From aderuwe at aqs-carcontrol.be  Wed Oct  2 04:54:56 2002
From: aderuwe at aqs-carcontrol.be (Alexander Deruwe)
Date: Wed, 2 Oct 2002 10:54:56 +0200
Subject: Anti-alias problem
Message-ID: <20021002085456.GD6506@nbalexander.aqs-carcontrol.be>

Hey all,

I installed openbox today, and only then discovered that it did
anti-aliasing. Great feature. ;)
But when I restarted openbox (changed .xinitrc from just 'openbox' to
'ssh-agent openbox'), anti-aliasing didn't work anymore, even though it
is turned on.
Any ideas?

Thanks!

-- 
Alexander Deruwe
AQS-CarControl


From dbr at informactis.com  Wed Oct  2 08:09:12 2002
From: dbr at informactis.com (Didier Bretin)
Date: Wed, 2 Oct 2002 14:09:12 +0200
Subject: [openbox] problem for compiling the 2.1.0 version
In-Reply-To: <20020930150312.GC3614@marilyn>
References: <20020920093004.26a4ed34.didier@bretin.net>
	<20020927230834.GA9914@orodu.net>
	<20020930154418.7f6bd194.dbr@informactis.com>
	<20020930150312.GC3614@marilyn>
Message-ID: <20021002140912.113dee3e.dbr@informactis.com>

On Mon, 30 Sep 2002 11:03:12 -0400
Scott Moynes <smoynes at nexus.carleton.ca> wrote:

> * Didier Bretin (dbr at informactis.com) wrote:
> > Do I forget something ?
> 
> run ./boostrap.sh
> -- 
> Scott Moynes http://www.icculus.org/openbox/
> "Computer science is as much about computers
> as astronomy is about telescopes." -- Dijkstra

Cool,

  The compilation is OK :o).
  Thanks.

But I can't compile the 2.1.1 version. Same problem of the 2.1.0 version :o).

Regards.
-- 
            .------------------------------------------------.
    .^.     | Didier Bretin, France | dbr at informactis.com    |
    /V\     |-----------------------| www.informactis.com    |
   // \\    |                       `------------------------|
  /(   )\   | Visit: http://www.vim.org/                     |
   ^^-^^    `------------------------------------------------'


From smoynes at nexus.carleton.ca  Thu Oct  3 12:18:11 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Thu, 3 Oct 2002 12:18:11 -0400
Subject: CVS status
Message-ID: <20021003161811.GA27894@marilyn>

Okay,

I've finally committed my changes to HEAD.

First, there's options to set the bitmaps on the window buttons. I also
modified the man page to reflect this, so check that out if you want to use
it. Basically, you create an xbm file (gimp can do this, or you can create
one using a text editor if you want), set some resources in the rc or in the
style, and you get button that look like a baby in the bathtub.

Next, I've added style resources to change the look of a pressed, unfocused
button. To see an example of this, you can vertically maximize the default
nyzclone theme and unfocus. I haven't updated the documentation, but here's
the deal.
There are 2 new resources:
window.button.pressed.focus and window.button.pressed.unfocus and they are
textures just like the rest. If either are defined, they are used that widget.
If one of them is not defined, it inherits from window.button.pressed. This
allows both forward and backward compatibility of styles.

There is an example screenshot of these "features" in action at 
http://nexus.carleton.ca/~smoynes/pressedbuttons.png
and the style used is http://nexus.carleton.ca/~smoynes/nyzclone
Well, there will be when my host decides to contract Al Gore to connect them
to the interweb.

Any questions, comments, or pies, feel free to send them my way.
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021003/79f0a5ec/attachment.pgp>

From smoynes at nexus.carleton.ca  Fri Oct  4 00:11:54 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Fri, 4 Oct 2002 00:11:54 -0400
Subject: Testing 2.2.0 (clear) released
Message-ID: <20021004041154.GD26343@marilyn>

We've, finally, released a new testing branch. You can find it at
http://icculus.org/openbox/download.php
New in this release the aformentioned:
- bitmap buttons
- pressed button focus/unfocus style resources
So, if anyone wants to make styles that take advantage of these
features, send them my way. There's still quite a few things I'd like
to add to this branch, so stayed tuned.

Peace, love, and cotton candy.
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021004/e6ec65ae/attachment.pgp>

From james at grickle.org  Fri Oct  4 17:00:11 2002
From: james at grickle.org (James Harr)
Date: Fri, 4 Oct 2002 16:00:11 -0500 (CDT)
Subject: KDE configure options
Message-ID: <32960.204.182.60.2.1033765211.squirrel@webmail.grickle.org>

I noticed on some of the screenshots that you were running kde, but with
openbox as the window manager.

I was wandering what steps you took to do that. I know in gnome you can
select it in the config program, but I saw nothing like that in kde.

Later,
James Harr <james at grickle.org>




From marius at cs.pdx.edu  Fri Oct  4 17:56:29 2002
From: marius at cs.pdx.edu (Marius Nita)
Date: Fri, 4 Oct 2002 14:56:29 -0700
Subject: [openbox] KDE configure options
In-Reply-To: <32960.204.182.60.2.1033765211.squirrel@webmail.grickle.org>
References: <32960.204.182.60.2.1033765211.squirrel@webmail.grickle.org>
Message-ID: <20021004215629.GA20258@cs.pdx.edu>

On Fri, Oct 04, 2002 at 04:00:11PM -0500, James Harr wrote:
> I noticed on some of the screenshots that you were running kde, but with
> openbox as the window manager.
> 
> I was wandering what steps you took to do that. I know in gnome you can
> select it in the config program, but I saw nothing like that in kde.


From james at grickle.org  Fri Oct  4 14:58:56 2002
From: james at grickle.org (James Harr)
Date: 04 Oct 2002 13:58:56 -0500
Subject: [openbox] KDE configure options
In-Reply-To: <20021004215629.GA20258@cs.pdx.edu>
References: <32960.204.182.60.2.1033765211.squirrel@webmail.grickle.org> 
	<20021004215629.GA20258@cs.pdx.edu>
Message-ID: <1033757936.19207.1.camel@Zeus>

Thanks,

I think I'll use my .xinitrc file instead of modifying it globally for
everyone. I thought KDE had an option for window managers like gnome
does. You don't think it would be all that hard to implement..

Hopefully they do sometime in the future.

Thanks,
James Harr

On Fri, 2002-10-04 at 16:56, Marius Nita wrote:
> On Fri, Oct 04, 2002 at 04:00:11PM -0500, James Harr wrote:
> > I noticed on some of the screenshots that you were running kde, but with
> > openbox as the window manager.
> > 
> > I was wandering what steps you took to do that. I know in gnome you can
> > select it in the config program, but I saw nothing like that in kde.
> 
> From /usr/bin/startkde, toward the end of the file
> 
> # finally, give the session control to the session manager
> # Syntax:   ksmserver [--restore] [--windowmanager <wm>]
> # if no windowmanager is specified, ksmserver will ensure kwin is started.
> # [--restore] should be controlled by kdm
> # kwrapper is used to reduce startup time and memory usage
> kwrapper ksmserver --restore || \
>   xmessage -geometry 500x100 "Couldn't start ksmserver. Check your
>   installation."
> 
> 
> so add --windowmanager openbox to that line.
> 
> A lot of users, however, prefer not to run kde's fullblown services, and they
> run kicker and/or kdesktop separately, together with openbox. Whatever works
> for you.
> 
> > Later,
> > James Harr <james at grickle.org>
> > 




From ben at orodu.net  Fri Oct  4 20:03:01 2002
From: ben at orodu.net (Ben Jansens)
Date: Fri, 4 Oct 2002 19:03:01 -0500
Subject: [openbox] KDE configure options
In-Reply-To: <1033757936.19207.1.camel@Zeus>
References: <32960.204.182.60.2.1033765211.squirrel@webmail.grickle.org> <20021004215629.GA20258@cs.pdx.edu> <1033757936.19207.1.camel@Zeus>
Message-ID: <20021005000301.GA26054@orodu.net>

On Fri, Oct 04, 2002 at 01:58:56PM -0500, James Harr wrote:
> Thanks,
> 
> I think I'll use my .xinitrc file instead of modifying it globally for
> everyone. I thought KDE had an option for window managers like gnome
> does. You don't think it would be all that hard to implement..
> 
> Hopefully they do sometime in the future.
> 

The KDE developers see no reason why everyone doesnt run kwin, and so
I don't think they will add this. In KDE 3.1 tho, youll be able to set
the KDEWM environment variable and kde will use that if it is set
instead of kwin.

personally, ive found the best is to put this in my .xinitrc

LD_BIND_NOW=true kdeinit +kcminit +knotify
kicker & korgac & klipper &
epist & exec openbox

> On Fri, 2002-10-04 at 16:56, Marius Nita wrote:
> > On Fri, Oct 04, 2002 at 04:00:11PM -0500, James Harr wrote:
> > > I noticed on some of the screenshots that you were running kde, but with
> > > openbox as the window manager.
> > > 
> > > I was wandering what steps you took to do that. I know in gnome you can
> > > select it in the config program, but I saw nothing like that in kde.
> > 
> > From /usr/bin/startkde, toward the end of the file
> > 
> > # finally, give the session control to the session manager
> > # Syntax:   ksmserver [--restore] [--windowmanager <wm>]
> > # if no windowmanager is specified, ksmserver will ensure kwin is started.
> > # [--restore] should be controlled by kdm
> > # kwrapper is used to reduce startup time and memory usage
> > kwrapper ksmserver --restore || \
> >   xmessage -geometry 500x100 "Couldn't start ksmserver. Check your
> >   installation."
> > 
> > 
> > so add --windowmanager openbox to that line.
> > 
> > A lot of users, however, prefer not to run kde's fullblown services, and they
> > run kicker and/or kdesktop separately, together with openbox. Whatever works
> > for you.
> > 
> > > Later,
> > > James Harr <james at grickle.org>
> > > 
> 


-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021004/ef7639f6/attachment.pgp>

From james at grickle.org  Fri Oct  4 20:26:17 2002
From: james at grickle.org (James Harr)
Date: Fri, 4 Oct 2002 19:26:17 -0500 (CDT)
Subject: [openbox] KDE configure options
In-Reply-To: <20021005000301.GA26054@orodu.net>
References: <32960.204.182.60.2.1033765211.squirrel@webmail.grickle.org>
        <20021004215629.GA20258@cs.pdx.edu>
        <1033757936.19207.1.camel@Zeus>
        <20021005000301.GA26054@orodu.net>
Message-ID: <32783.204.182.60.2.1033777577.squirrel@webmail.grickle.org>

Hey again,

Just got openbox running, I'm impressed. It'll take a while to get used to
it as I'm an old-time window-maker user. I needed something different.

I actually lied, I'm not starting it from my .xinitrc (although I should
create a hook in the Xsession so that kdm can have a "xinitrc" option).
I'm doing it from kdm. I'll probably end up setting up that xinitrc deal I
was talking about.

Thanks for the suggestion...

Later,
jh

Ben Jansens said:
> On Fri, Oct 04, 2002 at 01:58:56PM -0500, James Harr wrote:
>> Thanks,
>>
>> I think I'll use my .xinitrc file instead of modifying it globally for
>> everyone. I thought KDE had an option for window managers like gnome
>> does. You don't think it would be all that hard to implement..
>>
>> Hopefully they do sometime in the future.
>>
>
> The KDE developers see no reason why everyone doesnt run kwin, and so I
> don't think they will add this. In KDE 3.1 tho, youll be able to set the
> KDEWM environment variable and kde will use that if it is set
> instead of kwin.
>
> personally, ive found the best is to put this in my .xinitrc
>
> LD_BIND_NOW=true kdeinit +kcminit +knotify
> kicker & korgac & klipper &
> epist & exec openbox
>
>> On Fri, 2002-10-04 at 16:56, Marius Nita wrote:
>> > On Fri, Oct 04, 2002 at 04:00:11PM -0500, James Harr wrote:
>> > > I noticed on some of the screenshots that you were running kde,
>> but with openbox as the window manager.
>> > >
>> > > I was wandering what steps you took to do that. I know in gnome
>> you can select it in the config program, but I saw nothing like
>> that in kde.
>> >
>> > From /usr/bin/startkde, toward the end of the file
>> >
>> > # finally, give the session control to the session manager
>> > # Syntax:   ksmserver [--restore] [--windowmanager <wm>]
>> > # if no windowmanager is specified, ksmserver will ensure kwin is
>> started. # [--restore] should be controlled by kdm
>> > # kwrapper is used to reduce startup time and memory usage
>> > kwrapper ksmserver --restore || \
>> >   xmessage -geometry 500x100 "Couldn't start ksmserver. Check your
>> installation."
>> >
>> >
>> > so add --windowmanager openbox to that line.
>> >
>> > A lot of users, however, prefer not to run kde's fullblown services,
>> and they run kicker and/or kdesktop separately, together with
>> openbox. Whatever works for you.
>> >
>> > > Later,
>> > > James Harr <james at grickle.org>
>> > >
>>
>
>
> --
> I am damn unsatisfied to be killed in this way.


