bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows

bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows

[Luca Capello]

[-- Attachment #1: Type: text/plain, Size: 4531 bytes --]


Please write in English if possible, because the Emacs maintainers
usually do not have translators to read other languages for them.

Your bug report will be posted to the emacs-pretest-bug@gnu.org mailing list.

Please describe exactly what actions triggered the bug
and the precise symptoms of the bug:

Hello,

please Cc: me, I'm not subscribed to the list.

In Debian emacs-snapshot_1:20080606-1 screwed up buffers and frames, but
most of those problems are fixed in emacs-snapshot_1:20080613-1.

However, it seems that `display-buffer' now opens vertical windows
instead of horizontal ones.

Romain Francoise discovered that this happens when the width of the
window to be split is bigger than its height.  He also privately
confirmed me that this bug is still present in the current CVS, so most
of the credits for this bug goes to him ;-)

This is clearly reproducible with both `emacs -Q` (GTK+) and `emacs -Q
-nw`.

Thx, bye,
Gismo / Luca

If Emacs crashed, and you have the Emacs process in the gdb debugger,
please include the output from the following gdb commands:
    `bt full' and `xbacktrace'.
If you would like to further debug the crash, please read the file
/usr/share/emacs/23.0.60/etc/DEBUG for instructions.


In GNU Emacs 23.0.60.1 (x86_64-pc-linux-gnu, GTK+ Version 2.12.10)
 of 2008-06-13 on elegiac, modified by Debian
 (emacs-snapshot package, version 1:20080613-1)
configured using `configure  '--build' 'x86_64-linux-gnu' '--host' 'x86_64-linux-gnu' '--prefix=/usr' '--sharedstatedir=/var/lib' '--libexecdir=/usr/lib' '--localstatedir=/var' '--infodir=/usr/share/info' '--mandir=/usr/share/man' '--with-pop=yes' '--enable-locallisppath=/etc/emacs-snapshot:/etc/emacs:/usr/local/share/emacs/23.0.60/site-lisp:/usr/local/share/emacs/site-lisp:/usr/share/emacs/23.0.60/site-lisp:/usr/share/emacs/site-lisp' '--with-x=yes' '--with-x-toolkit=gtk' 'build_alias=x86_64-linux-gnu' 'host_alias=x86_64-linux-gnu' 'CFLAGS=-DDEBIAN -DSITELOAD_PURESIZE_EXTRA=5000 -g -O2' 'LDFLAGS=-g -Wl,--as-needed' 'CPPFLAGS=''

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: en_US.UTF-8
  value of $XMODIFIERS: nil
  locale-coding-system: utf-8-unix
  default-enable-multibyte-characters: t

Major mode: Group

Minor modes in effect:
  shell-dirtrack-mode: t
  erc-truncate-mode: t
  erc-track-mode: t
  erc-track-minor-mode: t
  erc-services-mode: t
  erc-networks-mode: t
  erc-ring-mode: t
  erc-netsplit-mode: t
  erc-match-mode: t
  erc-pcomplete-mode: t
  erc-button-mode: t
  erc-fill-mode: t
  erc-autojoin-mode: t
  erc-autoaway-mode: t
  gnus-topic-mode: t
  gnus-undo-mode: t
  display-battery-mode: t
  display-time-mode: t
  show-paren-mode: t
  erc-stamp-mode: t
  erc-log-mode: t
  erc-scrolltobottom-mode: t
  mouse-wheel-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  global-auto-composition-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  column-number-mode: t
  line-number-mode: t
  transient-mark-mode: t

Recent input:
SPC i n SPC F e b r u a r y SPC I I R C C-c ESC f g 
L C-s I B E R C-s C-s DEL DEL DEL DEL DEL b e r i RET 
RET ESC [ B RET ESC [ B ESC [ B RET C-c ESC f g C-c 
ESC f i ESC [ D ESC [ D ESC [ D ESC [ D ESC [ D DEL 
DEL DEL DEL DEL DEL DEL DEL l a t e SPC M a r c h C-a 
a n a , SPC RET C-c ESC f g C-x o C-v C-v TAB RET C-g 
C-x o q l C-c ESC f i a n a TAB I SPC g o t SPC s DEL 
t h e s a m DEL DEL DEL SPC s a m e SPC DEL DEL DEL 
DEL DEL DEL DEL DEL DEL DEL DEL DEL DEL f o u n d SPC 
t h e SPC s a m e SPC w h e n SPC I SPC w a n t e d 
SPC t o SPC r e s e r v e : SPC a t SPC 1 6 p DEL P 
M DEL DEL ; DEL : 0 0 SPC h DEL t h e y SPC w a s SPC 
1 7 0 0 C H F , SPC a t SPC 2 0 : 0 0 SPC 1 9 0 0 C 
H F SPC ; SPC - DEL DEL DEL : ( DEL - ( RET C-c ESC 
f g ESC x r e p o r t TAB d e TAB RET C-g C-g ESC x 
r e p o r t - e m TAB b u TAB RET

