Change in handling of the Windows key

Change in handling of the Windows key

[Richard Copley]

In recent-ish builds of the emacs-25 branch, I find that the Windows
key is intercepted by Emacs and no longer opens the start menu. (The
Windows key still opens the start menu if Emacs is not the foreground
application.) Is this a deliberate change?

(This is on Windows 7, which I have to use at work. On Windows 10 I
can still open the start screen in the usual way.)

I know the lack of a usable super key on Windows has always annoyed
those who are used to super key bindings, so I understand why one
might want to intercept the Windows key, but some users (for example,
me) don't use any such bindings in Emacs and like to be able to open
the start menu by pressing the Windows key.

Isn't it too late in the release cycle to be adding such a disruptive change?

That aside, whenever the change is released, please will someone add
instructions to the NEWS stanza to explain how to configure Emacs so
that the Windows key behaves the way it used to?

Re: Change in handling of the Windows key

[Eli Zaretskii]

> From: Richard Copley <rcopley@gmail.com>
> Date: Wed, 22 Jun 2016 11:46:14 +0100
> 
> In recent-ish builds of the emacs-25 branch, I find that the Windows
> key is intercepted by Emacs and no longer opens the start menu. (The
> Windows key still opens the start menu if Emacs is not the foreground
> application.) Is this a deliberate change?
> 
> (This is on Windows 7, which I have to use at work. On Windows 10 I
> can still open the start screen in the usual way.)

I cannot reproduce this.  I tried Emacs 25.0.95 on 2 Windows 7
systems, and the latest emacs-25 branch on XP, and in all 3 cases the
Windows key still works as it always did.

So some other factor must be at work here.

For starters, does this happen to you in "emacs -Q"?

> Isn't it too late in the release cycle to be adding such a disruptive change?

It is, which is why no one did any such changes.  In fact the Windows
keyboard support was not touched since April.

Re: Change in handling of the Windows key

[Óscar Fuentes]

Eli Zaretskii <eliz@gnu.org> writes:

>> In recent-ish builds of the emacs-25 branch, I find that the Windows
>> key is intercepted by Emacs and no longer opens the start menu. (The
>> Windows key still opens the start menu if Emacs is not the foreground
>> application.) Is this a deliberate change?
>> 
>> (This is on Windows 7, which I have to use at work. On Windows 10 I
>> can still open the start screen in the usual way.)
>
> I cannot reproduce this.  I tried Emacs 25.0.95 on 2 Windows 7
> systems, and the latest emacs-25 branch on XP, and in all 3 cases the
> Windows key still works as it always did.
>
> So some other factor must be at work here.

I can reproduce with 25.1.50.1 (i686-w64-mingw32) of 2016-06-06 on
Windows XP. C-h k says

<lwindow> runs the command ignore [rest of text elided]

> For starters, does this happen to you in "emacs -Q"?

Yes.

RE: Change in handling of the Windows key

[Drew Adams]

> I can reproduce with 25.1.50.1 (i686-w64-mingw32) of 2016-06-06 on
> Windows XP. C-h k says
> 
> <lwindow> runs the command ignore [rest of text elided]
> 
> > For starters, does this happen to you in "emacs -Q"?
> 
> Yes.

In all releases of Emacs (20-24), `C-h k <lwindow>' prints help
showing that "lwindow runs the command ignore" or similar text
(e.g. "runs the command ignore (found in global-map), which is an
interactive compiled Lisp function in 'subr.el'").

But it also opens the Windows start menu.

(In the latest Emacs 25 build I have, which is from 2015-12-10,
this is still the case.)

Re: Change in handling of the Windows key

[Óscar Fuentes]

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

[snip]
> But it also opens the Windows start menu.
[snip]

Just to clarify: I can reproduce the behavior the OP described wrt the
Windows Start menu *not* opening when pressing the Windows key.

Re: Change in handling of the Windows key

[Eli Zaretskii]

> From: Óscar Fuentes <ofv@wanadoo.es>
> Date: Wed, 22 Jun 2016 19:23:17 +0200
> 
> I can reproduce with 25.1.50.1 (i686-w64-mingw32) of 2016-06-06 on
> Windows XP. C-h k says

In 25.1.50, I see this as well (and the handling of this and other
similar keys was indeed changed on master).  But the OP talked about
emacs-25 and the pretest, so either there's a confusion somewhere, or
there's another problem.

Jussi, can you look into this?  w32-pass-lwindow-to-system is t, and
yet the left Windows key doesn't open the Start menu.  Curiously, the
right Windows key does.

Re: Change in handling of the Windows key

[Richard Copley]

On 22 June 2016 at 19:06, Eli Zaretskii <eliz@gnu.org> wrote:
>> From: Óscar Fuentes <ofv@wanadoo.es>
>> Date: Wed, 22 Jun 2016 19:23:17 +0200
>>
>> I can reproduce with 25.1.50.1 (i686-w64-mingw32) of 2016-06-06 on
>> Windows XP. C-h k says
>
> In 25.1.50, I see this as well (and the handling of this and other
> similar keys was indeed changed on master).  But the OP talked about
> emacs-25 and the pretest, so either there's a confusion somewhere, or
> there's another problem.
>
> Jussi, can you look into this?  w32-pass-lwindow-to-system is t, and
> yet the left Windows key doesn't open the Start menu.  Curiously, the
> right Windows key does.

Since this isn't an intentional feature, I've created a proper bug
report (#23830) to answer Eli's questions. (I hadn't seen the other
replies until now.)