bug#25330: emacs -q always produces an emacs that was partially off screen.

bug#25330: emacs -q always produces an emacs that was partially off screen.

[積丹尼 Dan Jacobson]

[-- Attachment #1: Type: text/plain, Size: 99 bytes --]

For the past 1000 years, doing
$ emacs -q
always produces an emacs that was partially off screen.


[-- Attachment #2: screenshot --]
[-- Type: image/jpeg, Size: 131968 bytes --]

[-- Attachment #3: Type: text/plain, Size: 303 bytes --]


Today I read
http://unix.stackexchange.com/questions/14159/how-do-i-find-the-window-dimensions-and-position-accurately-including-decoration
and I think the reason might be related to it.

Yes I workaround it in http://jidanni.org/comp/configuration/ , but I
would just like to let you know.

Emacs 25.

bug#25330: emacs -q always produces an emacs that was partially off screen.

[martin rudalics]

 > For the past 1000 years, doing
 > $ emacs -q
 > always produces an emacs that was partially off screen.

Could you, in that frame, please evaluate the following three forms:

(display-monitor-attributes-list)

(frame-geometry)

(frame-parameters)

and post the results here?

 > Today I read
 > http://unix.stackexchange.com/questions/14159/how-do-i-find-the-window-dimensions-and-position-accurately-including-decoration
 > and I think the reason might be related to it.

Hardly.  Without any customizations, the first frame you see has 36
lines and 80 columns.  Emacs does not query the size of the workspace
and leaves it to the window manager to make the entire frame visible.

 > Yes I workaround it in http://jidanni.org/comp/configuration/ ,

The only frame-related setting I found there is that you maximize the
default frame.  Am I missing anything else?

Thanks, martin

bug#25330: emacs -q always produces an emacs that was partially off screen.

[積丹尼 Dan Jacobson]

(I had to go back and do C-u 0 C-x C-e to get rid of truncation.)

In the bad initial off window state:

icewm's size command (ALT+F8) shows
79x35+606+41

(display-monitor-attributes-list)(((name . "LVDS-1") (geometry 0 0 1366
768) (workarea 0 0 1366 768) (mm-size 344 193) (frames #<frame
emacs@jidanni5 0x12ffc30>) (source . "Gdk")))

(frame-geometry)((outer-position 606 . 41) (outer-size 760 . 743)
(external-border-size 4 . 4) (title-bar-size 0 . 20) (menu-bar-external
. t) (menu-bar-size 752 . 25) (tool-bar-external . t) (tool-bar-position
. top) (tool-bar-size 752 . 42) (internal-border-width . 0))

(frame-parameters)

(frame-parameters)((parent-id . 12603744) (explicit-name) (display .
":0") (visibility . t) (icon-name) (outer-window-id . "25165848")
(window-id . "25165852") (top . 41) (left . 606) (buried-buffer-list)
(buffer-list #<buffer *scratch*> #<buffer *GNU Emacs*> #<buffer
*Minibuf-1*>) (unsplittable) (minibuffer . #<window 4 on *Minibuf-0*>)
(modeline . t) (width . 80) (height . 36) (name . "emacs@jidanni5")
(environment) (sticky) (cursor-color . "black") (background-mode .
light) (display-type . color) (window-system . x) (fullscreen) (alpha)
(scroll-bar-height . 0) (scroll-bar-width . 16) (cursor-type . box)
(auto-lower) (auto-raise) (icon-type) (tool-bar-position . top)
(wait-for-wm . t) (title) (buffer-predicate) (tool-bar-lines . 1)
(menu-bar-lines . 1) (scroll-bar-background) (scroll-bar-foreground)
(right-fringe . 8) (left-fringe . 8) (line-spacing) (screen-gamma)
(border-color . "black") (mouse-color . "black") (background-color .
"white") (foreground-color . "black") (horizontal-scroll-bars)
(vertical-scroll-bars . right) (bottom-divider-width . 0)
(right-divider-width . 0) (internal-border-width . 0) (border-width . 0)
(font-parameter . "Monospace 11") (font . "-unknown-DejaVu Sans
Mono-normal-normal-normal-*-15-*-*-*-m-0-iso10646-1") (font-backend xft
x))

