modeline colors in emacs 24.2.1

modeline colors in emacs 24.2.1

[Ken Goldman]

I just installed emacs 24.2.1 (on Windows 7 if that matters) and am 
going through the usual task of porting my .emacs.

I notice that the modeline seems to be reversed.  That is, what used to 
be the foreground is now the background and vice versa.  The text is 
appearing in the background color.  Is this a known change?  It seems odd.

Re: modeline colors in emacs 24.2.1

[Eli Zaretskii]

> From: Ken Goldman <kgoldman@us.ibm.com>
> Date: Tue, 04 Dec 2012 14:17:36 -0500
> 
> I just installed emacs 24.2.1 (on Windows 7 if that matters) and am 
> going through the usual task of porting my .emacs.
> 
> I notice that the modeline seems to be reversed.  That is, what used to 
> be the foreground is now the background and vice versa.  The text is 
> appearing in the background color.  Is this a known change?  It seems odd.

There's no change in colors of the mode line.  It's something in your
.emacs, or in Windows setup, or whatever.

Re: modeline colors in emacs 24.2.1

[Bernardo]

It was a dark and stormy night when Ken Goldman said  (05/12/12 06:17):
> I just installed emacs 24.2.1 (on Windows 7 if that matters) and am 
> going through the usual task of porting my .emacs.
> 
> I notice that the modeline seems to be reversed.  That is, what used to 
> be the foreground is now the background and vice versa.  The text is 
> appearing in the background color.  Is this a known change?  It seems odd.
> 

i use half a dozen (mostly MS Windows, mostly under Remote Desktop) PCs 
at work, and have seen foreground/background modeline colours swapped -- 
on *some* of them; they all use almost identical .emacs; this "feature" 
has been around for a while (probably since version 22.x) if indeed it 
is an Emacs feature and not a configuration problem;

i've never had a closer look and just used a manually set variable in 
.emacs to swap the modeline colours

(not much help, i know)

Re: modeline colors in emacs 24.2.1

[Ken Goldman]

On 12/5/2012 5:09 AM, Bernardo wrote:
> It was a dark and stormy night when Ken Goldman said  (05/12/12 06:17):
>> I just installed emacs 24.2.1 (on Windows 7 if that matters) and am
>> going through the usual task of porting my .emacs.
>>
>> I notice that the modeline seems to be reversed.  That is, what used
>> to be the foreground is now the background and vice versa.  The text
>> is appearing in the background color.  Is this a known change?  It
>> seems odd.
>>
>
> i use half a dozen (mostly MS Windows, mostly under Remote Desktop) PCs
> at work, and have seen foreground/background modeline colours swapped --
> on *some* of them; they all use almost identical .emacs; this "feature"
> has been around for a while (probably since version 22.x) if indeed it
> is an Emacs feature and not a configuration problem;
>
> i've never had a closer look and just used a manually set variable in
> .emacs to swap the modeline colours

Yes, that was it.  When I view it locally, the mode line colors are 
correct.  When I use Remote Desktop Connection, they are reversed.

The local machine is XP.  The remote machines are XP/emacs 22 and 
Win7/emacs 24.

Is this a bug that should be reported?  Is there a work around?

I have a fast connection, and the colors can be displayed when I edit 
.emacs.

Re: modeline colors in emacs 24.2.1

[Eli Zaretskii]

> From: Ken Goldman <kgoldman@us.ibm.com>
> Date: Wed, 05 Dec 2012 10:24:15 -0500
> 
> Yes, that was it.  When I view it locally, the mode line colors are 
> correct.  When I use Remote Desktop Connection, they are reversed.

Ah, that.  Remote Desktop connections lie to Emacs about the number of
colors that are available, so I think this is the reason.

What does x-display-color-cells return on that remote desktop?

> The local machine is XP.  The remote machines are XP/emacs 22 and 
> Win7/emacs 24.
> 
> Is this a bug that should be reported?  Is there a work around?