Recent messages:
Mark saved where search started
Retrieving newsgroup: nnml:support.Iberia...
Fetching headers for nnml:support.Iberia...done
Generating summary...done
No more unread articles
mouse-2: show the MIME part; down-mouse-3: more options
Quit
Exiting summary buffer and applying spam rules
Report a bug for a [P]ackage or [F]ile: (default P) 
Quit
Quit

[-- Attachment #2: Type: application/pgp-signature, Size: 314 bytes --]

bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows

[martin rudalics]

 > However, it seems that `display-buffer' now opens vertical windows
 > instead of horizontal ones.

Come mai ... sorry, but I don't understand what you mean here.  From the
text below I suppose "open" means "split".  But is a horizontal window a
window that spans the entire frame or just a wide window?

 > Romain Francoise discovered that this happens when the width of the
 > window to be split is bigger than its height.  He also privately
 > confirmed me that this bug is still present in the current CVS, so most
 > of the credits for this bug goes to him ;-)

Romain could you please post your observations here.

 > This is clearly reproducible with both `emacs -Q` (GTK+) and `emacs -Q
 > -nw`.

Can you please give me a recipe with emacs -Q so I can reproduce it.

martin

bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows

[Romain Francoise]

Hi martin,

martin rudalics <rudalics@gmx.at> writes:

> Romain could you please post your observations here.

Sure, here goes:

 1) Start Emacs with ./src/emacs -Q -g 80x24
 2) C-x 4 f ~/.emacs
 3) Observe that the frame is split vertically, i.e. the new buffer
    is displayed in a window *below* the scratch buffer.

Now repeat this with 160x24 instead of 80x24.  The frame is now
split horizontally and the new buffer is in a window to the *right*
of the original window.  It's a bug because the size of the frame
should not affect the direction of the default split.

I just bootstrapped from CVS and it still happens.

Hope this clarifies things a bit,
-r

bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows

[martin rudalics]

 >  1) Start Emacs with ./src/emacs -Q -g 80x24
 >  2) C-x 4 f ~/.emacs
 >  3) Observe that the frame is split vertically, i.e. the new buffer
 >     is displayed in a window *below* the scratch buffer.
 >
 > Now repeat this with 160x24 instead of 80x24.  The frame is now
 > split horizontally and the new buffer is in a window to the *right*
 > of the original window.  It's a bug because the size of the frame
 > should not affect the direction of the default split.
 >
 > I just bootstrapped from CVS and it still happens.
 >
 > Hope this clarifies things a bit,

Users can avoid horizontal splitting by setting `split-width-threshold'
to nil.  They also can avoid vertical splitting (but for single-window
frames) by setting `split-height-threshold' to nil.

We can set the default of `split-width-threshold' to nil to get the
behavior you consider correct.  IIRC some people wanted the default
split to be horizontal for very wide frames.  I have no preferences
here.

martin

bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows

[Romain Francoise]

martin rudalics <rudalics@gmx.at> writes:

> Users can avoid horizontal splitting by setting `split-width-threshold'
> to nil.  They also can avoid vertical splitting (but for single-window
> frames) by setting `split-height-threshold' to nil.

Oh, sorry.  I hadn't realized that this is actually a new feature.  :)

> We can set the default of `split-width-threshold' to nil to get the
> behavior you consider correct.  IIRC some people wanted the default
> split to be horizontal for very wide frames.  I have no preferences
> here.

Me neither.  Perhaps we should try to make Emacs 23's default
behavior identical to that of Emacs 22: as this bug report shows,
it's not immediately obvious that the windows get split this way on
purpose, and changing the default split will confuse the user if
she's used to look for the new window at a given place.  Old habits
die hard.