Thank You,
James Harr <james at grickle.org>




From james at grickle.org  Fri Oct  4 21:10:11 2002
From: james at grickle.org (James Harr)
Date: Fri, 4 Oct 2002 20:10:11 -0500 (CDT)
Subject: [openbox] KDE configure options
In-Reply-To: <32783.204.182.60.2.1033777577.squirrel@webmail.grickle.org>
References: <32960.204.182.60.2.1033765211.squirrel@webmail.grickle.org>
        <20021004215629.GA20258@cs.pdx.edu>
        <1033757936.19207.1.camel@Zeus>
        <20021005000301.GA26054@orodu.net>
        <32783.204.182.60.2.1033777577.squirrel@webmail.grickle.org>
Message-ID: <32934.204.182.60.2.1033780211.squirrel@webmail.grickle.org>

Seem to have gotton it working, now only one issue remains..

The openbox menu vs. kde's desktop. I know there's some shortcut keys, but
the problem is that I can't seem to find any of them. Is there more to the
config than the rc file?, or at least what openbox's menu puts in there?

Later,
jh

James Harr said:
> Hey again,
>
> Just got openbox running, I'm impressed. It'll take a while to get used
> to it as I'm an old-time window-maker user. I needed something
> different.
>
> I actually lied, I'm not starting it from my .xinitrc (although I should
> create a hook in the Xsession so that kdm can have a "xinitrc" option).
> I'm doing it from kdm. I'll probably end up setting up that xinitrc deal
> I was talking about.
>
> Thanks for the suggestion...
>
> Later,
> jh
>
> Ben Jansens said:
>> On Fri, Oct 04, 2002 at 01:58:56PM -0500, James Harr wrote:
>>> Thanks,
>>>
>>> I think I'll use my .xinitrc file instead of modifying it globally
>>> for everyone. I thought KDE had an option for window managers like
>>> gnome does. You don't think it would be all that hard to implement..
>>>
>>> Hopefully they do sometime in the future.
>>>
>>
>> The KDE developers see no reason why everyone doesnt run kwin, and so
>> I don't think they will add this. In KDE 3.1 tho, youll be able to set
>> the KDEWM environment variable and kde will use that if it is set
>> instead of kwin.
>>
>> personally, ive found the best is to put this in my .xinitrc
>>
>> LD_BIND_NOW=true kdeinit +kcminit +knotify
>> kicker & korgac & klipper &
>> epist & exec openbox
>>
>>> On Fri, 2002-10-04 at 16:56, Marius Nita wrote:
>>> > On Fri, Oct 04, 2002 at 04:00:11PM -0500, James Harr wrote:
>>> > > I noticed on some of the screenshots that you were running kde,
>>> but with openbox as the window manager.
>>> > >
>>> > > I was wandering what steps you took to do that. I know in gnome
>>> you can select it in the config program, but I saw nothing like that
>>> in kde.
>>> >
>>> > From /usr/bin/startkde, toward the end of the file
>>> >
>>> > # finally, give the session control to the session manager
>>> > # Syntax:   ksmserver [--restore] [--windowmanager <wm>]
>>> > # if no windowmanager is specified, ksmserver will ensure kwin is
>>> started. # [--restore] should be controlled by kdm
>>> > # kwrapper is used to reduce startup time and memory usage
>>> > kwrapper ksmserver --restore || \
>>> >   xmessage -geometry 500x100 "Couldn't start ksmserver. Check your
>>> installation."
>>> >
>>> >
>>> > so add --windowmanager openbox to that line.
>>> >
>>> > A lot of users, however, prefer not to run kde's fullblown
>>> services,
>>> and they run kicker and/or kdesktop separately, together with
>>> openbox. Whatever works for you.
>>> >
>>> > > Later,
>>> > > James Harr <james at grickle.org>
>>> > >
>>>
>>
>>
>> --
>> I am damn unsatisfied to be killed in this way.
>
>
> Thank You,
> James Harr <james at grickle.org>


Thank You,
James Harr <james at grickle.org>




From ben at orodu.net  Fri Oct  4 21:12:01 2002
From: ben at orodu.net (Ben Jansens)
Date: Fri, 4 Oct 2002 20:12:01 -0500
Subject: [openbox] KDE configure options
In-Reply-To: <32934.204.182.60.2.1033780211.squirrel@webmail.grickle.org>
References: <32960.204.182.60.2.1033765211.squirrel@webmail.grickle.org> <20021004215629.GA20258@cs.pdx.edu> <1033757936.19207.1.camel@Zeus> <20021005000301.GA26054@orodu.net> <32783.204.182.60.2.1033777577.squirrel@webmail.grickle.org> <32934.204.182.60.2.1033780211.squirrel@webmail.grickle.org>
Message-ID: <20021005011200.GB26054@orodu.net>

On Fri, Oct 04, 2002 at 08:10:11PM -0500, James Harr wrote:
> Seem to have gotton it working, now only one issue remains..
> 
> The openbox menu vs. kde's desktop. I know there's some shortcut keys, but
> the problem is that I can't seem to find any of them. Is there more to the
> config than the rc file?, or at least what openbox's menu puts in there?

Good resources for configuration:
http://icculus.org/openbox/rc.php
man epist
man epistrc   <-- keybingings config file

Its all in there.

Later,
Ben
 
> Later,
> jh
> 
> James Harr said:
> > Hey again,
> >
> > Just got openbox running, I'm impressed. It'll take a while to get used
> > to it as I'm an old-time window-maker user. I needed something
> > different.
> >
> > I actually lied, I'm not starting it from my .xinitrc (although I should
> > create a hook in the Xsession so that kdm can have a "xinitrc" option).
> > I'm doing it from kdm. I'll probably end up setting up that xinitrc deal
> > I was talking about.
> >
> > Thanks for the suggestion...
> >
> > Later,
> > jh
> >
> > Ben Jansens said:
> >> On Fri, Oct 04, 2002 at 01:58:56PM -0500, James Harr wrote:
> >>> Thanks,
> >>>
> >>> I think I'll use my .xinitrc file instead of modifying it globally
> >>> for everyone. I thought KDE had an option for window managers like
> >>> gnome does. You don't think it would be all that hard to implement..
> >>>
> >>> Hopefully they do sometime in the future.
> >>>
> >>
> >> The KDE developers see no reason why everyone doesnt run kwin, and so
> >> I don't think they will add this. In KDE 3.1 tho, youll be able to set
> >> the KDEWM environment variable and kde will use that if it is set
> >> instead of kwin.
> >>
> >> personally, ive found the best is to put this in my .xinitrc
> >>
> >> LD_BIND_NOW=true kdeinit +kcminit +knotify
> >> kicker & korgac & klipper &
> >> epist & exec openbox
> >>
> >>> On Fri, 2002-10-04 at 16:56, Marius Nita wrote:
> >>> > On Fri, Oct 04, 2002 at 04:00:11PM -0500, James Harr wrote:
> >>> > > I noticed on some of the screenshots that you were running kde,
> >>> but with openbox as the window manager.
> >>> > >
> >>> > > I was wandering what steps you took to do that. I know in gnome
> >>> you can select it in the config program, but I saw nothing like that
> >>> in kde.
> >>> >
> >>> > From /usr/bin/startkde, toward the end of the file
> >>> >
> >>> > # finally, give the session control to the session manager
> >>> > # Syntax:   ksmserver [--restore] [--windowmanager <wm>]
> >>> > # if no windowmanager is specified, ksmserver will ensure kwin is
> >>> started. # [--restore] should be controlled by kdm
> >>> > # kwrapper is used to reduce startup time and memory usage
> >>> > kwrapper ksmserver --restore || \
> >>> >   xmessage -geometry 500x100 "Couldn't start ksmserver. Check your
> >>> installation."
> >>> >
> >>> >
> >>> > so add --windowmanager openbox to that line.
> >>> >
> >>> > A lot of users, however, prefer not to run kde's fullblown
> >>> services,
> >>> and they run kicker and/or kdesktop separately, together with
> >>> openbox. Whatever works for you.
> >>> >
> >>> > > Later,
> >>> > > James Harr <james at grickle.org>
> >>> > >
> >>>
> >>
> >>
> >
> >
> > Thank You,
> > James Harr <james at grickle.org>
> 
> 
> Thank You,
> James Harr <james at grickle.org>
> 


-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021004/cbe4a4d9/attachment.pgp>

From james at grickle.org  Sat Oct  5 11:06:38 2002
From: james at grickle.org (James Harr)
Date: Sat, 5 Oct 2002 10:06:38 -0500 (CDT)
Subject: [openbox] KDE configure options
In-Reply-To: <20021005011200.GB26054@orodu.net>
References: <32960.204.182.60.2.1033765211.squirrel@webmail.grickle.org>
        <20021004215629.GA20258@cs.pdx.edu>
        <1033757936.19207.1.camel@Zeus>
        <20021005000301.GA26054@orodu.net>
        <32783.204.182.60.2.1033777577.squirrel@webmail.grickle.org>
        <32934.204.182.60.2.1033780211.squirrel@webmail.grickle.org>
        <20021005011200.GB26054@orodu.net>
Message-ID: <1143.204.182.60.2.1033830398.squirrel@webmail.grickle.org>

Oh, didn't know about the epis part.

Thanks,
jh


Ben Jansens said:
> On Fri, Oct 04, 2002 at 08:10:11PM -0500, James Harr wrote:
>> Seem to have gotton it working, now only one issue remains..
>>
>> The openbox menu vs. kde's desktop. I know there's some shortcut keys,
>> but the problem is that I can't seem to find any of them. Is there
>> more to the config than the rc file?, or at least what openbox's menu
>> puts in there?
>
> Good resources for configuration:
> http://icculus.org/openbox/rc.php
> man epist
> man epistrc   <-- keybingings config file
>
> Its all in there.
>
> Later,
> Ben
>
>> Later,
>> jh
>>
>> James Harr said:
>> > Hey again,
>> >
>> > Just got openbox running, I'm impressed. It'll take a while to get
>> used to it as I'm an old-time window-maker user. I needed something
>> different.
>> >
>> > I actually lied, I'm not starting it from my .xinitrc (although I
>> should create a hook in the Xsession so that kdm can have a
>> "xinitrc" option). I'm doing it from kdm. I'll probably end up
>> setting up that xinitrc deal I was talking about.
>> >
>> > Thanks for the suggestion...
>> >
>> > Later,
>> > jh
>> >
>> > Ben Jansens said:
>> >> On Fri, Oct 04, 2002 at 01:58:56PM -0500, James Harr wrote:
>> >>> Thanks,
>> >>>
>> >>> I think I'll use my .xinitrc file instead of modifying it globally
>> for everyone. I thought KDE had an option for window managers like
>> gnome does. You don't think it would be all that hard to
>> implement..
>> >>>
>> >>> Hopefully they do sometime in the future.
>> >>>
>> >>
>> >> The KDE developers see no reason why everyone doesnt run kwin, and
>> so I don't think they will add this. In KDE 3.1 tho, youll be able
>> to set the KDEWM environment variable and kde will use that if it
>> is set instead of kwin.
>> >>
>> >> personally, ive found the best is to put this in my .xinitrc
>> >>
>> >> LD_BIND_NOW=true kdeinit +kcminit +knotify
>> >> kicker & korgac & klipper &
>> >> epist & exec openbox
>> >>
>> >>> On Fri, 2002-10-04 at 16:56, Marius Nita wrote:
>> >>> > On Fri, Oct 04, 2002 at 04:00:11PM -0500, James Harr wrote:
>> >>> > > I noticed on some of the screenshots that you were running
>> kde,
>> >>> but with openbox as the window manager.
>> >>> > >
>> >>> > > I was wandering what steps you took to do that. I know in
>> gnome
>> >>> you can select it in the config program, but I saw nothing like
>> that in kde.
>> >>> >
>> >>> > From /usr/bin/startkde, toward the end of the file
>> >>> >
>> >>> > # finally, give the session control to the session manager #
>> Syntax:   ksmserver [--restore] [--windowmanager <wm>]
>> >>> > # if no windowmanager is specified, ksmserver will ensure kwin
>> is
>> >>> started. # [--restore] should be controlled by kdm
>> >>> > # kwrapper is used to reduce startup time and memory usage
>> kwrapper ksmserver --restore || \
>> >>> >   xmessage -geometry 500x100 "Couldn't start ksmserver. Check
>> your
>> >>> installation."
>> >>> >
>> >>> >
>> >>> > so add --windowmanager openbox to that line.
>> >>> >
>> >>> > A lot of users, however, prefer not to run kde's fullblown
>> >>> services,
>> >>> and they run kicker and/or kdesktop separately, together with
>> openbox. Whatever works for you.
>> >>> >
>> >>> > > Later,
>> >>> > > James Harr <james at grickle.org>
>> >>> > >
>> >>>
>> >>
>> >>
>> >
>> >
>> > Thank You,
>> > James Harr <james at grickle.org>
>>
>>
>> Thank You,
>> James Harr <james at grickle.org>
>>
>
>
> --
> I am damn unsatisfied to be killed in this way.


