Geometry and temp-buffer-max-height

Geometry and temp-buffer-max-height

[Andrea Vettorello]

I've a trouble setting "temp-buffer-max-height" under X and Emacs 22.1.

If i start Emacs with a geometry parameter, via .Xdefaults or passing
the option from the command line, in the first created frame
(initial-frame) the value assigned to "temp-buffer-max-height" is not
applied, i.e. the temp buffer height spans the default half frame.
This doesn't happen if i resize the frame before invoking a temp
buffer (for example using the help) or if inside a new frame (C-x 5
2).

I've looked the documentation about "initial-frame-alist" and
"default-frame-alist" but i didn't found nothing interesting.

It's a bug or i'm missing something, like some other variable to poke?


-- 
Andrea

Re: Geometry and temp-buffer-max-height

[Kevin Rodgers]

Andrea Vettorello wrote:
> I've a trouble setting "temp-buffer-max-height" under X and Emacs 22.1.
> 
> If i start Emacs with a geometry parameter, via .Xdefaults or passing
> the option from the command line, in the first created frame
> (initial-frame) the value assigned to "temp-buffer-max-height" is not
> applied, i.e. the temp buffer height spans the default half frame.
> This doesn't happen if i resize the frame before invoking a temp
> buffer (for example using the help) or if inside a new frame (C-x 5
> 2).
> 
> I've looked the documentation about "initial-frame-alist" and
> "default-frame-alist" but i didn't found nothing interesting.
> 
> It's a bug or i'm missing something, like some other variable to poke?

The documentation for temp-buffer-max-height seems to suggest that you
should call (temp-buffer-max-height 1) to ensure that
temp-buffer-max-height is effective.

-- 
Kevin Rodgers
Denver, Colorado, USA

Re: Geometry and temp-buffer-max-height

[Andrea Vettorello]

On 7/31/07, Kevin Rodgers <kevin.d.rodgers@gmail.com> wrote:


> The documentation for temp-buffer-max-height seems to suggest that you
> should call (temp-buffer-max-height 1) to ensure that
> temp-buffer-max-height is effective.

If it's a function to call i can't find it, for the variable with the
same name, i've tried to set it both as a fixed or a variable value
(2/3 of the current frame, as i've found a suggestion somewhere).

If you mean the "temp-buffer-resize-mode" variable, then is set to
true. I thought about a bug because this happens only when i pass a
geometry option, via command line or ~/.Xdefaults, but it's not a big
deal, i let the ini files resizing the frame and i'm done with it...


-- 
Andrea

Re: Geometry and temp-buffer-max-height

[Kevin Rodgers]

Andrea Vettorello wrote:
> On 7/31/07, Kevin Rodgers <kevin.d.rodgers@gmail.com> wrote:
> 
> 
>> The documentation for temp-buffer-max-height seems to suggest that you
>> should call (temp-buffer-max-height 1) to ensure that
>> temp-buffer-max-height is effective.
> 
> If it's a function to call i can't find it, for the variable with the
> same name, i've tried to set it both as a fixed or a variable value
> (2/3 of the current frame, as i've found a suggestion somewhere).
> 
> If you mean the "temp-buffer-resize-mode" variable, then is set to
> true. I thought about a bug because this happens only when i pass a
> geometry option, via command line or ~/.Xdefaults, but it's not a big
> deal, i let the ini files resizing the frame and i'm done with it...

It's a function:

,----[ C-h f temp-buffer-resize-mode RET ]
| temp-buffer-resize-mode is an interactive compiled Lisp function in 
`help.el'.
| (temp-buffer-resize-mode &optional arg)
|
| Toggle the mode which makes windows smaller for temporary buffers.
| With prefix argument arg, turn the resizing of windows displaying 
temporary
| buffers on if arg is positive or off otherwise.
| This makes the window the right height for its contents, but never
| more than `temp-buffer-max-height' nor less than `window-min-height'.
| This applies to `help', `apropos' and `completion' buffers, and some 
others.
|
| [back]
`----

-- 
Kevin Rodgers
Denver, Colorado, USA

Re: Geometry and temp-buffer-max-height

[Andrea Vettorello]

On 8/2/07, Kevin Rodgers <kevin.d.rodgers@gmail.com> wrote:
> Andrea Vettorello wrote:
> > On 7/31/07, Kevin Rodgers <kevin.d.rodgers@gmail.com> wrote:
> >
> >
> >> The documentation for temp-buffer-max-height seems to suggest that you
> >> should call (temp-buffer-max-height 1) to ensure that
> >> temp-buffer-max-height is effective.
> >
> > If it's a function to call i can't find it, for the variable with the
> > same name, i've tried to set it both as a fixed or a variable value
> > (2/3 of the current frame, as i've found a suggestion somewhere).
> >
> > If you mean the "temp-buffer-resize-mode" variable, then is set to
> > true. I thought about a bug because this happens only when i pass a
> > geometry option, via command line or ~/.Xdefaults, but it's not a big
> > deal, i let the ini files resizing the frame and i'm done with it...
>
> It's a function:
>
> ,----[ C-h f temp-buffer-resize-mode RET ]
> | temp-buffer-resize-mode is an interactive compiled Lisp function in
> `help.el'.
> | (temp-buffer-resize-mode &optional arg)
> |
> | Toggle the mode which makes windows smaller for temporary buffers.
> | With prefix argument arg, turn the resizing of windows displaying
> temporary
> | buffers on if arg is positive or off otherwise.
> | This makes the window the right height for its contents, but never
> | more than `temp-buffer-max-height' nor less than `window-min-height'.
> | This applies to `help', `apropos' and `completion' buffers, and some
> others.
> |
> | [back]
> `----