Luca, what do you think?

bug#431: marked as done (23.0.60; `display-buffer' opens vertical instead of horizontal windows)

[Emacs bug Tracking System]

[-- Attachment #1: Type: text/plain, Size: 929 bytes --]


Your message dated Thu, 19 Jun 2008 09:15:23 -0400
with message-id <jwvwsklr2h0.fsf-monnier+emacsbugreports@gnu.org>
and subject line Re: bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows
has caused the Emacs bug report #431,
regarding 23.0.60; `display-buffer' opens vertical instead of horizontal windows
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact don@donarmstrong.com
immediately.)


-- 
431: http://emacsbugs.donarmstrong.com/cgi-bin/bugreport.cgi?bug=431
Emacs Bug Tracking System
Contact don@donarmstrong.com with problems

[-- Attachment #2: Type: message/rfc822, Size: 7000 bytes --]

[-- Attachment #2.1.1: Type: text/plain, Size: 4531 bytes --]


Please write in English if possible, because the Emacs maintainers
usually do not have translators to read other languages for them.

Your bug report will be posted to the emacs-pretest-bug@gnu.org mailing list.

Please describe exactly what actions triggered the bug
and the precise symptoms of the bug:

Hello,

please Cc: me, I'm not subscribed to the list.

In Debian emacs-snapshot_1:20080606-1 screwed up buffers and frames, but
most of those problems are fixed in emacs-snapshot_1:20080613-1.

However, it seems that `display-buffer' now opens vertical windows
instead of horizontal ones.

Romain Francoise discovered that this happens when the width of the
window to be split is bigger than its height.  He also privately
confirmed me that this bug is still present in the current CVS, so most
of the credits for this bug goes to him ;-)

This is clearly reproducible with both `emacs -Q` (GTK+) and `emacs -Q
-nw`.

Thx, bye,
Gismo / Luca

If Emacs crashed, and you have the Emacs process in the gdb debugger,
please include the output from the following gdb commands:
    `bt full' and `xbacktrace'.
If you would like to further debug the crash, please read the file
/usr/share/emacs/23.0.60/etc/DEBUG for instructions.


In GNU Emacs 23.0.60.1 (x86_64-pc-linux-gnu, GTK+ Version 2.12.10)
 of 2008-06-13 on elegiac, modified by Debian
 (emacs-snapshot package, version 1:20080613-1)
configured using `configure  '--build' 'x86_64-linux-gnu' '--host' 'x86_64-linux-gnu' '--prefix=/usr' '--sharedstatedir=/var/lib' '--libexecdir=/usr/lib' '--localstatedir=/var' '--infodir=/usr/share/info' '--mandir=/usr/share/man' '--with-pop=yes' '--enable-locallisppath=/etc/emacs-snapshot:/etc/emacs:/usr/local/share/emacs/23.0.60/site-lisp:/usr/local/share/emacs/site-lisp:/usr/share/emacs/23.0.60/site-lisp:/usr/share/emacs/site-lisp' '--with-x=yes' '--with-x-toolkit=gtk' 'build_alias=x86_64-linux-gnu' 'host_alias=x86_64-linux-gnu' 'CFLAGS=-DDEBIAN -DSITELOAD_PURESIZE_EXTRA=5000 -g -O2' 'LDFLAGS=-g -Wl,--as-needed' 'CPPFLAGS=''

Important settings:
  value of $LC_ALL: nil
  value of $LC_COLLATE: nil
  value of $LC_CTYPE: nil
  value of $LC_MESSAGES: nil
  value of $LC_MONETARY: nil
  value of $LC_NUMERIC: nil
  value of $LC_TIME: nil
  value of $LANG: en_US.UTF-8
  value of $XMODIFIERS: nil
  locale-coding-system: utf-8-unix
  default-enable-multibyte-characters: t

Major mode: Group

Minor modes in effect:
  shell-dirtrack-mode: t
  erc-truncate-mode: t
  erc-track-mode: t
  erc-track-minor-mode: t
  erc-services-mode: t
  erc-networks-mode: t
  erc-ring-mode: t
  erc-netsplit-mode: t
  erc-match-mode: t
  erc-pcomplete-mode: t
  erc-button-mode: t
  erc-fill-mode: t
  erc-autojoin-mode: t
  erc-autoaway-mode: t
  gnus-topic-mode: t
  gnus-undo-mode: t
  display-battery-mode: t
  display-time-mode: t
  show-paren-mode: t
  erc-stamp-mode: t
  erc-log-mode: t
  erc-scrolltobottom-mode: t
  mouse-wheel-mode: t
  file-name-shadow-mode: t
  global-font-lock-mode: t
  font-lock-mode: t
  global-auto-composition-mode: t
  auto-composition-mode: t
  auto-encryption-mode: t
  auto-compression-mode: t
  column-number-mode: t
  line-number-mode: t
  transient-mark-mode: t