Thank You,
James Harr <james at grickle.org>




From awol at home.nl  Mon Oct  7 06:45:35 2002
From: awol at home.nl (Alexander Volovics)
Date: Mon, 7 Oct 2002 12:45:35 +0200
Subject: compiling openbox-2.1.2 on redhat-8.0
Message-ID: <20021007104535.GA1466@home.nl>

Hello,

Yesterday i tried to compile openbox-2.1.2 on redhat-8.0
with xft support.

I get a working openbox except for xft support.
Configure claims it can't find XftFontOpenXlfd in -lXft.

I have both Xft and Xft-devel installed.
Can somebody who managed to get everything working on redhat-8.0
please give me some suggestions on how to proceed.

Alexander




From john at hl.lu.se  Mon Oct  7 07:04:20 2002
From: john at hl.lu.se (John Eckerdal)
Date: Mon, 7 Oct 2002 13:04:20 +0200 (CEST)
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021007104535.GA1466@home.nl>
Message-ID: <Pine.LNX.4.44.0210071257210.17264-100000@maestro.hl.lu.se>

On Mon, 7 Oct 2002, Alexander Volovics wrote:

> Hello,
> 
> Yesterday i tried to compile openbox-2.1.2 on redhat-8.0
> with xft support.
> 
> I get a working openbox except for xft support.
> Configure claims it can't find XftFontOpenXlfd in -lXft.
> 
> I have both Xft and Xft-devel installed.
> Can somebody who managed to get everything working on redhat-8.0
> please give me some suggestions on how to proceed.

Since RH (I use it myself) places Xft in a place where Openbox does
not look for it some tweaking has to be performed.

I did this by manually editing the two Makefiles after I ran configure.
I know, this is a hack but I don't know how the autoconf/automake stuff 
works, someone with more knowledge with these two might have a better
(and more premanent soultion).

However:
in src/Makefile
find the LDFLAGS line and append `xft-config --libs` to it.
find the CXXFLAGS line and append `xft-config --cflags` to it.

in util/Make file
find the xftlsfonts_LDFLADS and append `xft-config --libs` to it.
find the CXXFLAGS and append `xft-config --cflags` to it.

Thats all I had to do.

-- 
John Eckerdal



From awol at home.nl  Mon Oct  7 14:48:43 2002
From: awol at home.nl (Alexander Volovics)
Date: Mon, 7 Oct 2002 20:48:43 +0200
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <Pine.LNX.4.44.0210071257210.17264-100000@maestro.hl.lu.se>
References: <20021007104535.GA1466@home.nl> <Pine.LNX.4.44.0210071257210.17264-100000@maestro.hl.lu.se>
Message-ID: <20021007184843.GA1131@home.nl>

On Mon, Oct 07, 2002 at 01:04:20PM +0200, John Eckerdal wrote:

> > Yesterday i tried to compile openbox-2.1.2 on redhat-8.0
> > with xft support.
> > I get a working openbox except for xft support.
> > Configure claims it can't find XftFontOpenXlfd in -lXft.

Thanks for your answer.
 
> Since RH (I use it myself) places Xft in a place where Openbox does
> not look for it some tweaking has to be performed.

I had already understood that Openbox was looking in the wrong
places and had tried to correct this using environment variables
like LDFLAGS with configure. Couldn't get it working.
(I know nothing about programming).
 
> I did this by manually editing the two Makefiles after I ran configure.
> I know, this is a hack but I don't know how the autoconf/automake stuff 
> works, someone with more knowledge with these two might have a better
> (and more premanent soultion).

> However:
> in src/Makefile
> find the LDFLAGS line and append `xft-config --libs` to it.
> find the CXXFLAGS line and append `xft-config --cflags` to it.
> 
> in util/Make file
> find the xftlsfonts_LDFLADS and append `xft-config --libs` to it.
> find the CXXFLAGS and append `xft-config --cflags` to it.

However your suggestions also do not work. There seems to be
something missing. The result now is that make looks for
files (called something like 'dcflags') in the wrong places.

If you have overlooked something can you please let me know.

Alexander



From john at hl.lu.se  Mon Oct  7 15:28:04 2002
From: john at hl.lu.se (John Eckerdal)
Date: Mon, 7 Oct 2002 21:28:04 +0200 (CEST)
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021007184843.GA1131@home.nl>
Message-ID: <Pine.LNX.4.44.0210072127350.20389-100000@maestro.hl.lu.se>

On Mon, 7 Oct 2002, Alexander Volovics wrote:

> However your suggestions also do not work. There seems to be
> something missing. The result now is that make looks for
> files (called something like 'dcflags') in the wrong places.
> 
> If you have overlooked something can you please let me know.

Can't hurt to ask:

Do you have the Xft-devel package installed?

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From awol at home.nl  Mon Oct  7 15:38:03 2002
From: awol at home.nl (Alexander Volovics)
Date: Mon, 7 Oct 2002 21:38:03 +0200
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <Pine.LNX.4.44.0210072127350.20389-100000@maestro.hl.lu.se>
References: <20021007184843.GA1131@home.nl> <Pine.LNX.4.44.0210072127350.20389-100000@maestro.hl.lu.se>
Message-ID: <20021007193803.GA1207@home.nl>

On Mon, Oct 07, 2002 at 09:28:04PM +0200, John Eckerdal wrote:
 
> > However your suggestions also do not work. There seems to be
> > something missing. The result now is that make looks for
> > files (called something like 'dcflags') in the wrong places.
> > 
> > If you have overlooked something can you please let me know.
 
> Can't hurt to ask:
> 
> Do you have the Xft-devel package installed?

Yes, both Xft-2.0-1 and Xft-devel-2.0-1 are installed.

Alexander



From john at hl.lu.se  Mon Oct  7 16:11:06 2002
From: john at hl.lu.se (John Eckerdal)
Date: Mon, 7 Oct 2002 22:11:06 +0200 (CEST)
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021007193803.GA1207@home.nl>
Message-ID: <Pine.LNX.4.44.0210072210250.20650-100000@maestro.hl.lu.se>

On Mon, 7 Oct 2002, Alexander Volovics wrote:

> Yes, both Xft-2.0-1 and Xft-devel-2.0-1 are installed.

Ok. That is good. What does the compiler say when you try to
compile it?

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From awol at home.nl  Mon Oct  7 18:02:25 2002
From: awol at home.nl (Alexander Volovics)
Date: Tue, 8 Oct 2002 00:02:25 +0200
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <Pine.LNX.4.44.0210072210250.20650-100000@maestro.hl.lu.se>
References: <20021007193803.GA1207@home.nl> <Pine.LNX.4.44.0210072210250.20650-100000@maestro.hl.lu.se>
Message-ID: <20021007220225.GA1292@home.nl>

On Mon, Oct 07, 2002 at 10:11:06PM +0200, John Eckerdal wrote:
 
> > Yes, both Xft-2.0-1 and Xft-devel-2.0-1 are installed.
 
> Ok. That is good. What does the compiler say when you try to
> compile it?


I would like to ask some questions first before I answer because
I don't understand your suggestions.

To adapt the 'Makefile' in /src and /util you first have to run
./configure to create them.

But in config.log (after running configure) I see the following:

configure:5232: gcc -o conftest -g -O2  -I/usr/X11R6/include
+-L/usr/X11R6/lib  -lSM -lICE conftest.c -lXft    -L/usr/X11R6/lib -lX11  -lXext
+>&5
/usr/bin/ld: cannot find -lXft
collect2: ld returned 1 exit status
configure:5235: $? = 1
configure: failed program was:
ne 5206 "configure"
#include "confdefs.h"

If I understand this correctly it says that necessary headers/libs from
Xft have not been found.
Can this still be corrected 'afterwards' by adapting the makefiles?
Must it not be corrected during configure itself?

Also if I understand correctly the linker /usr/bin/ld cannot find
libXft.so.2 or libXft.so.2.0 (which are in /usr/lib). 
Does the linker not look in /etc/ld.so.conf for paths. If the necessary
paths were included there would it work?

In your solution you make use of a bash script from Xft-devel:
/usr/bin/xft-config.
Looking at the script it prescribes the syntax:
xft-config [OPTIONS] [LIBRARIES] but you only only suggest adding
'xft-config --libs' and 'xft-config --cflags'.
Did you really mean that only this should be included or was I meant
to add the libraries, thus:
'xft-config --libs /usr/lib' and 'xft-config --cflags /usr/include/Xft2/X11'.

I suspect that only including 'xft-config --libs' gives the necessary
libs to link to so that the paths themselves do not have to be added.

Adding the libraries seems more logical but both ways do not yield the
expected result. (Or should the libraries have been included like for
example '-L/usr/lib' and '-I/usr/include/Xft2/X11'.

As you can see I am not a programmer. But I like to understand at least
something of what is going on.

Alexander



From ben at orodu.net  Mon Oct  7 18:14:39 2002
From: ben at orodu.net (Ben Jansens)
Date: Mon, 7 Oct 2002 17:14:39 -0500
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021007220225.GA1292@home.nl>
References: <20021007193803.GA1207@home.nl> <Pine.LNX.4.44.0210072210250.20650-100000@maestro.hl.lu.se> <20021007220225.GA1292@home.nl>
Message-ID: <20021007221439.GA17750@orodu.net>

On Tue, Oct 08, 2002 at 12:02:25AM +0200, Alexander Volovics wrote:
> On Mon, Oct 07, 2002 at 10:11:06PM +0200, John Eckerdal wrote:
>  
> > > Yes, both Xft-2.0-1 and Xft-devel-2.0-1 are installed.
>  
> > Ok. That is good. What does the compiler say when you try to
> > compile it?
> 
> 
> I would like to ask some questions first before I answer because
> I don't understand your suggestions.
> 
> To adapt the 'Makefile' in /src and /util you first have to run
> ./configure to create them.
> 
> But in config.log (after running configure) I see the following:
> 
> configure:5232: gcc -o conftest -g -O2  -I/usr/X11R6/include
> +-L/usr/X11R6/lib  -lSM -lICE conftest.c -lXft    -L/usr/X11R6/lib -lX11  -lXext
> +>&5
> /usr/bin/ld: cannot find -lXft
> collect2: ld returned 1 exit status
> configure:5235: $? = 1
> configure: failed program was:
> ne 5206 "configure"
> #include "confdefs.h"
> 
> If I understand this correctly it says that necessary headers/libs from
> Xft have not been found.
> Can this still be corrected 'afterwards' by adapting the makefiles?
> Must it not be corrected during configure itself?
> 
> Also if I understand correctly the linker /usr/bin/ld cannot find
> libXft.so.2 or libXft.so.2.0 (which are in /usr/lib). 
> Does the linker not look in /etc/ld.so.conf for paths. If the necessary
> paths were included there would it work?

Openbox 2 uses Xft 1, which is shipped with XFree86 4. Xft 2 is a new
and different beast, and is not meant to be entire backwards
compatible i beleive.
 
> In your solution you make use of a bash script from Xft-devel:
> /usr/bin/xft-config.
> Looking at the script it prescribes the syntax:
> xft-config [OPTIONS] [LIBRARIES] but you only only suggest adding
> 'xft-config --libs' and 'xft-config --cflags'.
> Did you really mean that only this should be included or was I meant
> to add the libraries, thus:
> 'xft-config --libs /usr/lib' and 'xft-config --cflags /usr/include/Xft2/X11'.
> 
> I suspect that only including 'xft-config --libs' gives the necessary
> libs to link to so that the paths themselves do not have to be added.
> 
> Adding the libraries seems more logical but both ways do not yield the
> expected result. (Or should the libraries have been included like for
> example '-L/usr/lib' and '-I/usr/include/Xft2/X11'.
> 
> As you can see I am not a programmer. But I like to understand at least
> something of what is going on.