Now after I grab the window and move it to its proper place on screen:
I get
79x35+606+3

(display-monitor-attributes-list)(((name . "LVDS-1") (geometry 0 0 1366
768) (workarea 0 0 1366 768) (mm-size 344 193) (frames #<frame
emacs@jidanni5 0x12ffc30>) (source . "Gdk")))

(frame-geometry)((outer-position 606 . 3) (outer-size 760 . 743)
(external-border-size 4 . 4) (title-bar-size 0 . 20) (menu-bar-external
. t) (menu-bar-size 752 . 25) (tool-bar-external . t) (tool-bar-position
. top) (tool-bar-size 752 . 42) (internal-border-width . 0))

(frame-parameters)((parent-id . 12603744) (explicit-name) (display .
":0") (visibility . t) (icon-name) (outer-window-id . "25165848")
(window-id . "25165852") (top + -4) (left . 606) (buried-buffer-list)
(buffer-list #<buffer *scratch*> #<buffer *GNU Emacs*> #<buffer
*Minibuf-1*>) (unsplittable) (minibuffer . #<window 4 on *Minibuf-0*>)
(modeline . t) (width . 80) (height . 36) (name . "emacs@jidanni5")
(environment) (sticky) (cursor-color . "black") (background-mode .
light) (display-type . color) (window-system . x) (fullscreen) (alpha)
(scroll-bar-height . 0) (scroll-bar-width . 16) (cursor-type . box)
(auto-lower) (auto-raise) (icon-type) (tool-bar-position . top)
(wait-for-wm . t) (title) (buffer-predicate) (tool-bar-lines . 1)
(menu-bar-lines . 1) (scroll-bar-background) (scroll-bar-foreground)
(right-fringe . 8) (left-fringe . 8) (line-spacing) (screen-gamma)
(border-color . "black") (mouse-color . "black") (background-color .
"white") (foreground-color . "black") (horizontal-scroll-bars)
(vertical-scroll-bars . right) (bottom-divider-width . 0)
(right-divider-width . 0) (internal-border-width . 0) (border-width . 0)
(font-parameter . "Monospace 11") (font . "-unknown-DejaVu Sans
Mono-normal-normal-normal-*-15-*-*-*-m-0-iso10646-1") (font-backend xft
x))

This last frame-parameters I had to redo to get rid of truncation "..."
so it is in a slightly different position.

mr> The only frame-related setting I found there is that you maximize the
mr> default frame.  Am I missing anything else?

That's all. The only special thing about my icewm set up is
$ find .icewm/ -size 0
.icewm/menu
.icewm/toolbar
.icewm/keys

In my maximized window, there is no icewm size command available, and I get

(display-monitor-attributes-list)(((name . "LVDS-1") (geometry 0 0 1366
768) (workarea 0 0 1366 768) (mm-size 344 193) (frames #<frame
E06:53macs 0x12ffc30>) (source . "Gdk")))

(frame-geometry)((outer-position -4 . -4) (outer-size 1368 . 748)
(external-border-size 4 . 4) (title-bar-size 0 . 20) (menu-bar-external
. t) (menu-bar-size 0 . 0) (tool-bar-external . t) (tool-bar-position .
top) (tool-bar-size 0 . 0) (internal-border-width . 0))

