TongKe Xue | 1 Jan 2003 07:19
Picon
Favicon

Having workspaces/frames load + unload with applications.

Greetings!
 Happy New Year!

  Any one know if it's possible to have ION
load/unload
workspaces + frames with appliations? For example, I
have
an "GIMP" workspace, with a main frame and numerous
small
frames containing the various toolboxes. However, I
don't
want this frame when most of the time (when I switch
between
workspaces with forward/backward, I want the least
distance).
Instead, I want the GIMP frame to load when I run
GIMP, and
to unload when I unload GIMP.

  Of course, GIMP probably doesn't have function calls
to
communicate to ion for this -- is it possible to
create a
script like loadGimp; which would create the new
workspaces,
and then run Gimp?

  Hints/Ideas would be very welcome and apperciated.

Thanks in advance,
(Continue reading)

Per Olofsson | 1 Jan 2003 10:43
Picon
Picon
Gravatar

Re: [Q] Is Emacs slow with ion-devel ?

On Tue, Dec 31, 2002 at 08:36:59PM +0100, Xavier MAILLARD wrote:
> 
> I don't know if I can tell here but Xft compilation doesn't work at all
> here with my Debian Sid (freshly updated). There are many compilation
> errors.
> 
> Did you try to enable/compile it ?
>   

I tried now and compilation worked fine. With my package, that
is. What compilation errors do you get? If you didn't use my package
you'll have to change -pedantic-errors to -pedantic in system.mk since
it breaks on /usr/include/regex.h.

/Pelle
Xavier MAILLARD | 1 Jan 2003 18:59
X-Face
Picon

Re: [Q] Is Emacs slow with ion-devel ?

On Wed, 1 Jan 2003, Per Olofsson said:

On Tue, Dec 31, 2002 at 08:36:59PM +0100, Xavier MAILLARD wrote: >
> > I don't know if I can tell here but Xft compilation doesn't work at
> > all here with my Debian Sid (freshly updated). There are many
> > compilation errors.
> > 
> > Did you try to enable/compile it ?
> >   
>  
>  I tried now and compilation worked fine. With my package, that
>  is. What compilation errors do you get? If you didn't use my package
>  you'll have to change -pedantic-errors to -pedantic in system.mk
>  since it breaks on /usr/include/regex.h.

Yup, I actually try to build against your package  decommenting Xft
stuff from system.mk.

This is the output of the errors:

,----
| 
| gcc -g -O2 -W -Wimplicit -Wreturn-type -Wswitch -Wcomment -Wtrigraphs -Wformat -Wchar-subscripts
-Wparentheses -pedantic -Wuninitialized -DCF_XFT -DETCDIR=\"/etc/X11\" -I../libtu/include
-I/usr/X11R6/include -I..    -c binding.c -o binding.o
| In file included from common.h:20,
|                  from binding.c:9:
| /usr/X11R6/include/X11/Xft/Xft.h:52: parse error before `_XftFTlibrary'
| /usr/X11R6/include/X11/Xft/Xft.h:52: warning: type defaults to `int' in declaration of `_XftFTlibrary'
| /usr/X11R6/include/X11/Xft/Xft.h:52: ANSI C forbids data definition with no type or storage class
(Continue reading)

Xavier MAILLARD | 1 Jan 2003 19:03
X-Face
Picon

Re: Having workspaces/frames load + unload with applications.

On Tue, 31 Dec 2002, TongKe Xue told this:

>  Greetings!
>   Happy New Year!

:) Happy new Year too :)

>    Any one know if it's possible to have ION
>  load/unload
>  workspaces + frames with appliations? For example, I
>  have
>  an "GIMP" workspace, with a main frame and numerous
>  small
>  frames containing the various toolboxes. However, I
>  don't
>  want this frame when most of the time (when I switch
>  between
>  workspaces with forward/backward, I want the least
>  distance).
>  Instead, I want the GIMP frame to load when I run
>  GIMP, and
>  to unload when I unload GIMP.
>  
>    Of course, GIMP probably doesn't have function calls
>  to
>  communicate to ion for this -- is it possible to
>  create a
>  script like loadGimp; which would create the new
>  workspaces,
>  and then run Gimp?
(Continue reading)

TongKe Xue | 1 Jan 2003 19:57
Picon
Favicon

Re: Having workspaces/frames load + unload with applications.


<Cutting & Pasting from response>
> AFAIK I don't know how this would be possible :) But
> it can be a new
> feature to implement. Maybe you can develop this for
> us :)
</cutting & pasting from response>

If someone's already posted this idea, sorry for the
repost (I haven't read through all of the archives).

Perhaps, in that case, it would be better to develop a
generalized "scripting" program. The program could be
called something like "ion-manip-frames", called with
say, "--dump", it would return all workspaces, frames,
and the current frame the user was in, all IDed, with
all of the names/classes/PIDs of the programs running
within the frames.

Then, called with "--create ID DIR", it would split
frame/workspace with ID blah into frames horizontally
or vertically. And so forth, with "--blah" for each
generalized form of function calls within ion.

BTW, I know some C/C++, but haven't taken a look at
the ion-devel code yet (other than the README and
Makefile), so I might be able to write a patch, but
it'll be a while. ;)

--TongKe
(Continue reading)

Per Olofsson | 1 Jan 2003 22:57
Picon
Picon
Gravatar

Re: [Q] Is Emacs slow with ion-devel ?

