Starting emacs without initial frame

Starting emacs without initial frame

[Rainer Trusch]

Hi,

Is there a way to start emacs without an initial frame like the
--unmapped option in xemacs?


Cheers

Rainer

Re: Starting emacs without initial frame

[David Kastrup]

Rainer Trusch <rainer.trusch@students.uni-mainz.de> writes:

> Is there a way to start emacs without an initial frame like the
> --unmapped option in xemacs?

Given that Emacs does not have multiple-tty capability right now,
there would be little sense in putting such an option in now: Emacs
would not be able to guess which tty was going to ask for it first.
So an unmapped Emacs does not make sense at the current point of time.

Merging the multi-tty branch is slated for after the release of Emacs
22.1.  I think that this would then be reason to incorporate some
functionality of gnuclient, and provide such an option.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Starting emacs without initial frame

[Stefan Monnier]

> Is there a way to start emacs without an initial frame like the
> --unmapped option in xemacs?

Not that I know.  You can make the initial frame unmapped, but it'll be
unmapped only after the .emacs is read, so it'll temporarily appear.


        Stefan

Re: Starting emacs without initial frame

[Miles Bader]

Stefan Monnier <monnier@iro.umontreal.ca> writes:
>> Is there a way to start emacs without an initial frame like the
>> --unmapped option in xemacs?
>
> Not that I know.  You can make the initial frame unmapped, but it'll be
> unmapped only after the .emacs is read, so it'll temporarily appear.

Kind of stupid that, isn't it?  I vaguely recall that Gerd successfully
changed Emacs to map the initial frame only _after_ .emacs was read, but
people who did weird things in their .emacs complained so this eminently
reasonable behavior was dropped for "compatibility".

It seems to me that it would much nicer to have Emacs map the frame only
after .emacs is read, and add a (map-frame) function or something that
people could use to force it to map earlier.

-Miles
-- 
"Whatever you do will be insignificant, but it is very important that
 you do it."  Mahatma Ghandi

Re: Starting emacs without initial frame

[David Kastrup]

Miles Bader <miles@gnu.org> writes:

> Stefan Monnier <monnier@iro.umontreal.ca> writes:
>>> Is there a way to start emacs without an initial frame like the
>>> --unmapped option in xemacs?
>>
>> Not that I know.  You can make the initial frame unmapped, but it'll be
>> unmapped only after the .emacs is read, so it'll temporarily appear.
>
> Kind of stupid that, isn't it?  I vaguely recall that Gerd
> successfully changed Emacs to map the initial frame only _after_
> .emacs was read, but people who did weird things in their .emacs
> complained so this eminently reasonable behavior was dropped for
> "compatibility".
>
> It seems to me that it would much nicer to have Emacs map the frame
> only after .emacs is read, and add a (map-frame) function or
> something that people could use to force it to map earlier.

At the current point of time, the only sensible way of turning the
toolbar off (and similar manipulations affecting display geometry) is
by manipulating X resources.  All other options will lead to window
resizes, meaning that the chosen startup frame layout is wrong.
Basically most of the "Hide/Show" menu is affected.

Mapping late would help a lot.

-- 
David Kastrup, Kriemhildstr. 15, 44793 Bochum

Re: Starting emacs without initial frame

[Rainer Trusch]

Hi David,

On 18 Mär 2005, David Kastrup wrote:
> Rainer Trusch <rainer.trusch@students.uni-mainz.de> writes:
>
>> Is there a way to start emacs without an initial frame like the
>> --unmapped option in xemacs?
>
> Given that Emacs does not have multiple-tty capability right now,
> there would be little sense in putting such an option in now: Emacs
> would not be able to guess which tty was going to ask for it first.
> So an unmapped Emacs does not make sense at the current point of
> time.
>
> Merging the multi-tty branch is slated for after the release of
> Emacs 22.1.  I think that this would then be reason to incorporate
> some functionality of gnuclient, and provide such an option.

Thanks for the info. I have to figure out how to deal with it.


Rainer

Re: Starting emacs without initial frame

[Stefan Monnier]

> Kind of stupid that, isn't it?  I vaguely recall that Gerd successfully
> changed Emacs to map the initial frame only _after_ .emacs was read, but
> people who did weird things in their .emacs complained so this eminently
> reasonable behavior was dropped for "compatibility".

My own local Emacs only creates the initial frame after reading .emacs ;-)
It indeed creates a few incompatibilities (e.g. startup.el can't properly
set the default of normal-erase-is-backspace before reading the .emacs
because it needs to query the X server for that, but the connection is not
yet opened).

> It seems to me that it would be much nicer to have Emacs map the frame
> only after .emacs is read, and add a (map-frame) function or something
> that people could use to force it to map earlier.

People may be annoyed at the "loading foo..." messages on stdout, tho.
Also my code currently breaks --debug-init.


        Stefan

Re: Starting emacs without initial frame

[Miles Bader]

Stefan Monnier <monnier@iro.umontreal.ca> writes:
>> Kind of stupid that, isn't it?  I vaguely recall that Gerd successfully
>> changed Emacs to map the initial frame only _after_ .emacs was read, but
>> people who did weird things in their .emacs complained so this eminently
>> reasonable behavior was dropped for "compatibility".
>
> My own local Emacs only creates the initial frame after reading .emacs ;-)
> It indeed creates a few incompatibilities (e.g. startup.el can't properly
> set the default of normal-erase-is-backspace before reading the .emacs
> because it needs to query the X server for that, but the connection is not
> yet opened).

Cool.  Obviously not for this release, but how about trying to get those
changes committed after the release?

> People may be annoyed at the "loading foo..." messages on stdout, tho.
> Also my code currently breaks --debug-init.

Those sound like the sort of detail that could be sorted out easily
enough though...  [personally I always find those initial "loading..."
messages annoyinga anyway; I think they should just be turned off unless
--debug-init etc.]

[Maybe there needs to be a queue of callbacks that is run when the
display finally shows up.]

-Miles
-- 
(\(\
(^.^)
(")")
*This is the cute bunny virus, please copy this into your sig so it can spread.

Re: Starting emacs without initial frame

[Cristian Gutierrez]

Este viernes, Rainer Trusch dijo:
> Hi,

Hello,

> Is there a way to start emacs without an initial frame like the
> --unmapped option in xemacs?

Not exactly that, but you can start it under Xvfb (or vnc, or similar)
in the background, and then get a frame in your current display by
calling `make-frame-on-display' (either by putting it at the end of your
.emacs or by using gnuserv/gnuclient).

Works well so far, and it makes Emacs survive X crashes :)

-- 
Cristian Gutierrez			http://www.dcc.uchile.cl/~crgutier
crgutier[@]dcc.uchile.cl                        Jabber:crgutier@jabber.org

To err is human, but to really foul things up requires a computer.