'Strange' changes in status line

'Strange' changes in status line

[Angelo Graziosi]

In the status line, now I see 'Emacs' near the buffer name

--:----Emacs foo.f90 59% ....

Just a curiosity, Why this change and which usefulness?

     Angelo

Re: 'Strange' changes in status line

[Glenn Morris]

Angelo Graziosi wrote:

> In the status line, now I see 'Emacs' near the buffer name
>
> --:----Emacs foo.f90 59% ....
>
> Just a curiosity, Why this change and which usefulness?

I believe it's an unintended consequence of this:

2008-08-25  Eli Zaretskii  <eliz@gnu.org>

    * bindings.el (mode-line-frame-control): New function, caters to
    `pc' ``window system''.
    (mode-line-frame-identification): Use it instead of
    accessing window-system directly.


In X:

emacs -Q

C-h v mode-line-frame-identification
Its value is ("-%F  ")

M-: (mode-line-frame-control)
-> " "

ie, the variable is being set at the wrong time.

Re: 'Strange' changes in status line

[Angelo Graziosi]

Glenn Morris ha scritto:
> Angelo Graziosi wrote:
> 
>> In the status line, now I see 'Emacs' near the buffer name
>>
>> --:----Emacs foo.f90 59% ....
>>
>> Just a curiosity, Why this change and which usefulness?
> 
> I believe it's an unintended consequence of this:
> 
> 2008-08-25  Eli Zaretskii  <eliz@gnu.org>
> 
>     * bindings.el (mode-line-frame-control): New function, caters to
>     `pc' ``window system''.
>     (mode-line-frame-identification): Use it instead of
>     accessing window-system directly.
> 
> 
> In X:
> 
> emacs -Q
> 
> C-h v mode-line-frame-identification
> Its value is ("-%F  ")
> 
> M-: (mode-line-frame-control)
> -> " "
> 
> ie, the variable is being set at the wrong time.

In other words: is it a BUG?

    Angelo

Re: 'Strange' changes in status line

[Eli Zaretskii]

> Date: Thu, 28 Aug 2008 10:09:44 +0200
> From: Angelo Graziosi <angelo.graziosi@alice.it>
> Cc: Eli Zaretskii <eliz@gnu.org>, emacs-devel@gnu.org
> 
> > C-h v mode-line-frame-identification
> > Its value is ("-%F  ")
> > 
> > M-: (mode-line-frame-control)
> > -> " "
> > 
> > ie, the variable is being set at the wrong time.
> 
> In other words: is it a BUG?

Yes, of course, and I'm sorry for introducing it.

I'll try to debug it soon, unless someone beats me to it.

Re: 'Strange' changes in status line

[Eli Zaretskii]

> From: Glenn Morris <rgm@gnu.org>
> Date: Thu, 28 Aug 2008 02:00:51 -0400
> Cc: Eli Zaretskii <eliz@gnu.org>, emacs-devel@gnu.org
> 
> Angelo Graziosi wrote:
> 
> > In the status line, now I see 'Emacs' near the buffer name
> >
> > --:----Emacs foo.f90 59% ....
> >
> > Just a curiosity, Why this change and which usefulness?
> 
> I believe it's an unintended consequence of this:
> 
> 2008-08-25  Eli Zaretskii  <eliz@gnu.org>
> 
>     * bindings.el (mode-line-frame-control): New function, caters to
>     `pc' ``window system''.
>     (mode-line-frame-identification): Use it instead of
>     accessing window-system directly.
> 
> 
> In X:
> 
> emacs -Q
> 
> C-h v mode-line-frame-identification
> Its value is ("-%F  ")
> 
> M-: (mode-line-frame-control)
> -> " "
> 
> ie, the variable is being set at the wrong time.

I don't see this problem on MS-Windows: the frame name is not
displayed in the mode line, and the value of
mode-line-frame-identification is "  ", as I'd expect.

I won't have access to a system with X for a couple of days.  If you
still see this with the current CVS, please put a breakpoint in
decode_mode_spec and tell me when and from where it is called ``at the
wrong time''.

Thanks.