Nope, even explicitly setting this one doesn't change it. Maybe it's
only in the Debian build, i haven't checked from CVS/SVN, but as i've
said above, it's a small glitch that shows up if i use the "geometry"
option, i can easily workaround it.

Thanks anyway for your help.


-- 
Andrea

Re: Geometry and temp-buffer-max-height

[martin rudalics]

> Nope, even explicitly setting this one doesn't change it. Maybe it's
> only in the Debian build, i haven't checked from CVS/SVN, but as i've
> said above, it's a small glitch that shows up if i use the "geometry"
> option, i can easily workaround it.

Could you please tell here whether the window is resized at all with
`temp-buffer-resize-mode' activated?  WOW, is the bug you observe a bug
where only the value of `temp-buffer-max-height' is not respected, or is
it a bug where `temp-buffer-resize-mode' is not applied?

Re: Geometry and temp-buffer-max-height

[Andrea Vettorello]

On 8/2/07, martin rudalics <rudalics@gmx.at> wrote:

> Could you please tell here whether the window is resized at all with
> `temp-buffer-resize-mode' activated?  WOW, is the bug you observe a bug
> where only the value of `temp-buffer-max-height' is not respected, or is
> it a bug where `temp-buffer-resize-mode' is not applied?


I've tried with a different user and the value
"temp-buffer-max-height" is applied, so it's a local config issue, aka
my fault.

Sorry for the noise... (=


-- 
Andrea

Re: Geometry and temp-buffer-max-height

[Andrea Vettorello]

On 8/2/07, Andrea Vettorello <andrea.vettorello@gmail.com> wrote:

> I've tried with a different user and the value
> "temp-buffer-max-height" is applied, so it's a local config issue, aka
> my fault.

Ah, not my fault. (=

I've isolated the conflicting parameter, modifying the fringe exhibit
the problem.

This was reported today on the bug mailing list:
http://lists.gnu.org/archive/html/bug-gnu-emacs/2007-08/msg00007.html


-- 
Andrea

Re: Geometry and temp-buffer-max-height

[martin rudalics]

 > I've isolated the conflicting parameter, modifying the fringe exhibit
 > the problem.
 >
 > This was reported today on the bug mailing list:
 > http://lists.gnu.org/archive/html/bug-gnu-emacs/2007-08/msg00007.html

Then you probably read my answer too.  Can you please now confirm that
the window is _not_ resized on your "local configuration" when you have
`temp-buffer-resize-mode' enabled and `temp-buffer-max-height' is left
at the default value.

Re: Geometry and temp-buffer-max-height

[Andrea Vettorello]

On 8/2/07, martin rudalics <rudalics@gmx.at> wrote:

 > Then you probably read my answer too.  Can you please now confirm that
> the window is _not_ resized on your "local configuration" when you have
> `temp-buffer-resize-mode' enabled and `temp-buffer-max-height' is left
> at the default value.
>

I confirm it, the temp buffer doesn't apply the
"temp-buffer-max-height" if i pass the geometry option, from command
line or .Xdefaults/.Xrerources, and if the fringe-mode is used in any
other way than the default.

If useful, i can post my Emacs init files.


-- 
Andrea

Re: Geometry and temp-buffer-max-height

[martin rudalics]

>> Can you please now confirm that
>> the window is _not_ resized on your "local configuration" when you have
>> `temp-buffer-resize-mode' enabled and `temp-buffer-max-height' is left
>> at the default value.
>>
>
> I confirm it, the temp buffer doesn't apply the
> "temp-buffer-max-height" if i pass the geometry option, from command
> line or .Xdefaults/.Xrerources, and if the fringe-mode is used in any
> other way than the default.

That doesn't answer my question but let's ignore this for the moment.
Could you please add the lines below at the end of your .emacs file,
restart Emacs, and tell me whether the problem persists (this is by no
means the correct fix but merely a temporary workaround):

(require 'help)

(defun resize-temp-buffer-window ()
   ""
   (unless (or (one-window-p 'nomini)
               (not (pos-visible-in-window-p (point-min)))
               (> (frame-width) (window-width)))
     (fit-window-to-buffer
      (selected-window)
      (if (functionp temp-buffer-max-height)
	 (funcall temp-buffer-max-height (current-buffer))
        temp-buffer-max-height))))