So, this is why its not working. You're trying to make it use Xft2,
which redhatapparently has moved to before anyone else. See if you
still have xft stuff around in /usr/X11R6 or if there are other xft 1
packages. Or, if you do get it to go with Xft 2, let me know. But
thats why the default build process is not working for you.

Ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021007/a542ddc3/attachment.pgp>

From john at hl.lu.se  Mon Oct  7 18:26:53 2002
From: john at hl.lu.se (John Eckerdal)
Date: Tue, 8 Oct 2002 00:26:53 +0200 (CEST)
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021007221439.GA17750@orodu.net>
Message-ID: <Pine.LNX.4.44.0210080018460.21263-100000@maestro.hl.lu.se>

On Mon, 7 Oct 2002, Ben Jansens wrote:

Me answering to two mails at the same time:

> > To adapt the 'Makefile' in /src and /util you first have to run
> > ./configure to create them.

You modify the Makefiles after configure creates them.

> > But in config.log (after running configure) I see the following:
> > 
> > configure:5232: gcc -o conftest -g -O2  -I/usr/X11R6/include
> > +-L/usr/X11R6/lib  -lSM -lICE conftest.c -lXft    -L/usr/X11R6/lib -lX11  -lXext
> > +>&5
> > /usr/bin/ld: cannot find -lXft
> > collect2: ld returned 1 exit status
> > configure:5235: $? = 1
> > configure: failed program was:
> > ne 5206 "configure"
> > #include "confdefs.h"

This is an expected error since RH ships Xft2, not Xft1 and also
Xft2 is not placed in the same directory as Xft1.

> > If I understand this correctly it says that necessary headers/libs from
> > Xft have not been found.
> > Can this still be corrected 'afterwards' by adapting the makefiles?
> > Must it not be corrected during configure itself?

No. configure only creates the Makefiles, what happens to them later on
is not something configure cares about.

> > In your solution you make use of a bash script from Xft-devel:
> > /usr/bin/xft-config.
> > Looking at the script it prescribes the syntax:
> > xft-config [OPTIONS] [LIBRARIES] but you only only suggest adding
> > 'xft-config --libs' and 'xft-config --cflags'.
> > Did you really mean that only this should be included or was I meant
> > to add the libraries, thus:
> > 'xft-config --libs /usr/lib' and 'xft-config --cflags /usr/include/Xft2/X11'.

No. Just
`xft-config --libs` to LDFLAGS
and
`xft-config --cflags` to CXXFLAGS
(there are a few to choose from, choose wisely. See the first mail).

Note that this is not ''-pair but a ``-pair. These are back-quotes
and using them will cause the Makefile to run the application.
My ` are at shift+(the key next to backspace)

> > I suspect that only including 'xft-config --libs' gives the necessary
> > libs to link to so that the paths themselves do not have to be added.

Yes. If you enclose it in ``, not ''.

[And finally]

> So, this is why its not working. You're trying to make it use Xft2,
> which redhatapparently has moved to before anyone else. See if you
> still have xft stuff around in /usr/X11R6 or if there are other xft 1
> packages. Or, if you do get it to go with Xft 2, let me know. But
> thats why the default build process is not working for you.

Xft2 works here OK here. The application compiles and runs and the
fonts looks OK I think. One of the utilities fails to compile though.
Have not looked into that one.
Openbox can be installed inspite of this failed compilation.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From awol at home.nl  Mon Oct  7 19:08:39 2002
From: awol at home.nl (Alexander Volovics)
Date: Tue, 8 Oct 2002 01:08:39 +0200
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021007221439.GA17750@orodu.net>
References: <20021007193803.GA1207@home.nl> <Pine.LNX.4.44.0210072210250.20650-100000@maestro.hl.lu.se> <20021007220225.GA1292@home.nl> <20021007221439.GA17750@orodu.net>
Message-ID: <20021007230839.GA1323@home.nl>

On Mon, Oct 07, 2002 at 05:14:39PM -0500, Ben Jansens and John Eckerdal wrote:

I am replying to both your mails at the same time.

Ben Jansens: 
> Openbox 2 uses Xft 1, which is shipped with XFree86 4. Xft 2 is a new
> and different beast, and is not meant to be entire backwards
> compatible i beleive.

John Eckerdal:
> This is an expected error since RH ships Xft2, not Xft1 and also
> Xft2 is not placed in the same directory as Xft1.

These remarks made me take a closer look. In my naivete I thought the
libraries would be supplied by Xft itself, so I only looked at where
Xft put its files. However now I did a 'locate libXft.so' and lo and behold
locate finds the following files:

 /usr/lib/libXft.so.2
 /usr/lib/libXft.so.2.0
 /usr/X11R6/lib/libXft.so.1
 /usr/X11R6/lib/libXft.so.1.2

The first two from Xft-2.0.1 and the second two from XFree86-libs-4.2.0-72.
So Red Hat ships both versions!

Ben Jansens: 
> So, this is why its not working. You're trying to make it use Xft2,
> which redhatapparently has moved to before anyone else. See if you
> still have xft stuff around in /usr/X11R6 or if there are other xft 1
> packages. Or, if you do get it to go with Xft 2, let me know. But
> thats why the default build process is not working for you.

*I* am not trying to make it use anything. I first unpacked openbox-2.1.1.tar.gz
and did a standard "./configure, make, make install" and I found the
error in config.log that '/usr/bin/ld could not find -lXft'.
So openbox compiled but without support for xft.

Why did this happen when the standard libXft.so.1's from XFree86 are
available? Are the newer redhat libs interfering in some way?

My question now is: if the "correct" libs are available how can I
go about compiling so that the linker does find them and openbox
compiles with xft support.

John Eckerdal:
> Xft2 works here OK here. The application compiles and runs and the
> fonts looks OK I think. One of the utilities fails to compile though.
> Have not looked into that one.
> Openbox can be installed inspite of this failed compilation.

I might indeed have made an error by not using the correct pair from
'' or ``. I will try again after I send this mail.

If openbox compiles OK with both libs which version is to be prefered?

Alexander



From john at hl.lu.se  Mon Oct  7 19:34:43 2002
From: john at hl.lu.se (John Eckerdal)
Date: Tue, 8 Oct 2002 01:34:43 +0200 (CEST)
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021007230839.GA1323@home.nl>
Message-ID: <Pine.LNX.4.44.0210080129110.21659-100000@maestro.hl.lu.se>


On Tue, 8 Oct 2002, Alexander Volovics wrote:
 
> These remarks made me take a closer look. In my naivete I thought the
> libraries would be supplied by Xft itself, so I only looked at where
> Xft put its files. However now I did a 'locate libXft.so' and lo and behold
> locate finds the following files:
> 
>  /usr/lib/libXft.so.2
>  /usr/lib/libXft.so.2.0
>  /usr/X11R6/lib/libXft.so.1
>  /usr/X11R6/lib/libXft.so.1.2
> 
> The first two from Xft-2.0.1 and the second two from XFree86-libs-4.2.0-72.
> So Red Hat ships both versions!

Yes, but they only ship the headerfiles for Xft2. The Xft1 libraries are
probably shipped to maintain compatiblity with precompiled applications
for earlier RH versions.
 
> My question now is: if the "correct" libs are available how can I
> go about compiling so that the linker does find them and openbox
> compiles with xft support.

Unless there is some compat packages you probably can't get openbox
to use Xft1 since the headerfiles are missing (atleast they are here
but then again my RH install is not a normal RH install it has been
tweaked for several years).

All I can say is that Openbox compiles and works OK with Xft2.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From awol at home.nl  Tue Oct  8 14:59:25 2002
From: awol at home.nl (Alexander Volovics)
Date: Tue, 8 Oct 2002 20:59:25 +0200
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <Pine.LNX.4.44.0210080018460.21263-100000@maestro.hl.lu.se>
References: <20021007221439.GA17750@orodu.net> <Pine.LNX.4.44.0210080018460.21263-100000@maestro.hl.lu.se>
Message-ID: <20021008185925.GA1253@home.nl>

On Tue, Oct 08, 2002 at 12:26:53AM +0200, John Eckerdal wrote:

> No. Just
> `xft-config --libs` to LDFLAGS
> and
> `xft-config --cflags` to CXXFLAGS
 
> Note that this is not ''-pair but a ``-pair. These are back-quotes
> and using them will cause the Makefile to run the application.

OK, I started again from scratch today using openbox-2.1.1.

1) did ./configure
2) appended `xft-config --libs` and `xft-config --cflags` to the
   Makefiles in /src and /util at LDFLAGS and CXXFLAGS (making
   sure I used the correct ``-pair this time).
3) ran 'make' and 'make install'

Indeed openbox compiles and runs BUT it is still not xft enabled!

Running 'openbox -h' shows:
 Compile time options:
   Debugging                     No
   Shape:                        Yes
   Xft:                          No
   Xinerama:                     No
   8bpp Ordered Dithering:       No

And running 'xftlsfonts' shows:
 Openbox was built without support for Xft fonts. This utility must
 be built with Xft support in order to function.

Are you really sure that you can ignore the linking with -lXft
that ./configure is supposed to supply 

By the way:
RH-8.0 does contain 2 header files, both supplied by Xft-devel-2.0.1,
namely '/usr/include/Xft2/X11/Xft/Xft.h' and 
'/usr/include/Xft2/X11/Xft/XftCompat.h'.
Xft.h somewhat resembles the Xft.h supplied by XFree86-devel-4.2.0-8 in
RH-7.3. XFtCompat.h is a smaller file and has a different structure.
I can't find any info describing its function.

Also:
Under RH-8.0 XFree86-devel-4.2.0-72 installs a completely empty
directory: /usr/X11R6/include/X11/Xft ?

Alexander






From lenix at lenix.de  Tue Oct  8 15:39:19 2002
From: lenix at lenix.de (Guido 'lenix' Boehm)
Date: Tue, 8 Oct 2002 21:39:19 +0200
Subject: strange behavior of epist
Message-ID: <20021008213919.206623f1.lenix@lenix.de>

Hi everyone,

would like to say hello to the mailinglist 'cause i just subscribed,
and report a problem i'm just recognized while using openbox +
epist. i'll paste here what i just wrote down in #openbox:

> i have some bindings together with my mod1 key, mod1+left for example
> to cycle to my next window
> sometimes epic 'blocks' my left key although mod1 isn't pressed
> same thing with mod1+down, which is happening right now. if i'm in a
> term (aterm has focus) and press up or down it behaves as if the 
> key-event was just grapped by bbkeys (cursor goes to 'terminal-window
> unfocused' look once but returns to 'focused' look immediatly
> afterwards)

i'll include my epistrc + ./openbox/rc + .Xmodmap with this email.

additional info:
  <lenix at lenix ~> epist -version
  epist - shipped with openbox 2.2.0
  <lenix at lenix ~> X -version
  XFree86 Version 3.3.6 / X Window System
  <lenix at lenix ~> uname
  FreeBSD
  <lenix at lenix ~> uname -srpn
  FreeBSD lenix.unixuni.de 4.7-RC i386
  <lenix at lenix ~> aterm -version
  aterm version 0.4.0 from 02 July 00

oh, and it really looks like an epist-issue because if i kill
epist everything works fine again, but if i start epist after-
wards the keys are blocked again. restarting X (and openbox+
epist of course) fixes the problem. it seems to occur random,
but i'll check if i can find any special events.

regards, lenix
-- 
www: http://www.lenix.de
fon: +49 - 173 - 80 99 196

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: .Xmodmap
URL: <http://icculus.org/pipermail/openbox/attachments/20021008/df9425d4/attachment.diff>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: epistrc
URL: <http://icculus.org/pipermail/openbox/attachments/20021008/df9425d4/attachment-0001.diff>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: rc
URL: <http://icculus.org/pipermail/openbox/attachments/20021008/df9425d4/attachment-0002.diff>

From vR at movingparts.net  Tue Oct  8 15:43:22 2002
From: vR at movingparts.net (Jason 'vanRijn' Kasper)
Date: 08 Oct 2002 15:43:22 -0400
Subject: [openbox] strange behavior of epist
In-Reply-To: <20021008213919.206623f1.lenix@lenix.de>
References: <20021008213919.206623f1.lenix@lenix.de>
Message-ID: <1034106202.31276.94.camel@tobias>

On Tue, 2002-10-08 at 15:39, Guido 'lenix' Boehm wrote:
> Hi everyone,
> 
> would like to say hello to the mailinglist 'cause i just subscribed,
> and report a problem i'm just recognized while using openbox +
> epist. i'll paste here what i just wrote down in #openbox:
> 
> > i have some bindings together with my mod1 key, mod1+left for example
> > to cycle to my next window
> > sometimes epic 'blocks' my left key although mod1 isn't pressed
> > same thing with mod1+down, which is happening right now. if i'm in a
> > term (aterm has focus) and press up or down it behaves as if the 
> > key-event was just grapped by bbkeys (cursor goes to 'terminal-window
> > unfocused' look once but returns to 'focused' look immediatly
> > afterwards)
> 
> i'll include my epistrc + ./openbox/rc + .Xmodmap with this email.
> 
> additional info:
>   <lenix at lenix ~> epist -version
>   epist - shipped with openbox 2.2.0
>   <lenix at lenix ~> X -version
>   XFree86 Version 3.3.6 / X Window System
>   <lenix at lenix ~> uname
>   FreeBSD
>   <lenix at lenix ~> uname -srpn
>   FreeBSD lenix.unixuni.de 4.7-RC i386
>   <lenix at lenix ~> aterm -version
>   aterm version 0.4.0 from 02 July 00
> 
> oh, and it really looks like an epist-issue because if i kill
> epist everything works fine again, but if i start epist after-
> wards the keys are blocked again. restarting X (and openbox+
> epist of course) fixes the problem. it seems to occur random,
> but i'll check if i can find any special events.

