This is untidy.

Yes, it is.

-- 
John Wiegley                  GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com                          60E1 46C4 BD1A 7AC1 4BA2

Re: Disappearance of hi-lock bindings from global key map.

[Stefan Monnier]

> That led me to believe the C-x h bindings had been removed.  It would
> appear there's a difference between how the C-x w bindings are set and
> how the M-s h bindings are set.

Yes.  AFAIK the C-x w bindings are "legacy" bindings kept for "user
compatibility", and we just tried to preserve their earlier behavior
for now (until we remove them in some future version).

The new M-s h bindings OTOH behave slightly differently because we took
the opportunity to try and improve the behavior.


        Stefan

Re: Disappearance of hi-lock bindings from global key map.

[John Wiegley]

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

> Yes. AFAIK the C-x w bindings are "legacy" bindings kept for "user
> compatibility", and we just tried to preserve their earlier behavior for now
> (until we remove them in some future version).

> The new M-s h bindings OTOH behave slightly differently because we took the
> opportunity to try and improve the behavior.

Excellent, the deprecated they now are. What is the way to indicate that for a
keybinding; just the NEWS file? We can then remove in 27.1, to give everyone
time to switch.

-- 
John Wiegley                  GPG fingerprint = 4710 CF98 AF9B 327B B80F
http://newartisans.com                          60E1 46C4 BD1A 7AC1 4BA2

Re: Disappearance of hi-lock bindings from global key map.

[Eli Zaretskii]

> From: Stefan Monnier <monnier@iro.umontreal.ca>
> Date: Mon, 08 Feb 2016 08:54:42 -0500
> 
> > That led me to believe the C-x h bindings had been removed.  It would
> > appear there's a difference between how the C-x w bindings are set and
> > how the M-s h bindings are set.
> 
> Yes.  AFAIK the C-x w bindings are "legacy" bindings kept for "user
> compatibility", and we just tried to preserve their earlier behavior
> for now (until we remove them in some future version).
> 
> The new M-s h bindings OTOH behave slightly differently because we took
> the opportunity to try and improve the behavior.

Shouldn't this piece of info be in NEWS?  If so, could you please add
it?

Thanks.

Re: Disappearance of hi-lock bindings from global key map.

[Stefan Monnier]

> Shouldn't this piece of info be in NEWS?  If so, could you please add
> it?

Done,


        Stefan

Re: Disappearance of hi-lock bindings from global key map.

[Eli Zaretskii]

> From: Stefan Monnier <monnier@IRO.UMontreal.CA>
> Date: Fri, 12 Feb 2016 22:33:47 -0500
> Cc: emacs-devel@gnu.org
> 
> > Shouldn't this piece of info be in NEWS?  If so, could you please add
> > it?
> 
> Done,

Thanks.