On Wed, Jan 01, 2003 at 06:59:15PM +0100, Xavier MAILLARD wrote:
> 
> Yup, I actually try to build against your package  decommenting Xft
> stuff from system.mk.
> 
> This is the output of the errors:
> 

-snip-

> This is the list of the packages installed on my box about Xft related
> packages:
> 
> ,----
> | ii  libxft2                         2.1-5                           advanced font drawing library for X
> | ii  libxft2-dev                     2.1-5                           advanced font drawing library for X (development files)
> `----
> 

The problem is that you have libxft2-dev installed. This package
replaces (using a diversion) the header file
/usr/X11R6/include/X11/Xft/Xft.h which normally belongs to xlibs-dev
which contains Xft version 1. Xft2 expects header files from
freetype2, which are not used by default when <freetype/freetype.h> is
included. Anyway, you have these options:

1. Remove libxft2-dev. The diversion will be removed and Xft1 will be
used.

2. Change all references to Xft.h in the Ion source code to Xft1.h
(Continue reading)

Xavier MAILLARD | 2 Jan 2003 06:16
X-Face
Picon

Re: [Q] Is Emacs slow with ion-devel ?

On Wed, 1 Jan 2003, Per Olofsson spake thusly:

On Wed, Jan 01, 2003 at 06:59:15PM +0100, Xavier MAILLARD wrote: >
> > Yup, I actually try to build against your package  decommenting Xft
> > stuff from system.mk.
> > 
> > This is the output of the errors:
> > 
>  
>  -snip-
>  
> > This is the list of the packages installed on my box about Xft
> > related packages:
> > 
> > ,----
> > | ii libxft2 2.1-5 advanced font drawing library for X ii
> > | libxft2-dev 2.1-5 advanced font drawing library for X (development
> > | files)
> > `----
> > 
>  
>  The problem is that you have libxft2-dev installed. This package
>  replaces (using a diversion) the header file
>  /usr/X11R6/include/X11/Xft/Xft.h which normally belongs to xlibs-dev
>  which contains Xft version 1. Xft2 expects header files from
>  freetype2, which are not used by default when <freetype/freetype.h>
>  is included. Anyway, you have these options:
>  
>  1. Remove libxft2-dev. The diversion will be removed and Xft1 will be
>  used.
(Continue reading)

Xavier MAILLARD | 3 Jan 2003 06:02
X-Face
Picon

[OT]: I've been hacked :(


Hi,

I know this is off topic but I have to apologize for the mess I have
done tonight and maybe the nights before.

In fact someone unfair has hacked my box such a way that whenever I go
onto a site, a program  is started on each URL I visit and try
to copy the whole web site tree on my box.

I don't really know how long it has occur, but I have detected it
reading my logs and also when I saw that my FSes were fullfilled :(

One more time, I apologize. The problem has been fixed for now and
shouldn't occur in the future.

The targetted web site was the 'ion wiki' and AFAIK, the site should
have been disturbed during this period.

Best regards,

zeDek
Konstantin Kletschke | 3 Jan 2003 17:05
Picon

crash of ion? was: can wm cause sig11?

Hi there!

I asked some time before if my Xserver crashed or (only) ion itself.
Somebody said, his ion-devel did not crash ever. So I am really
irritated. Today it happened again.
Another geek advised me to add an "ulimit -c 1000000" to my .xinitrc to
get a corefile, but it did not work :-( No corefile on the whole
filesystem :-) 

Is there an error?

cat .xinitrc
xplanetbg -projection hemisphere & ulimit -c 1000000 & exec ion

So, only Thing is what console was telling me afterwards:
Improper call to JPEG library in state 205
Improper call to JPEG library in state 205
Fatal error loading PNG: Too much data in IDAT chunks
libpng error: Too much data in IDAT chunks
GetModeLine - scrn: 0 clock: 148500
GetModeLine - hdsp: 2432 hbeg: 1344 hend: 1504 httl: 1728
              vdsp: 1400 vbeg: 961 vend: 964 vttl: 1011 flags: 0
GetModeLine - scrn: 0 clock: 148500
GetModeLine - hdsp: 2432 hbeg: 1344 hend: 1504 httl: 1728
              vdsp: 1400 vbeg: 961 vend: 964 vttl: 1011    flags: 0
ion: Caught fatal signal 11. Dying without deinit.
waiting for X server to shut down Gdk-ERROR **: X connection to :0.0 broken (explicit kill or server shutdown).
Gdk-ERROR **: X connection to :0.0 broken (explicit kill or server shutdown).
The application 'gkrellm2' lost its connection to the display :0.0;
most likely the X server was shut down or you killed/destroyed the application.
(Continue reading)

Paul Jarc | 3 Jan 2003 17:59
Picon

Re: crash of ion? was: can wm cause sig11?

Konstantin Kletschke <konsti <at> ludenkalle.de> wrote:
> Is there an error?
>
> cat .xinitrc
> xplanetbg -projection hemisphere & ulimit -c 1000000 & exec ion

ulimit should run in the foreground.  Otherwise it will have no effect
on the following commands, because it will run in its own short-lived
subprocess, and its only purpose is to change the state of the process
it's in.
xplanetbg -projection hemisphere &
ulimit -c 1000000
exec ion

paul

Gmane