Does this happen with bbkeys too?

> 
> regards, lenix
> -- 
> www: http://www.lenix.de
> fon: +49 - 173 - 80 99 196
> 
> ----
> 

> ! map windows keys:
> keycode 115 = F13
> keycode 116 = F14
> keycode 117 = F15
> ! make AltGr a modifier
> keycode 113 = Mode_switch Mode_switch Multi_key
> ! add euro + cent symbols
> keycode  26 = e  E  EuroSign
> keycode  54 = c  C  cent
> ! add german umlauts
> keycode  47 = semicolon colon odiaeresis
> keycode  48 = apostrophe quotedbl adiaeresis
> keycode  34 = bracketleft braceleft udiaeresis
> ----
> 

> Mod1-Down toggleshade;
> Control-Shift-Down moveWindowUp 10;
> F13 execute "aterm";
> F14 execute "netscape";
> F15 execute "aterm -g 59x17+0+0 -T xtrans -e /bin/sh -c "/home/lenix/bin/translate -f; read foo;"";
> Mod1-F4 close;
> Control-Shift-Left moveWindowLeft 10;
> Control-Left prevWorkspace;
> Mod1-Left prevWindow;
> Control-Shift-Right moveWindowRight 10;
> Control-Right nextWorkspace;
> Mod1-Right nextWindow;
> Control-Shift-Up moveWindowUp, 10;
> Mod1-Up toggleMaximizeFull;
> Mod1-Control-s toggleomnipresent;
> Mod2-Down toggleshade;
> Mod2-F4 close;
> Mod2-Left prevWindow;
> Mod2-Right nextWindow;
> Mod2-Up toggleMaximizeFull;
> Mod2-Control-s toggleomnipresent;
> ----
> 

> session.screen0.slit.placement:	BottomRight
> session.screen0.slit.direction:	Horizontal
> session.screen0.slit.onTop:	False
> session.screen0.slit.autoHide:	False
> session.screen0.toolbar.onTop:	False
> session.screen0.toolbar.autoHide:	False
> session.screen0.toolbar.placement:	TopRight
> session.screen0.toolbar.widthPercent:	33
> session.screen0.workspaceMenuButton:	Middle
> session.screen0.placementIgnoreShaded:	False
> session.screen0.focusLastWindow:	False
> session.screen0.windowPlacement:	CascadePlacement
> session.screen0.edgeSnapOffset:	0
> session.screen0.toolbarWidthPercent:	66
> session.screen0.focusModel:	SloppyFocus
> session.screen0.rootScrollDirection:	Normal
> session.screen0.antialiasFonts:	False
> session.screen0.workspaceNames:	coms,terms,docs,browser,edits
> session.screen0.workspaces:	5
> session.screen0.windowToEdgeSnap:	Snap
> session.screen0.disableBindingsWithScrollLock:	True
> session.screen0.resistanceSize:	20
> session.screen0.colPlacementDirection:	TopToBottom
> session.screen0.focusNewWindows:	False
> session.screen0.hideToolbar:	False
> session.screen0.imageDither:	True
> session.screen0.slitPlacement:	CenterRight
> session.screen0.resizeZones:	1
> session.screen0.windowToWindowSnap:	Snap
> session.screen0.windowCornerSnap:	False
> session.screen0.rowPlacementDirection:	LeftToRight
> session.screen0.strftimeFormat:	%H:%M
> session.screen0.workspaceWarping:	False
> session.screen0.placementIgnoreMaximized:	True
> session.screen0.edgeSnapThreshold:	20
> session.screen0.toolbarOnTop:	False
> session.screen0.dropShadowFonts:	False
> session.screen0.opaqueMove:	False
> session.screen0.fullMaximization:	False
> session.screen0.rootMenuButton:	Right
> session.windowCycleModifier:	Mod1
> session.imageDither:	True
> session.colorsPerChannel:	4
> session.autoRaiseDelay:	250
> session.titlebarLayout:	ILMC
> session.colormapFocusModel:	Click
> session.cacheLife:	5
> session.cacheMax:	200
> session.doubleClickInterval:	250
> session.styleFile:	/home/lenix/.blackbox/styles/calm
> session.modifierMask:	Mod1
> session.opaqueMove:	False
> session.menuFile:	/home/lenix/.blackbox/menu
-- 

,-----------------------------------------------------------------//
| Jason 'vanRijn' Kasper ::  Numbers 6:22-26 
 `
 | All brontosauruses are thin at one end, much MUCH thicker 
 | in the middle, and then thin again at the far end.  That is 
 | the theory that I have and which is mine, and what it is too.  
 ,
| bash$ :(){ :|:&};:
`----------------------//


From john at hl.lu.se  Tue Oct  8 16:01:28 2002
From: john at hl.lu.se (John Eckerdal)
Date: Tue, 8 Oct 2002 22:01:28 +0200 (CEST)
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021008185925.GA1253@home.nl>
Message-ID: <Pine.LNX.4.44.0210082158400.27854-100000@maestro.hl.lu.se>

On Tue, 8 Oct 2002, Alexander Volovics wrote:

> On Tue, Oct 08, 2002 at 12:26:53AM +0200, John Eckerdal wrote:
> 
> > No. Just
> > `xft-config --libs` to LDFLAGS
> > and
> > `xft-config --cflags` to CXXFLAGS
>  
> > Note that this is not ''-pair but a ``-pair. These are back-quotes
> > and using them will cause the Makefile to run the application.
> 
> OK, I started again from scratch today using openbox-2.1.1.
> 
> 1) did ./configure
> 2) appended `xft-config --libs` and `xft-config --cflags` to the
>    Makefiles in /src and /util at LDFLAGS and CXXFLAGS (making
>    sure I used the correct ``-pair this time).
> 3) ran 'make' and 'make install'
> 
> Indeed openbox compiles and runs BUT it is still not xft enabled!
> 
> Running 'openbox -h' shows:
>  Compile time options:
>    Debugging                     No
>    Shape:                        Yes
>    Xft:                          No
>    Xinerama:                     No
>    8bpp Ordered Dithering:       No

Strange:

My shows:

Compile time options:
  Debugging:                    no
  Shape:                        yes
  Xft:                          yes
  Xinerama:                     no
  8bpp Ordered Dithering:       no


Are you sure you added it to the correct lines? There are a few to
choose from.

> Are you really sure that you can ignore the linking with -lXft
> that ./configure is supposed to supply 

xft-config --libs

actually provides -lXft2
 
> By the way:
> RH-8.0 does contain 2 header files, both supplied by Xft-devel-2.0.1,
> namely '/usr/include/Xft2/X11/Xft/Xft.h' and 
> '/usr/include/Xft2/X11/Xft/XftCompat.h'.
> Xft.h somewhat resembles the Xft.h supplied by XFree86-devel-4.2.0-8 in
> RH-7.3. XFtCompat.h is a smaller file and has a different structure.
> I can't find any info describing its function.

Didn't know that.

> 
> Also:
> Under RH-8.0 XFree86-devel-4.2.0-72 installs a completely empty
> directory: /usr/X11R6/include/X11/Xft ?

Yes.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From lenix at lenix.de  Tue Oct  8 16:26:10 2002
From: lenix at lenix.de (Guido 'lenix' Boehm)
Date: Tue, 8 Oct 2002 22:26:10 +0200
Subject: [openbox] strange behavior of epist
In-Reply-To: <1034106202.31276.94.camel@tobias>
References: <20021008213919.206623f1.lenix@lenix.de>
	<1034106202.31276.94.camel@tobias>
Message-ID: <20021008222610.645dcb23.lenix@lenix.de>

quoting "Jason 'vanRijn' Kasper" <vR at movingparts.net> 
(08 Oct 2002 15:43:22 -0400):

> > oh, and it really looks like an epist-issue because if i kill
> > epist everything works fine again, but if i start epist after-
> > wards the keys are blocked again. restarting X (and openbox+
> > epist of course) fixes the problem. it seems to occur random,
> > but i'll check if i can find any special events.
> 
> Does this happen with bbkeys too?
> 

oops, sorry, i forgot to test that one before i restarted X, but
if i happens next time i'll try out and let you know :)

regards, lenix
-- 
www: http://www.lenix.de
fon: +49 - 173 - 80 99 196



From john at hl.lu.se  Tue Oct  8 18:36:51 2002
From: john at hl.lu.se (John Eckerdal)
Date: Wed, 9 Oct 2002 00:36:51 +0200 (CEST)
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021008185925.GA1253@home.nl>
Message-ID: <Pine.LNX.4.44.0210090034260.28553-100000@maestro.hl.lu.se>

On Tue, 8 Oct 2002, Alexander Volovics wrote:

> On Tue, Oct 08, 2002 at 12:26:53AM +0200, John Eckerdal wrote:
> 
> > No. Just
> > `xft-config --libs` to LDFLAGS
> > and
> > `xft-config --cflags` to CXXFLAGS
>  
> > Note that this is not ''-pair but a ``-pair. These are back-quotes
> > and using them will cause the Makefile to run the application.
> 
> OK, I started again from scratch today using openbox-2.1.1.
> 
> 1) did ./configure
> 2) appended `xft-config --libs` and `xft-config --cflags` to the
>    Makefiles in /src and /util at LDFLAGS and CXXFLAGS (making
>    sure I used the correct ``-pair this time).
> 3) ran 'make' and 'make install'

Dang!

I forgot something vitally importat!

You will not get Xft support unless modify config.h after you execute
configure.

There is a line probably something like:
/* #undef XFT */

Change that one to:
#define XFT 1

This is a must since configure never finds Xft and will therefor remove
support for it. Sorry, don't know why I forgot that one.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From ben at orodu.net  Tue Oct  8 18:52:14 2002
From: ben at orodu.net (Ben Jansens)
Date: Tue, 8 Oct 2002 17:52:14 -0500
Subject: [openbox] strange behavior of epist
In-Reply-To: <20021008213919.206623f1.lenix@lenix.de>
References: <20021008213919.206623f1.lenix@lenix.de>
Message-ID: <20021008225214.GB17750@orodu.net>

On Tue, Oct 08, 2002 at 09:39:19PM +0200, Guido 'lenix' Boehm wrote:
> Hi everyone,
> 
> would like to say hello to the mailinglist 'cause i just subscribed,
> and report a problem i'm just recognized while using openbox +
> epist. i'll paste here what i just wrote down in #openbox:
> 
> > i have some bindings together with my mod1 key, mod1+left for example
> > to cycle to my next window
> > sometimes epic 'blocks' my left key although mod1 isn't pressed
> > same thing with mod1+down, which is happening right now. if i'm in a
> > term (aterm has focus) and press up or down it behaves as if the 
> > key-event was just grapped by bbkeys (cursor goes to 'terminal-window
> > unfocused' look once but returns to 'focused' look immediatly
> > afterwards)

Coincidentally, I experienced this myself today when i added a Mod1-F1
binding to my epistrc. Using it would cause epist to grab the whole
keyboard and hold it.

Ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021008/67f3759c/attachment.pgp>

From kevin at desertsol.com  Wed Oct  9 12:06:38 2002
From: kevin at desertsol.com (Kevin Geiss)
Date: Wed, 9 Oct 2002 09:06:38 -0700
Subject: xft font help!
Message-ID: <20021009160638.GA20057@raven.localdomain>

long time blackbox user, drawn to openbox for the antialiased fonts and
net wm compliance. But:

I'm trying to get antialiased fonts in openbox, and I need some help.