(frame-parameters)((parent-id . 12583383) (explicit-name . t) (display .
":0") (visibility . t) (icon-name) (outer-window-id . "20971544")
(window-id . "20971548") (top + -4) (left + -4) (buried-buffer-list
#<buffer *Completions*> #<buffer jidanni> #<buffer .icewm> #<buffer
.newsrc-dribble>) (buffer-list #<buffer *unsent wide reply to martin
rudalics*> #<buffer *Help*> #<buffer *Minibuf-1*> #<buffer *Article
nnml:mail.misc*> #<buffer *Buffer List*> #<buffer preferences*> #<buffer
*Summary nnml:mail.misc*> #<buffer *Group*> #<buffer hhh.txt>)
(unsplittable) (minibuffer . #<window 4 on *Minibuf-0*>) (modeline . t)
(width . 166) (height . 45) (name . "E06:53macs") (environment) (sticky)
(cursor-color . "White") (background-mode . dark) (display-type . color)
(window-system . x) (fullscreen . maximized) (alpha) (scroll-bar-height
. 0) (scroll-bar-width . 16) (cursor-type . box) (auto-lower)
(auto-raise) (icon-type) (tool-bar-position . top) (wait-for-wm . t)
(title) (buffer-predicate) (tool-bar-lines . 0) (menu-bar-lines . 0)
(scroll-bar-background) (scroll-bar-foreground) (right-fringe . 8)
(left-fringe . 8) (line-spacing) (screen-gamma) (border-color . "black")
(mouse-color . "black") (background-color . "DarkSlateGray")
(foreground-color . "Wheat") (horizontal-scroll-bars)
(vertical-scroll-bars . right) (bottom-divider-width . 0)
(right-divider-width . 0) (internal-border-width . 0) (border-width . 0)
(font-parameter .
"-sony-fixed-medium-r-normal--16-120-100-100-c-80-iso8859-1") (font .
"-sony-fixed-medium-r-normal--16-120-100-100-c-80-iso8859-1")
(font-backend xft x))

bug#25330: emacs -q always produces an emacs that was partially off screen.

[martin rudalics]

 > (I had to go back and do C-u 0 C-x C-e to get rid of truncation.)

I evaluate idioms like

(insert (format "\n%s" (pp (frame-parameters))))

in *scratch* instead.

 > In the bad initial off window state:
 >
 > icewm's size command (ALT+F8) shows
 > 79x35+606+41

I don't know what the first two values stand for.  The 3rd and 4th
values are later reported by ‘frame-geometry’ as

 > (frame-geometry)((outer-position 606 . 41) (outer-size 760 . 743)

so Emacs and icewm agree about this.  Now 41 + 743 give 784 which is
clearly larger than the 768 reported by

 > (display-monitor-attributes-list)(((name . "LVDS-1") (geometry 0 0 1366
 > 768) (workarea 0 0 1366 768) (mm-size 344 193) (frames #<frame
 > emacs@jidanni5 0x12ffc30>) (source . "Gdk")))

so icewm is (IMHO) to blame for positioning the frame with a vertical
offset of 41.  In fact, 743 < 768 so the frame should fit if it were
positioned "sensibly".

I also don't like that your height of the workarea is the same as the
height of the monitor.  In your snapshot I see a taskbar - is this a
taskbar that gets hidden automatically?

 > (frame-parameters)(... (top . 41) (left . 606) ...
 > (width . 80) (height . 36)

Here you see the position values as before and the default width and
height values for your initial frame, 80 x 36.

 > Now after I grab the window and move it to its proper place on screen:
 > I get
 > 79x35+606+3
 >
 > (display-monitor-attributes-list)(((name . "LVDS-1") (geometry 0 0 1366
 > 768) (workarea 0 0 1366 768) (mm-size 344 193) (frames #<frame
 > emacs@jidanni5 0x12ffc30>) (source . "Gdk")))
 >
 > (frame-geometry)((outer-position 606 . 3) (outer-size 760 . 743)
 > (external-border-size 4 . 4) (title-bar-size 0 . 20) (menu-bar-external
 > . t) (menu-bar-size 752 . 25) (tool-bar-external . t) (tool-bar-position
 > . top) (tool-bar-size 752 . 42) (internal-border-width . 0))

Clearly with a vertical offset of 3 the frame fits.

 > That's all. The only special thing about my icewm set up is
 > $ find .icewm/ -size 0
 > .icewm/menu
 > .icewm/toolbar
 > .icewm/keys

Do you have any other "fairly large window" say in the 750 pixels height
category that gets positioned off screen in a similar way?  Also, do you
have any Emacs related settings in some .Xresources?