Recent input:
SPC i n SPC F e b r u a r y SPC I I R C C-c ESC f g 
L C-s I B E R C-s C-s DEL DEL DEL DEL DEL b e r i RET 
RET ESC [ B RET ESC [ B ESC [ B RET C-c ESC f g C-c 
ESC f i ESC [ D ESC [ D ESC [ D ESC [ D ESC [ D DEL 
DEL DEL DEL DEL DEL DEL DEL l a t e SPC M a r c h C-a 
a n a , SPC RET C-c ESC f g C-x o C-v C-v TAB RET C-g 
C-x o q l C-c ESC f i a n a TAB I SPC g o t SPC s DEL 
t h e s a m DEL DEL DEL SPC s a m e SPC DEL DEL DEL 
DEL DEL DEL DEL DEL DEL DEL DEL DEL DEL f o u n d SPC 
t h e SPC s a m e SPC w h e n SPC I SPC w a n t e d 
SPC t o SPC r e s e r v e : SPC a t SPC 1 6 p DEL P 
M DEL DEL ; DEL : 0 0 SPC h DEL t h e y SPC w a s SPC 
1 7 0 0 C H F , SPC a t SPC 2 0 : 0 0 SPC 1 9 0 0 C 
H F SPC ; SPC - DEL DEL DEL : ( DEL - ( RET C-c ESC 
f g ESC x r e p o r t TAB d e TAB RET C-g C-g ESC x 
r e p o r t - e m TAB b u TAB RET

Recent messages:
Mark saved where search started
Retrieving newsgroup: nnml:support.Iberia...
Fetching headers for nnml:support.Iberia...done
Generating summary...done
No more unread articles
mouse-2: show the MIME part; down-mouse-3: more options
Quit
Exiting summary buffer and applying spam rules
Report a bug for a [P]ackage or [F]ile: (default P) 
Quit
Quit

[-- Attachment #2.1.2: Type: application/pgp-signature, Size: 314 bytes --]

[-- Attachment #3: Type: message/rfc822, Size: 2213 bytes --]

From: Stefan Monnier <monnier@iro.umontreal.ca>
To: 431-close@emacsbugs.donarmstrong.com
Subject: Re: bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows
Date: Thu, 19 Jun 2008 09:15:23 -0400
Message-ID: <jwvwsklr2h0.fsf-monnier+emacsbugreports@gnu.org>

>> Now repeat this with 160x24 instead of 80x24.  The frame is now
>> split horizontally and the new buffer is in a window to the *right*
>> of the original window.

This is the desired default behavior.

>> It's a bug because the size of the frame
>> should not affect the direction of the default split.

Actually it does, now.  If you don't like it, please change
split-width-threshold.


        Stefan

bug#431: closed by Stefan Monnier <monnier@iro.umontreal.ca> (Re: bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows)

[Luca Capello]

[-- Attachment #1: Type: text/plain, Size: 1411 bytes --]

reopen 431
thanks

Hi Stefan!

Sorry for the lag, real life (and work) comes first.

I re-added the people involved in this bug.  However, please Cc: me, I'm
not subscribed to the list nor to the bug.

On Thu, 19 Jun 2008 15:15:23 +02, Stefan Monnier wrote:
>>> Now repeat this with 160x24 instead of 80x24.  The frame is now
>>> split horizontally and the new buffer is in a window to the *right*
>>> of the original window.
>
> This is the desired default behavior.

This is *now* the desired default behavior, but from my POV this is a
*regression*: on Debian emacs22_22.2+2.2 still has the old behavior and
I've some functions which rely on that.

Arbitrarily breaking stuff is not acceptable IMO and this is the reason
why I reopened this bug.

>>> It's a bug because the size of the frame
>>> should not affect the direction of the default split.
>
> Actually it does, now.  If you don't like it, please change
> split-width-threshold.