First, these are things I have done:

outside of X, I added
session.screen0.antialiasFonts: True
to my ~/.openbox/rc file and then "startx"ed 

put a line like this in a stylefile:
*xft.font:        Nimbus Roman
there are no other "font" lines in the style, and "Nimbus Roman" is
a name straight out of my /etc/X11/XftConfig file. (and it works with
"xterm -fa 'Nimbus Roman'")

then when I choose the style file, openbox says this:
BFont::BFont(): couldn't load font 'Nimbus Roman'
Falling back to default 'fixed'

help!


From smoynes at nexus.carleton.ca  Wed Oct  9 15:42:52 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Wed, 9 Oct 2002 15:42:52 -0400
Subject: [openbox] xft font help!
In-Reply-To: <20021009160638.GA20057@raven.localdomain>
References: <20021009160638.GA20057@raven.localdomain>
Message-ID: <20021009194252.GA20368@marilyn>

* Kevin Geiss (kevin at desertsol.com) wrote:
> I'm trying to get antialiased fonts in openbox, and I need some help.

Have you read http://www.xfce.org/Xft_and_Xfce_mini_Howto.html? Typically,
Xft problems come from configuration mistakes.
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021009/309ddcef/attachment.pgp>

From kevin at desertsol.com  Wed Oct  9 16:01:16 2002
From: kevin at desertsol.com (Kevin Geiss)
Date: Wed, 9 Oct 2002 13:01:16 -0700
Subject: [openbox] xft font help!
In-Reply-To: <20021009194252.GA20368@marilyn>
References: <20021009160638.GA20057@raven.localdomain> <20021009194252.GA20368@marilyn>
Message-ID: <20021009200116.GA27409@raven.localdomain>

I checked all the stuff there. the only thing I didn't check was if there
is a configure flag for openbox to use antialiasing at compile time. other
than that, I have everything listed on that page set up correctly. I
get antialiasing with QT, xterm -fa, and gtk (thanks to gdkxft).

Are there only certain font families that openbox will antialias?

On Wed, Oct 09, 2002 at 03:42:52PM -0400, Scott Moynes wrote:
> * Kevin Geiss (kevin at desertsol.com) wrote:
> > I'm trying to get antialiased fonts in openbox, and I need some help.
> 
> Have you read http://www.xfce.org/Xft_and_Xfce_mini_Howto.html? Typically,
> Xft problems come from configuration mistakes.


From smoynes at nexus.carleton.ca  Wed Oct  9 16:10:48 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Wed, 9 Oct 2002 16:10:48 -0400
Subject: [openbox] xft font help!
In-Reply-To: <20021009200116.GA27409@raven.localdomain>
References: <20021009160638.GA20057@raven.localdomain> <20021009194252.GA20368@marilyn> <20021009200116.GA27409@raven.localdomain>
Message-ID: <20021009201048.GB20368@marilyn>

* Kevin Geiss (kevin at desertsol.com) wrote:
> Are there only certain font families that openbox will antialias?

If everything is configured properly, it will use them. Check the website for
fonts used and known to work with the shipped styles, and look at that
XftConfig
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021009/6ae46b69/attachment.pgp>

From ben at orodu.net  Wed Oct  9 16:41:20 2002
From: ben at orodu.net (Ben Jansens)
Date: Wed, 9 Oct 2002 15:41:20 -0500
Subject: [openbox] xft font help!
In-Reply-To: <20021009160638.GA20057@raven.localdomain>
References: <20021009160638.GA20057@raven.localdomain>
Message-ID: <20021009204120.GC17750@orodu.net>

On Wed, Oct 09, 2002 at 09:06:38AM -0700, Kevin Geiss wrote:
> long time blackbox user, drawn to openbox for the antialiased fonts and
> net wm compliance. But:
> 
> I'm trying to get antialiased fonts in openbox, and I need some help.
> 
> First, these are things I have done:
> 
> outside of X, I added
> session.screen0.antialiasFonts: True
> to my ~/.openbox/rc file and then "startx"ed 

No, need, the default is true. And the old "you have to exit blackbox
to edit your rc file" philosophy does not hold with Openbox. Edit it
and hit reconfigure. Easy as that.
 
> put a line like this in a stylefile:
> *xft.font:        Nimbus Roman
> there are no other "font" lines in the style, and "Nimbus Roman" is
> a name straight out of my /etc/X11/XftConfig file. (and it works with
> "xterm -fa 'Nimbus Roman'")

A font cannot be specified without a size.
*xft.size: 10
or something like that, and then things should start working out
better for you.

Take a look at the styles shipped with Openbox for examples.

> then when I choose the style file, openbox says this:
> BFont::BFont(): couldn't load font 'Nimbus Roman'
> Falling back to default 'fixed'
> 
> help!

Ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021009/047bffe9/attachment.pgp>

From kevin at desertsol.com  Wed Oct  9 17:59:51 2002
From: kevin at desertsol.com (Kevin Geiss)
Date: Wed, 9 Oct 2002 14:59:51 -0700
Subject: [openbox] xft font help!
In-Reply-To: <20021009204120.GC17750@orodu.net>
References: <20021009160638.GA20057@raven.localdomain> <20021009204120.GC17750@orodu.net>
Message-ID: <20021009215951.GA31005@raven.localdomain>

that did it! the font size is what I was missing. I found the '*xft.font:'
line out of one of the examples, but didn't think I needed the size too.

thanks!

On Wed, Oct 09, 2002 at 03:41:20PM -0500, Ben Jansens wrote:
> A font cannot be specified without a size.
> *xft.size: 10
> or something like that, and then things should start working out
> better for you.
> 
> Take a look at the styles shipped with Openbox for examples.
> 


From ben at orodu.net  Thu Oct 10 10:40:00 2002
From: ben at orodu.net (Ben Jansens)
Date: Thu, 10 Oct 2002 09:40:00 -0500
Subject: [Diebert.Van.Rhijn@esa.int: openbox 2.1.1 on Solaris 8 = success]
Message-ID: <20021010144000.GD17750@orodu.net>

From: Diebert.Van.Rhijn at esa.int
Subject: openbox 2.1.1 on Solaris 8 = success
To: ben at orodu.net
Delivery-date: Thu, 10 Oct 2002 02:52:52 -0500
X-Lotus-FromDomain: ESA
X-Spam-Level: 
X-Spam-Status: No, hits=0.6 required=5.0 tests=NO_REAL_NAME version=2.20
X-Spam-Level: 

Hi, your great project Openbox 2.1.1 is compiling and running smoothly on my
Solaris 8 machine (Sparc Ultra 10 / 256 MB) just as version 1.2.2 by the way.

Using gcc 3.03 and gnu bintools.

Thank you very much!

Diebert
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021010/acb9833a/attachment.pgp>

From awol at home.nl  Fri Oct 11 07:14:37 2002
From: awol at home.nl (Alexander Volovics)
Date: Fri, 11 Oct 2002 13:14:37 +0200
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <Pine.LNX.4.44.0210090034260.28553-100000@maestro.hl.lu.se>
References: <20021008185925.GA1253@home.nl> <Pine.LNX.4.44.0210090034260.28553-100000@maestro.hl.lu.se>
Message-ID: <20021011111437.GB1419@home.nl>

On Wed, Oct 09, 2002 at 12:36:51AM +0200, John Eckerdal wrote:

> Dang!
>
> I forgot something vitally importat!
>
> You will not get Xft support unless modify config.h after you execute
> configure.
>
> There is a line probably something like:
> /* #undef XFT */
>
> Change that one to:
> #define XFT 1
>
> This is a must since configure never finds Xft and will therefor remove
> support for it. Sorry, don't know why I forgot that one.
  
Dear John,
  
That did the trick!
I have a working openbox now with xft support.
The TT fonts appear nice enough.
  
Thank you very much for your time and help.
  
I have one last question. Openbox and epist compiled and work OK but
not bsetbg, bsetroot and xftlsfonts. Maybe I was not complete/correct
enough in adapting the Makefiles.
Do you get working binaries of these.
  
(There were still some statements about "undeclared" things like:
 "FcListFontsPatternObjects", "struct _FcPattern", "XftValueList",
 "FcTypecharset", "FcTypeTTFace", etc)
  
Alexander



From john at hl.lu.se  Fri Oct 11 07:44:50 2002
From: john at hl.lu.se (John Eckerdal)
Date: Fri, 11 Oct 2002 13:44:50 +0200 (CEST)
Subject: [openbox] compiling openbox-2.1.2 on redhat-8.0
In-Reply-To: <20021011111437.GB1419@home.nl>
Message-ID: <Pine.LNX.4.44.0210111340330.15861-100000@maestro.hl.lu.se>

On Fri, 11 Oct 2002, Alexander Volovics wrote:

> Dear John,
>   
> That did the trick!
> I have a working openbox now with xft support.
> The TT fonts appear nice enough.
>   
> Thank you very much for your time and help.
>   
> I have one last question. Openbox and epist compiled and work OK but
> not bsetbg, bsetroot and xftlsfonts. Maybe I was not complete/correct
> enough in adapting the Makefiles.
> Do you get working binaries of these.

No.

As far as I know bsetbg is a script (atleast it was in blackbox, bsetroot
compiles OK here atleast), xftlsfonts fails to compile probably due to 
incompatibilities between Xft1 and Xft2.

A make install should install all the correctly compiled utilities.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e





From marius at cs.pdx.edu  Tue Oct 15 21:39:36 2002
From: marius at cs.pdx.edu (Marius Nita)
Date: Tue, 15 Oct 2002 18:39:36 -0700
Subject: Ninja Foo & Spaghetti Sauce
Message-ID: <20021016013935.GI6161@cs.pdx.edu>

So regarding Openbox 2. Woodblock and I talked about further development a
little bit this morning, and I want to outline some thoughts/ideas that would
be good[tm] but which may also seem far fetched and weird, bummy-like. Anyway.

Otk
----

We need a toolkit. Xor and I talked about this long long ago before openbox
existed; we need a non-sucky way to build visual WM elements. Current ob/bb
sucks hard at this, as many items are drawn "in place"--instead of saying "hey,
nice class that i wrote, pls give me this kind of visual element" it says "oh
i need to draw xyz. let's use the X primitives right here in the middle of
this 1200 line function to draw it." That's bad, weird, and weird. Fast, you
say? PFFT!

We need high level classes like the following:

	OtkLabel
	OtkButton
	OtkTitle
	OtkMenu
	...

which rely on low level classes like the following:

	OtkTexture
	OtkFont
	OtkTextContainer
	OtkImageContainer
	OtkImage
	...

Some would probably be abstract, and there would be an inheritance hierarchy
with OtkObject at the top, probably. This hierarchy will not be huge, and the
classes will stay small.

Then, putting a toolbar on the screen, would be something like (inside the
Toolbar class):

  OtkTitle *title = new OtkTitle(x, y, width, height);
  OtkLabel *wksp = new OtkLabel(small_lbl_width, lbl_height);
  OtkLabel *win_name = new OtkLabel(lbl_width, lbl_height);
  OtkLabel *clock = new OtkLabel(small_lbl_width, lbl_height);
  OtkButton *leftwksp = new OtkButton(btn_width, lbl_height,
                                      foo->getPixmap(OtkButton::LeftArrow));
  OtkButton *rightwksp = new OtkButton(btn_width, lbl_height,
                                       foo->getPixmap(OtkButton::RightArrow));
  // same for next_win, prev_win

  title->setJustify(Otk::JustifyLeft);
  title->addItem(wksp);
  title->addItem(leftwksp);
  title->addItem(rightwksp);
  title->addItem(win_name);
  title->addItem(prev_win);
  title->addItem(next_win);
  title->addItem(clock);

  wksp->setText(current_workspace_name);
  clock->setText(clock_string);
  win_name->setText(getCurrentWindow()->getLabel()->getText());

  title->map();

  // or something like that.

Far fetched? Probably. Cool? Damn right. You can now imagine how the rest of
openbox' visual presence gets put together. Now, the toolkit would be
incorporated into libob. The obtools can then take full advantage of the
drawing niceties that openbox takes advantage of. They'd be using exactly the
same code. Menus are themselves composed of Otk elements, and are offered as
high level elements themselves. You can instantiate menus, add items to them,
even nest them in other structures. You can nest a menu into the toolbar, or a
toolbar like element.