 > In my maximized window, there is no icewm size command available, and I get
 >
 > (display-monitor-attributes-list)(((name . "LVDS-1") (geometry 0 0 1366
 > 768) (workarea 0 0 1366 768) (mm-size 344 193) (frames #<frame
 > E06:53macs 0x12ffc30>) (source . "Gdk")))
 >
 > (frame-geometry)((outer-position -4 . -4) (outer-size 1368 . 748)
 > (external-border-size 4 . 4) (title-bar-size 0 . 20) (menu-bar-external
 > . t) (menu-bar-size 0 . 0) (tool-bar-external . t) (tool-bar-position .
 > top) (tool-bar-size 0 . 0) (internal-border-width . 0))

These figures look good: The outer window is set off by -4 pixels so the
external border is not shown on either side.  Correct?  In addition it
seems to leave 768 - 748 - 8 that is 28 pixels for the taskbar (the
lower external border is drawn below the taskbar, I presume).

So there isn't much Emacs itself can do about this.  From you snapshot
it isn't clear to me whether the xterm window is fully visible - IIUC it
also gets partly hidden beneath the taskbar.  This would hint at a
problem with icewm not recognizing your taskbar.  (I don't assume that
the xterm window is maximized - this would contradict your statement
that a maximized window like the Emacs one is shown correctly.)

Obviously, you could try, in your .emacs, to either reduce the initial
or default height of your frame but this sounds ludicrous given the fact
that icewm deliberately throws away part of the available space.  Also
you could explicitly ask for an initial position which is, however,
silly because it overrides the placement strategy of the window manager.
You could explicitly specify the ‘top’ value alone like in

(setq default-frame-alist '((top . 3)))

to get rid of the manual readjustment but I have no idea whether your
window manager will like it.  Or try another window manager ...

martin

bug#25330: emacs -q always produces an emacs that was partially off screen.

[積丹尼 Dan Jacobson]

>>>>> "mr" == martin rudalics <rudalics@gmx.at> writes:
>> 79x35+606+41

mr> I don't know what the first two values stand for.  The 3rd and 4th
mr> values are later reported by ‘frame-geometry’ as

I always assumed they stood for width and height measured in ASCII
columns and rows.

mr> In your snapshot I see a taskbar - is this a
mr> taskbar that gets hidden automatically?

All I know is I have
$ grep bar- .emacs
(tool-bar-mode 0)
(menu-bar-mode 0)

mr> Do you have any other "fairly large window" say in the 750 pixels height
mr> category that gets positioned off screen in a similar way? 
Never.

mr> do you have any Emacs related settings in some .Xresources?

$ grep -i emacs .Xresources
Emacs.bitmapIcon: off
http://jidanni.org/comp/configuration/.Xresources

mr> These figures look good: The outer window is set off by -4 pixels so the
mr> external border is not shown on either side.  Correct?  In addition it
mr> seems to leave 768 - 748 - 8 that is 28 pixels for the taskbar (the
mr> lower external border is drawn below the taskbar, I presume).

All I know is I'm happy with what I finally use. I just feel bad for
first time users.

mr> So there isn't much Emacs itself can do about this.  From you snapshot
mr> it isn't clear to me whether the xterm window is fully visible - IIUC it
mr> also gets partly hidden beneath the taskbar.  This would hint at a
mr> problem with icewm not recognizing your taskbar.  (I don't assume that
mr> the xterm window is maximized - this would contradict your statement
mr> that a maximized window like the Emacs one is shown correctly.)

All I know is the line of boxes at the bottom of the screen (taskbar?)
is part of icewm itself. Yes, xterms are 1/2 a ASCII row too short, but
if they were 1/2 a row too long they would extend below the taskbar (if
that is what it is called.) So I am still happy with that. Anyway I am
very happy. It just took a long time to make it all that way.

Anyway, perhaps you could install icewm and run emacs inside it and see
if there is a problem with a vanilla set up. Thanks.

bug#25330: can't send directly to martin

[積丹尼 Dan Jacobson]

MDS> <rudalics@gmx.at>: delivery temporarily suspended: host
MDS>     mx00.emig.gmx.net[212.227.15.9] refused to talk to me: 421-gmx.net
MDS>     (mxgmx002) Nemesis ESMTP Service not available 421-Service unavailable
MDS>     421-Reject due to policy violations. 421 For explanation visit
MDS>     http://postmaster.gmx.com/en/error-messages?ip=66.33.216.122&c=poli

bug#25330: can't send directly to martin

[martin rudalics]