Re: 'Strange' changes in status line

[Eli Zaretskii]

> From: Glenn Morris <rgm@gnu.org>
> Cc: emacs-devel@gnu.org, Eli Zaretskii <eliz@gnu.org>
> Date: Thu, 28 Aug 2008 02:00:51 -0400
> 
> Angelo Graziosi wrote:
> 
> > In the status line, now I see 'Emacs' near the buffer name
> >
> > --:----Emacs foo.f90 59% ....
> >
> > Just a curiosity, Why this change and which usefulness?
> 
> I believe it's an unintended consequence of this:
> 
> 2008-08-25  Eli Zaretskii  <eliz@gnu.org>
> 
>     * bindings.el (mode-line-frame-control): New function, caters to
>     `pc' ``window system''.
>     (mode-line-frame-identification): Use it instead of
>     accessing window-system directly.
> 
> 
> In X:
> 
> emacs -Q
> 
> C-h v mode-line-frame-identification
> Its value is ("-%F  ")
> 
> M-: (mode-line-frame-control)
> -> " "
> 
> ie, the variable is being set at the wrong time.

Does this change fix the problem on X?  (I cannot try this on X right
now, but I verified that it works on a tty, MS-Windows, and MS-DOS.)

Index: lisp/bindings.el
===================================================================
RCS file: /cvsroot/emacs/emacs/lisp/bindings.el,v
retrieving revision 1.211
diff -u -r1.211 bindings.el
--- lisp/bindings.el	25 Aug 2008 21:50:55 -0000	1.211
+++ lisp/bindings.el	28 Aug 2008 19:30:19 -0000
@@ -219,8 +219,8 @@
 (defun mode-line-frame-control ()
   "Compute mode-line control for frame identification.
 Value is used for `mode-line-frame-identification', which see."
-  (if (or (null (window-system))
-	  (eq (window-system) 'pc))
+  (if (or (null initial-window-system)
+	  (eq initial-window-system 'pc))
       "-%F  "
     "  "))
 

Re: 'Strange' changes in status line

[Glenn Morris]

Eli Zaretskii wrote:

> Does this change fix the problem on X?  (I cannot try this on X right
> now, but I verified that it works on a tty, MS-Windows, and MS-DOS.)

Afraid not (guess it gets eval'd during dumping).

In any case, an approach like that is not going to work for multi-tty
with mixed X and tty frames. You need something that's eval'd at
runtime. This seems to work:

(setq mode-line-frame-identification '(:eval (mode-line-frame-control))) 
(put 'mode-line-frame-identification 'risky-local-variable t)

Re: 'Strange' changes in status line

[Eli Zaretskii]

> From: Glenn Morris <rgm@gnu.org>
> Cc: angelo.graziosi@alice.it,  emacs-devel@gnu.org
> Date: Fri, 29 Aug 2008 02:34:22 -0400
> 
> In any case, an approach like that is not going to work for multi-tty
> with mixed X and tty frames. You need something that's eval'd at
> runtime.

I thought I was doing just that ;-)  But quoting in Lisp was always my
weak spot; thanks for showing me the light.

> This seems to work:
> 
> (setq mode-line-frame-identification '(:eval (mode-line-frame-control))) 
> (put 'mode-line-frame-identification 'risky-local-variable t)

Installed.

Re: 'Strange' changes in status line

[Glenn Morris]

Eli Zaretskii wrote:

>> (setq mode-line-frame-identification '(:eval (mode-line-frame-control))) 
>> (put 'mode-line-frame-identification 'risky-local-variable t)
>
> Installed.

No need for a separate setq, you can do the setting in the defvar
(sorry for being unclear, that was an example rather than a patch).

Also, the change to use initial-window-system in
mode-line-frame-control is wrong, again because of multi-tty mixed X
and tty frames. Just using the variable (no need for the function)
window-system seems fine.

Re: 'Strange' changes in status line

[Eli Zaretskii]

> From: Glenn Morris <rgm@gnu.org>
> Cc: angelo.graziosi@alice.it,  emacs-devel@gnu.org
> Date: Fri, 29 Aug 2008 12:34:40 -0400
> 
> Eli Zaretskii wrote:
> 
> >> (setq mode-line-frame-identification '(:eval (mode-line-frame-control))) 
> >> (put 'mode-line-frame-identification 'risky-local-variable t)
> >
> > Installed.
> 
> No need for a separate setq, you can do the setting in the defvar
> (sorry for being unclear, that was an example rather than a patch).
> 
> Also, the change to use initial-window-system in
> mode-line-frame-control is wrong, again because of multi-tty mixed X
> and tty frames. Just using the variable (no need for the function)
> window-system seems fine.

Fixed.