The toolkit collaborates with the style engine under the hood. To keep
backward compatibility with blackbox styles, we'd need our Toolbar to access
the style engine, since it's a style element. However, a consistent style
would specify style elements for the titles, labels, buttons, frame, and the
other ob specific elements: grip, etc. The toolbar and window styles would be
deduced from these lower level specifications. even currently, the
overwhelming majority of styles dupe large amounts of style data so they
achieve consistency... However, i can understand the need for fine touches and
the various elements of the WM, and this should be able to be accomplished.
Even then, you would specify the primitive info, and if you say, want to add a
text shadow to the toolbar only, then that's the only additional element you
need to specify. Basically, this would eliminate redundancy from styles.

Now, since this toolkit is so versatile, the toolbar can easily be removed
from the window manager, and written in a couple of hundred of lines of code,
using libob and libob's netwm and Otk capabilities. Same goes for any obtool.
Obtools will write the code that composes their look from these toolkit
primitives, and the look will be enforced by libob. They won't be able to
change say, the gradient type/color of OtkLabel, but they will be able to
customize things where we find suitable to allow them.

Removing the drawing nastiness from the WM core will simplify things GREATLY,
making ob2 more fun to hack on in the future, and extremely versatile and easy
to modify and improve.

This is sort of like the engine concept in ob3, but not quite. There will be a
"main engine" and that is openbox itself. Openbox uses the toolkit to
construct its visual elements. So the separation between decoration and
Openbox is not as concrete as in ob3. We could separate the decoration code
from the WM code, as in different dirs, but the decoration code wouldn't be in
an .so or anything...

A better class hierarchy in the WM
----------------------------------

The current arrangement of classes in ob2 sucks. We all know this, we complain
about it all the time. It's part of what has driven us to start ob3. To
immediately improve ob2, we need 2 things:

	BaseWindow
	XEventHandler

Now these are affected by the toolkit above to a certain extent, but the
problems we need to address are:

	- We need to be able to treat most of the screen elements as Windows,
	  such as the slit, and that little annoying thing that displays the
	  coordinates. The current way of handling stacking is a nightmare.
	  The slit, toolbar, menu, little annoying coordinate thing, all need
	  a concrete place in the Z-stack so they need to be manipulated
	  similarly. Right now, it's a nasty combination of XLower* XRaise*
	  and throwing the ob-specific windows on top of the other windows
	  when the screen stacks things up.

	- We need event dispatch to not suck. searching through all sorts of
	  lists to find the window you need is bull. The dispatcher need not
	  know which windows handle which events. You get the event, you
	  get the element, you call handleEvent() on it. Done.

Other stuff
-----------

blackbox.cc and its relationship to BaseDisplay is a mess. We need to separate
these things, and probably move event dispatch into BaseDisplay.

And um, more later, as I think about it.

marius



From smoynes at nexus.carleton.ca  Tue Oct 15 21:59:40 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Tue, 15 Oct 2002 21:59:40 -0400
Subject: [openbox] Ninja Foo & Spaghetti Sauce
In-Reply-To: <20021016013935.GI6161@cs.pdx.edu>
References: <20021016013935.GI6161@cs.pdx.edu>
Message-ID: <20021016015940.GB2306@marilyn>

* Marius Nita (marius at cs.pdx.edu) wrote:
> We need a toolkit.

I agree. The only problem I could see is making it too big. I figure,
if we limit the interfaces to non-interactive (except maybe a button)
it won't grow to a monster bigger than openbox itself. I figure, what,
90% of QT is interactive elements, and virtually none of them are
required for the types of helper applications (pagers, tabs, toolbars,
dockapps) that would be stapled on to openbox. In fact, the only text
entry in openbox is the workspace name change in the toolbar, and it
still sucks.

What do you think?

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021015/a6e3dd50/attachment.pgp>

From smoynes at nexus.carleton.ca  Tue Oct 15 22:29:14 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Tue, 15 Oct 2002 22:29:14 -0400
Subject: Fireworks
Message-ID: <20021016022914.GD2306@marilyn>

In the spirit of the season, I thought I'd give anyone who cares an
idea of where I think I will be heading with openbox. I have a few
feature ideas, and would like some feedback, yeigh or neigh, on them.

- drop shadows: Make them not suck. First, make them per-style element
  so you can have drop shadows on the windows, but not the
  menus. Also, allow the style to specify the orientation of the
  shadow a little better. Maybe allow specifying the color to
  use. This causes some problems as a translucent color is used, so if
  I can get feedback from people who care, that'd help steer it.

- Menu wraping: Currently, if you have a really long menu, it will
  wrap to multiple columns if it bigger than the screen. However,
  people's screens are far too large, so you can specify a hard limit
  on some of the menus. Boring? Yes. Why do it? Because I hate having
  a long menu. QED.

- It doesn't appear like it'd be too difficult to add translucency to
  all textures. I don't know if such a thing would be wanted
  though. Seems sort of silly to me.

- I have a crazy idea to attempt to implement tabbed windows, or
  tabbed window-like functionality with a third-party app using
  netwm. Maybe it can't be done. Maybe it can. Maybe I will be
  kidnapped by billy goats. Two of three options will give me love
  from unexpected sources, and I like those chances.

- More stuff. If you have any ideas, slap them on me because I am
  sitting in maple syrup, so they are likely to stick.

Peace, love, and capitalism,
scott
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021015/7554b45c/attachment.pgp>

From mkeadle at gentoo.org  Wed Oct 16 02:10:21 2002
From: mkeadle at gentoo.org (Matt Keadle)
Date: Wed, 16 Oct 2002 01:10:21 -0500
Subject: [openbox] Fireworks
In-Reply-To: <20021016022914.GD2306@marilyn>
References: <20021016022914.GD2306@marilyn>
Message-ID: <20021016011021.10521d9e.mkeadle@gentoo.org>

On Tue, 15 Oct 2002 22:29:14 -0400
Scott Moynes <smoynes at nexus.carleton.ca> wrote:

> - It doesn't appear like it'd be too difficult to add translucency to
>   all textures. I don't know if such a thing would be wanted
>   though. Seems sort of silly to me.

Openbox is what it is partially because it's fast and liteweight. I tried Waimea and it certainly was flashy, but I didn't appreciate the fact that my cpu would spike to ~40% by moving translucent menus/titlebars/etc around. It's great eyecandy, but I have to keep editing style definitions to make solids, just so I can get some work done. And once you take that candy out of Waimea, there wasn't much more of a point to using it. Add it if you like, but imo it steals a bit from that fast-n-liteweight department.

> - I have a crazy idea to attempt to implement tabbed windows, or
>   tabbed window-like functionality with a third-party app using
>   netwm. Maybe it can't be done. Maybe it can. Maybe I will be
>   kidnapped by billy goats. Two of three options will give me love
>   from unexpected sources, and I like those chances.

I used Fluxbox for a loooong time and got VERY used to the tabs. I spend time in a handful of different wms, but I always end up back in Flux a lot for those tabs. Flux has tabs, Moz has tabs, now Konq has tabs iirc. I won't say yes to tab development here, but i will to " tabbed window-like functionality". I think there's an opportunity here to devise a new form of tab functionality.
 
> - More stuff. If you have any ideas, slap them on me because I am
>   sitting in maple syrup, so they are likely to stick.

Always on top! Am I missing something, or is this a difficult thing to do? Show me a box that has something "tab-like" and an always-on-top feature and I'll show you a convert. Currently I'm looking at Pekwm, which has both, but doesn't have anything slit-like for my wmdocapps.

-- 
Matt Keadle
Gentoo Linux Developer
mkeadle at gentoo.org
http://mkeadle.org/


From ben at orodu.net  Wed Oct 16 02:01:11 2002
From: ben at orodu.net (Ben Jansens)
Date: Wed, 16 Oct 2002 01:01:11 -0500
Subject: [openbox] Fireworks
In-Reply-To: <20021016011021.10521d9e.mkeadle@gentoo.org>
References: <20021016022914.GD2306@marilyn> <20021016011021.10521d9e.mkeadle@gentoo.org>
Message-ID: <20021016060111.GI293@orodu.net>

On Wed, Oct 16, 2002 at 01:10:21AM -0500, Matt Keadle wrote:
> On Tue, 15 Oct 2002 22:29:14 -0400
> Scott Moynes <smoynes at nexus.carleton.ca> wrote:
> 
> > - It doesn't appear like it'd be too difficult to add translucency to
> >   all textures. I don't know if such a thing would be wanted
> >   though. Seems sort of silly to me.
> 
> Openbox is what it is partially because it's fast and liteweight. I tried Waimea and it certainly was flashy, but I didn't appreciate the fact that my cpu would spike to ~40% by moving translucent menus/titlebars/etc around. It's great eyecandy, but I have to keep editing style definitions to make solids, just so I can get some work done. And once you take that candy out of Waimea, there wasn't much more of a point to using it. Add it if you like, but imo it steals a bit from that fast-n-liteweight department.

A definate need if this is added, is a global config option to
enable/disable translucency. That way noone has to edit themes on
their non-p9-5.9GHz machines.

> > - I have a crazy idea to attempt to implement tabbed windows, or
> >   tabbed window-like functionality with a third-party app using
> >   netwm. Maybe it can't be done. Maybe it can. Maybe I will be
> >   kidnapped by billy goats. Two of three options will give me love
> >   from unexpected sources, and I like those chances.
> 
> I used Fluxbox for a loooong time and got VERY used to the tabs. I spend time in a handful of different wms, but I always end up back in Flux a lot for those tabs. Flux has tabs, Moz has tabs, now Konq has tabs iirc. I won't say yes to tab development here, but i will to " tabbed window-like functionality". I think there's an opportunity here to devise a new form of tab functionality.

No real comment here, except, plz don't do it the way flux did :>

> > - More stuff. If you have any ideas, slap them on me because I am
> >   sitting in maple syrup, so they are likely to stick.
> 
> Always on top! Am I missing something, or is this a difficult thing to do? Show me a box that has something "tab-like" and an always-on-top feature and I'll show you a convert. Currently I'm looking at Pekwm, which has both, but doesn't have anything slit-like for my wmdocapps.

Shrimpx (Marius) has code for this. It'd be good to see that added in
some fashion for the next TESTING release. I, too, would enjoy that ability.

ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021016/0b3b6a66/attachment.pgp>

From awol at home.nl  Wed Oct 16 05:45:22 2002
From: awol at home.nl (Alexander Volovics)
Date: Wed, 16 Oct 2002 11:45:22 +0200
Subject: [openbox] Fireworks
In-Reply-To: <20021016022914.GD2306@marilyn>
References: <20021016022914.GD2306@marilyn>
Message-ID: <20021016094522.GA1392@home.nl>

On Tue, Oct 15, 2002 at 10:29:14PM -0400, Scott Moynes wrote:

> In the spirit of the season, I thought I'd give anyone who cares an
> idea of where I think I will be heading with openbox. I have a few
> feature ideas, and would like some feedback, yeigh or neigh, on them.

Featureism, feature creap, is always a nice topic for heated and long
discussions.

May I put in a plug for some more basic things:
uniformly working UTF-8 and Xft2 support!

At the moment this is based on the old xft (=xft1) included in XFree86-4.2.
However we already have Xft2 and if I understand it correctly ongoing Xft
development is not going to build on xft1.

Alexander



From ben at orodu.net  Wed Oct 16 05:53:04 2002
From: ben at orodu.net (Ben Jansens)
Date: Wed, 16 Oct 2002 04:53:04 -0500
Subject: Docker
Message-ID: <20021016095303.GK293@orodu.net>

I've released a program some of you Openboxer's might be interested
in.

Docker! It functions as a KDE system tray, in the Slit. So you can use
KDE system tray applications without running Kicker (the KDE panel).

There's a fm.net project page for it at
http://freshmeat.net/projects/docker
And it can be wgetted from
http://icculus.org/openbox/releases/docker-1.0.tar.gz

It depends on glib-2.0, so make sure you have it installed first.

Later,
Ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021016/dab45957/attachment.pgp>

From smoynes at nexus.carleton.ca  Wed Oct 16 07:38:04 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Wed, 16 Oct 2002 07:38:04 -0400
Subject: [openbox] Fireworks
In-Reply-To: <20021016094522.GA1392@home.nl>
References: <20021016022914.GD2306@marilyn> <20021016094522.GA1392@home.nl>
Message-ID: <20021016113804.GA4998@marilyn>

* Alexander Volovics (awol at home.nl) wrote:
> At the moment this is based on the old xft (=xft1) included in
> XFree86-4.2.  However we already have Xft2 and if I understand it
> correctly ongoing Xft development is not going to build on xft1.