It could be strange, but I really like the new default behavior: I tend
to have my Emacs buffer at a width < 72, so most of the time I'd like to
split horizontally.

However, what I'd prefer instead is something configurable every time I
want to split a window: there's a default behavior (let it be managed by
`split-width-threshold' or `split-height-threshold', I don't care), but
e.g. if I call C-u C-x 4 f the default behavior is reversed.

Thx, bye,
Gismo / Luca

[-- Attachment #2: Type: application/pgp-signature, Size: 314 bytes --]

bug#431: closed by Stefan Monnier <monnier@iro.umontreal.ca> (Re: bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows)

[Stefan Monnier]

>>>> Now repeat this with 160x24 instead of 80x24.  The frame is now
>>>> split horizontally and the new buffer is in a window to the *right*
>>>> of the original window.
>> This is the desired default behavior.
> This is *now* the desired default behavior, but from my POV this is a
> *regression*: on Debian emacs22_22.2+2.2 still has the old behavior and

Every change in the default behavior will be regression for someone.

> I've some functions which rely on that.

What kind of functions?  What do they look like?  What do they do?

> Arbitrarily breaking stuff is not acceptable IMO and this is the reason
> why I reopened this bug.

This is not arbitrary.

> It could be strange, but I really like the new default behavior: I tend

I'm not surprised: that's why I say the change is not arbitrary.

> However, what I'd prefer instead is something configurable every time I
> want to split a window: there's a default behavior (let it be managed by
> `split-width-threshold' or `split-height-threshold', I don't care), but
> e.g. if I call C-u C-x 4 f the default behavior is reversed.

I see what you mean.  I don't think "C-u C-x 4 f" is a good solution
because it's too specific.  But I've already toyed with some other idea
which would be just as useful: get rid of C-x 4 f and C-x 5 f etc.. and
instead introduce "other-window" and "other-frame" prefix keys, which
would apply to the next command (C-x C-f, C-x b, M-x calendar, M-x info,
...).  In this context, you could easily add "other-vertical-window" and
"other-horizontal-window" prefix keys.

In any case, this is out of the scope of this bug report.  If you want
to discuss such a thing, please move it to some other bug-report or to
emacs-devel.


        Stefan

bug#431: closed by Stefan Monnier <monnier@iro.umontreal.ca> (Re: bug#431: 23.0.60; `display-buffer' opens vertical instead of horizontal windows)

[Luca Capello]

[-- Attachment #1: Type: text/plain, Size: 1851 bytes --]

Hi Stefan!

I'm really sorry for the lag, but I'm not subscribed to the bug nor I
follow the emacs-* lists.  Can you please Cc: me on replies, as I
explicitally asked the last time?

Disclaimer: I haven't checked the emacs sources nor mailing lists to see
if there were other discussions about this subject.

On Thu, 19 Jun 2008 18:56:03 +0200, Stefan Monnier wrote:
>> I've some functions which rely on that.
>
> What kind of functions?  What do they look like?  What do they do?

These are custom functions, e.g.:

--8<---------------cut here---------------start------------->8---
(defun plan-gismo ()
  "Switch to the Planner frame (create it if it doesn't exist)
and show calendar, today plan and the TaskPool."
  (interactive)
  (unless (get-buffer "TaskPool.muse")
    (make-frame '((name . "Planner"))))
  (select-frame-by-name "Planner")
  (delete-other-windows)
  (split-window-vertically)
  (other-window +1)
  (calendar)
  (other-window +1)
  (split-window-vertically)
  (other-window +1)
  (plan))
--8<---------------cut here---------------end--------------->8---

(calendar) is the first source of problem if the window is vertically
splitted: the so-created buffer is empty for most of its vertical space.

> In any case, this is out of the scope of this bug report.  If you want
> to discuss such a thing, please move it to some other bug-report or to
> emacs-devel.

Sure, when I'll find the time to hack on it by myself and propose my
solution.  I didn't intend to discuss not-related improvements, instead
I was explaining why IMHO a general, i.e. not per command/action,
behavior is wrong.

Now, going back to the original bug, since it seems I'm the only one who
finds it a regression, you can close it.  Until I won't have coded my
proposal, I already have (setq split-width-threshold nil) in my
~/.emacs.

Thx, bye,
Gismo / Luca

[-- Attachment #2: Type: application/pgp-signature, Size: 314 bytes --]