 > MDS> <rudalics@gmx.at>: delivery temporarily suspended: host
 > MDS>     mx00.emig.gmx.net[212.227.15.9] refused to talk to me: 421-gmx.net
 > MDS>     (mxgmx002) Nemesis ESMTP Service not available 421-Service unavailable
 > MDS>     421-Reject due to policy violations. 421 For explanation visit
 > MDS>     http://postmaster.gmx.com/en/error-messages?ip=66.33.216.122&c=poli

Sorry.  It seems we had this or a similar issue already once, many years
ago.  IIRC you did change something then and it worked.  OTOH your first
reply passed through to me without any problems as:

Return-Path: <jidanni@jidanni.org>
Received: from hapkido.dreamhost.com ([66.33.216.122]) by mx-ha.gmx.net
  (mxgmx012 [212.227.15.9]) with ESMTPS (Nemesis) id 0MYbfW-1c3an943dK-00VSOi
  for <rudalics@gmx.at>; Thu, 05 Jan 2017 00:31:58 +0100
Received: from homiemail-a62.g.dreamhost.com (homie.mail.dreamhost.com [208.97.132.208])
	by hapkido.dreamhost.com (Postfix) with ESMTP id 5982F96127
	for <rudalics@gmx.at>; Mon,  2 Jan 2017 15:20:27 -0800 (PST)
Received: from homiemail-a62.g.dreamhost.com (localhost [127.0.0.1])
	by homiemail-a62.g.dreamhost.com (Postfix) with ESMTP id CAE2863406C;
	Mon,  2 Jan 2017 15:20:25 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha1; c=relaxed; d=jidanni.org; h=from:to:cc
	:subject:references:date:message-id:mime-version:content-type;
	 s=jidanni.org; bh=S+CT8LdESj6iOum1GPX54/8I6Io=; b=mU9al+K1NIWGc
	487MG7Vz8oBaz08nXmvVMNpmxY9g89lnN+7DgYDpbK1Uz6Q1O4SAv1DCsN7Ny5/g
	PUeUZry4NfHrvnmMg2lVhtSOeCgZMnR9WsCTgBHFqEe8Lc+XjZMFg9kaQRETxCKV
	l3RXKhrm/KA53CrP4KhnhhI2zu2NcU=
Received: from jidanni.org (114-41-5-15.dynamic.hinet.net [114.41.5.15])
	(using TLSv1 with cipher AES256-SHA (256/256 bits))
	(No client certificate requested)
	(Authenticated sender: jidanni@jidanni.org)
	by homiemail-a62.g.dreamhost.com (Postfix) with ESMTPSA id CEACD634075;
	Mon,  2 Jan 2017 15:20:24 -0800 (PST)
From: =?utf-8?B?56mN5Li55bC8?= Dan Jacobson <jidanni@jidanni.org>
To: martin rudalics <rudalics@gmx.at>
Cc: 25330@debbugs.gnu.org
Subject: Re: bug#25330: emacs -q always produces an emacs that was partially off screen.
References: <87o9zqggcu.fsf@jidanni.org>
Date: Tue, 03 Jan 2017 07:20:21 +0800
Message-ID: <87lgutqcay.fsf@jidanni.org>

So maybe it's a temporary issue only?

martin

bug#25330: emacs -q always produces an emacs that was partially off screen.

[Lars Ingebrigtsen]

martin rudalics <rudalics@gmx.at> writes:

>> (display-monitor-attributes-list)(((name . "LVDS-1") (geometry 0 0 1366
>> 768) (workarea 0 0 1366 768) (mm-size 344 193) (frames #<frame
>> emacs@jidanni5 0x12ffc30>) (source . "Gdk")))
>
> so icewm is (IMHO) to blame for positioning the frame with a vertical
> offset of 41.  In fact, 743 < 768 so the frame should fit if it were
> positioned "sensibly".

I think the conclusion here was that either the window manager places
the window in the wrong position, or there's a user error.  So I'm
closing this bug report.

-- 
(domestic pets only, the antidote for overdose, milk.)
   bloggy blog: http://lars.ingebrigtsen.no