It would be best if you debug the problem, as none of the Emacs
maintainers seem to have access to such configuration to debug it.

You may wish to try the latest pretest of Emacs 24.3, I think
something was done in that area lately.

Re: modeline colors in emacs 24.2.1

[Ken Goldman]

On 12/5/2012 11:24 AM, Eli Zaretskii wrote:
>> From: Ken Goldman <kgoldman@us.ibm.com>
>> Date: Wed, 05 Dec 2012 10:24:15 -0500
>>
>> Yes, that was it.  When I view it locally, the mode line colors are
>> correct.  When I use Remote Desktop Connection, they are reversed.
>
> Ah, that.  Remote Desktop connections lie to Emacs about the number of
> colors that are available, so I think this is the reason.
>
> What does x-display-color-cells return on that remote desktop?

M-x   x-display-color-cells   says (no match)

The help says it's in the C source code.

Re: modeline colors in emacs 24.2.1

[Eli Zaretskii]

> From: Ken Goldman <kgoldman@us.ibm.com>
> Date: Wed, 05 Dec 2012 15:56:55 -0500
> 
> > What does x-display-color-cells return on that remote desktop?
> 
> M-x   x-display-color-cells   says (no match)

It's not a command, it's a function.  Try

  M-: (x-display-color-cells) RET

Re: modeline colors in emacs 24.2.1

[Ken Goldman]

On 12/5/2012 4:14 PM, Eli Zaretskii wrote:
>> From: Ken Goldman <kgoldman@us.ibm.com>
>> Date: Wed, 05 Dec 2012 15:56:55 -0500
>>
>>> What does x-display-color-cells return on that remote desktop?
>>
>> M-x   x-display-color-cells   says (no match)
>
> It's not a command, it's a function.  Try
>
>    M-: (x-display-color-cells) RET
>
>

Debugger entered--Lisp error: (void-variable x-display-color-cells)
   eval(x-display-color-cells nil)
   eval-expression(x-display-color-cells nil)
   call-interactively(eval-expression nil nil)

On a local platform, I get a similar, but not exactly the same, response.

Re: modeline colors in emacs 24.2.1

[Peter Dyballa]

Am 06.12.2012 um 15:03 schrieb Ken Goldman:

> Debugger entered--Lisp error: (void-variable x-display-color-cells)
>  eval(x-display-color-cells nil)
>  eval-expression(x-display-color-cells nil)
>  call-interactively(eval-expression nil nil)

In *scratch* buffer insert exactly this text:

	(x-display-color-cells)

At the right-most parenthesis type C-j. This you will execute the function x-display-color-cells and insert the function's return value into the *scratch* buffer,

--
Greetings

  Pete

If you don't find it in the index, look very carefully through the entire catalogue.
		–  Sears, Roebuck, and Co., Consumer's Guide, 1897

Re: modeline colors in emacs 24.2.1

[Ken Goldman]

On 12/6/2012 10:47 AM, Peter Dyballa wrote:
>
> At the right-most parenthesis type C-j. This you will execute the
> function x-display-color-cells and insert the function's return value
> into the *scratch* buffer,

I get 20, which is evidently the problem.  Locally I get 16 meg.

Is there a way to change this through .emacs or elsewhere, or is it hard 
coded?

Re: modeline colors in emacs 24.2.1

[Eli Zaretskii]

> From: Ken Goldman <kgoldman@us.ibm.com>
> Date: Thu, 06 Dec 2012 11:16:21 -0500
> 
> On 12/6/2012 10:47 AM, Peter Dyballa wrote:
> >
> > At the right-most parenthesis type C-j. This you will execute the
> > function x-display-color-cells and insert the function's return value
> > into the *scratch* buffer,
> 
> I get 20, which is evidently the problem.  Locally I get 16 meg.

Yes, that's the reason for the problem.

> Is there a way to change this through .emacs or elsewhere, or is it hard 
> coded?

Like I said, try the latest pretest of 24.3, I think this problem was
fixed there.