I thought about this, but most distros don't ship with xft2 so we'd
have to support both which could get ugly. I'll look into it though.
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021016/94588098/attachment.pgp>

From awol at home.nl  Wed Oct 16 11:19:01 2002
From: awol at home.nl (Alexander Volovics)
Date: Wed, 16 Oct 2002 17:19:01 +0200
Subject: [openbox] Fireworks
In-Reply-To: <20021016113804.GA4998@marilyn>
References: <20021016022914.GD2306@marilyn> <20021016094522.GA1392@home.nl> <20021016113804.GA4998@marilyn>
Message-ID: <20021016151901.GA1139@home.nl>

On Wed, Oct 16, 2002 at 07:38:04AM -0400, Scott Moynes wrote:

> * Alexander Volovics (awol at home.nl) wrote:
> > At the moment this is based on the old xft (=xft1) included in
> > XFree86-4.2.  However we already have Xft2 and if I understand it
> > correctly ongoing Xft development is not going to build on xft1.
 
> I thought about this, but most distros don't ship with xft2 so we'd
> have to support both which could get ugly. I'll look into it though.

Indeed, as far as I know only RH-8.0 has switched to Xft2/fontconfig.

But it is already in XFree86-CVS, Gnome2-CVS, Ximian-gnome-CVS.
qt-3.1 and gtk+-2.0.5 support it and all recent Gtk apps (abiword,
galeon, gftp, etc). So you are going to encounter it sooner than later.
(if you use Debian it is definitely going to be later {^_-})

It makes installing and configuring freetype fonts almost as easy as dropping
them into, for example, ~/.fonts and you get "client side fonts" not the
wooden X server variety. (no more messing around with something like xftconfig).

I would appreciate very much if you could look into it.

Alexander





From smoynes at nexus.carleton.ca  Wed Oct 16 11:39:14 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Wed, 16 Oct 2002 11:39:14 -0400
Subject: [openbox] Fireworks
In-Reply-To: <20021016151901.GA1139@home.nl>
References: <20021016022914.GD2306@marilyn> <20021016094522.GA1392@home.nl> <20021016113804.GA4998@marilyn> <20021016151901.GA1139@home.nl>
Message-ID: <20021016153914.GA5885@marilyn>

* Alexander Volovics (awol at home.nl) wrote:
> Indeed, as far as I know only RH-8.0 has switched to Xft2/fontconfig.
> 
> But it is already in XFree86-CVS, Gnome2-CVS, Ximian-gnome-CVS.
> qt-3.1 and gtk+-2.0.5 support it and all recent Gtk apps (abiword,
> galeon, gftp, etc). So you are going to encounter it sooner than later.
> (if you use Debian it is definitely going to be later {^_-})
> 
> It makes installing and configuring freetype fonts almost as easy as dropping
> them into, for example, ~/.fonts and you get "client side fonts" not the
> wooden X server variety. (no more messing around with something like xftconfig).
> 

Well, first, since I run Debian stable, it would be a pain to develop. Also, 
the number of people have been unable to configure "AA text" properly, and 
blame openbox is astounding. To venture into having 3 separate font rendering
paths would cause me to pull my hair out trying to write documentation and
support users. I will look into what others have done, but until I can
easily install it, I don't think I'll actually be able to develop for xft2.
That said, if anyone has patches, they'll happily live in a CVS branch, and
I'd love to have them.

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021016/2cf679f7/attachment.pgp>

From ben at orodu.net  Wed Oct 16 12:46:13 2002
From: ben at orodu.net (Ben Jansens)
Date: Wed, 16 Oct 2002 11:46:13 -0500
Subject: [openbox] Fireworks
In-Reply-To: <20021016153914.GA5885@marilyn>
References: <20021016022914.GD2306@marilyn> <20021016094522.GA1392@home.nl> <20021016113804.GA4998@marilyn> <20021016151901.GA1139@home.nl> <20021016153914.GA5885@marilyn>
Message-ID: <20021016164612.GN293@orodu.net>

On Wed, Oct 16, 2002 at 11:39:14AM -0400, Scott Moynes wrote:
> * Alexander Volovics (awol at home.nl) wrote:
> > Indeed, as far as I know only RH-8.0 has switched to Xft2/fontconfig.
> > 
> > But it is already in XFree86-CVS, Gnome2-CVS, Ximian-gnome-CVS.
> > qt-3.1 and gtk+-2.0.5 support it and all recent Gtk apps (abiword,
> > galeon, gftp, etc). So you are going to encounter it sooner than later.
> > (if you use Debian it is definitely going to be later {^_-})
> > 
> > It makes installing and configuring freetype fonts almost as easy as dropping
> > them into, for example, ~/.fonts and you get "client side fonts" not the
> > wooden X server variety. (no more messing around with something like xftconfig).
> > 
> 
> Well, first, since I run Debian stable, it would be a pain to develop. Also, 
> the number of people have been unable to configure "AA text" properly, and 
> blame openbox is astounding. To venture into having 3 separate font rendering
> paths would cause me to pull my hair out trying to write documentation and
> support users. I will look into what others have done, but until I can
> easily install it, I don't think I'll actually be able to develop for xft2.
> That said, if anyone has patches, they'll happily live in a CVS branch, and
> I'd love to have them.

ts pretty easy to install. Grab the tgzs off keithp's site. I've got
it installed on my machine. It eve comes with an updated version of
Xrender! :)

Ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021016/5d9818e5/attachment.pgp>

From smoynes at nexus.carleton.ca  Wed Oct 16 12:57:41 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Wed, 16 Oct 2002 12:57:41 -0400
Subject: [openbox] Fireworks
In-Reply-To: <20021016164612.GN293@orodu.net>
References: <20021016022914.GD2306@marilyn> <20021016094522.GA1392@home.nl> <20021016113804.GA4998@marilyn> <20021016151901.GA1139@home.nl> <20021016153914.GA5885@marilyn> <20021016164612.GN293@orodu.net>
Message-ID: <20021016165741.GB5885@marilyn>

* Ben Jansens (ben at orodu.net) wrote:
> ts pretty easy to install. Grab the tgzs off keithp's site. I've got
> it installed on my machine. It eve comes with an updated version of
> Xrender! :)

Yeah, I'll check it out. I'm a little wary of trying to support it if it ends
up that RedHat has one version, the rest of the world has another, and 80%
of people have to grab a tarball to install.

cashmoneyblingbucks
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021016/23530704/attachment.pgp>

From xor at orodu.net  Fri Oct 18 06:39:43 2002
From: xor at orodu.net (Ben Jansens)
Date: Fri, 18 Oct 2002 05:39:43 -0500
Subject: Docker 1.3
Message-ID: <20021018103943.GY293@orodu.net>

Hello all,

I've been putting some time into this Docker app, and I'm proud to
report that it can do Gnome2 system tray icons now as well as KDE.

I've also put up a nice webpage for the application at:
http://icculus.org/openbox/docker/

Thought I'd let you all know in case anyone was interested.

Ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021018/1b22012a/attachment.pgp>

From john at hl.lu.se  Sat Oct 19 19:02:55 2002
From: john at hl.lu.se (John Eckerdal)
Date: Sun, 20 Oct 2002 01:02:55 +0200 (CEST)
Subject: Dropshadows for fonts?
Message-ID: <Pine.LNX.4.44.0210200046290.13283-100000@maestro.hl.lu.se>


Hi!

Have been playing around with Openbox for awhile now and like it.
It works like a charm (well, almost). I can't get drop shadows to
work.
I searched the source for any clues on how to enable these and I
found the session.screenX.dropShadowFonts (and corresponding menu
item). None of them make a difference (it can be enabled and disabled
but there is no visible difference). Normal anti-alias fonts work.

Since the default Openbox compilation does not support Xft 2, I sort
of hacked it. After I ./configure has completed config.h is modified so 
that XFT is defined and then I added a two calls in the Makefile to
xft-config.
The net result is that aa-fonts work, drop-shadows do not work.

GCC is RH8.0 gcc-3.2
libstdc++ is RH8.0 libstdc++-3.2
Xft is RH8.0 Xft 2.0
XFree86 is Rawhide XFree86-4.2.1

Either I have missed something trivial in the styles files (don't
think so but I am not entirely sure), or dropshadows fonts do not
work with Xft 2?

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From smoynes at nexus.carleton.ca  Sun Oct 20 02:13:14 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Sun, 20 Oct 2002 02:13:14 -0400
Subject: [openbox] Dropshadows for fonts?
In-Reply-To: <Pine.LNX.4.44.0210200046290.13283-100000@maestro.hl.lu.se>
References: <Pine.LNX.4.44.0210200046290.13283-100000@maestro.hl.lu.se>
Message-ID: <20021020061314.GB10640@marilyn>

* John Eckerdal (john at hl.lu.se) wrote:
> 
> Hi!
> 
> Have been playing around with Openbox for awhile now and like it.
> It works like a charm (well, almost). I can't get drop shadows to
> work.
> I searched the source for any clues on how to enable these and I
> found the session.screenX.dropShadowFonts (and corresponding menu
> item). None of them make a difference (it can be enabled and disabled
> but there is no visible difference). Normal anti-alias fonts work.
> 
> Since the default Openbox compilation does not support Xft 2, I sort
> of hacked it. After I ./configure has completed config.h is modified so 
> that XFT is defined and then I added a two calls in the Makefile to
> xft-config.
> The net result is that aa-fonts work, drop-shadows do not work.
> 
> GCC is RH8.0 gcc-3.2
> libstdc++ is RH8.0 libstdc++-3.2
> Xft is RH8.0 Xft 2.0
> XFree86 is Rawhide XFree86-4.2.1
> 
> Either I have missed something trivial in the styles files (don't
> think so but I am not entirely sure), or dropshadows fonts do not
> work with Xft 2?
> 

I'd like to have any patches for XFT2 you have. Partly to debug
this, partly to look at, and partly because I am a nerd and like that
sort of thing. I suspect it might be due to XFT2, but who knows?
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021020/87f2873e/attachment.pgp>

From neuhauser at bellavista.cz  Thu Oct 24 06:48:31 2002
From: neuhauser at bellavista.cz (Roman Neuhauser)
Date: Thu, 24 Oct 2002 12:48:31 +0200
Subject: assertion failed (coredump)
Message-ID: <20021024104831.GD29294@freepuppy.bellavista.cz>

here's the promised backtrace.

i'm not sure it will be usefull, however: i was dumb enough to wipe the
original bt (not sent yet at that time) *after* updating the sources,
without knowing the exact age of the original checkout.
i experimented with probable values for cvs up -D, and this *seems* to
be the right version. luckily enough, the HEAD was really quiet at that
time.

bt doesn't look very useful, so there's bt full as well.

Blackbox::process_event(): MapRequest for 0x1e0002d
BlackboxWindow::BlackboxWindow(): creating 0x1e0002d
BlackboxWindow::mapRequestEvent() for 0x1e0002d
assertion "dw >= base_width && dh >= base_height" failed: file "Window.cc", line 4232
Oct 23 16:14:50 freepuppy /kernel: pid 9611 (openbox), uid 1001: exited on signal 6 (core dumped)
epist: Restarting on request.
X connection to :0.0 broken (explicit kill or server shutdown).
Gdk-ERROR **: X connection to :0.0 broken (explicit kill or server shutdown).


-- 
If you cc me or take the list(s) out completely I'll most likely
ignore your message.
-------------- next part --------------
roman at freepuppy ~ 1003:0 > gdb install/openbox-head/src/openbox openbox.core
GNU gdb 4.18 (FreeBSD)
Copyright 1998 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386-unknown-freebsd"...Deprecated bfd_read called at /usr/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 2627 in elfstab_build_psymtabs
Deprecated bfd_read called at /usr/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 933 in fill_symbuf


warning: exec file is newer than core file.
Core was generated by `openbox'.
Program terminated with signal 6, Abort trap.
Reading symbols from /usr/X11R6/lib/libSM.so.6...done.
Reading symbols from /usr/X11R6/lib/libICE.so.6...done.
Reading symbols from /usr/X11R6/lib/libX11.so.6...done.
Reading symbols from /usr/X11R6/lib/libXft.so.1...done.
Reading symbols from /usr/lib/libxpg4.so.3...done.
Reading symbols from /usr/lib/libstdc++.so.3...done.
Reading symbols from /usr/lib/libm.so.2...done.
Reading symbols from /usr/lib/libc.so.4...done.
Reading symbols from /usr/X11R6/lib/libXThrStub.so.6...done.
Reading symbols from /usr/X11R6/lib/libXrender.so.1...done.
Reading symbols from /usr/X11R6/lib/libXext.so.6...done.
Reading symbols from /usr/local/lib/libfreetype.so.9...done.
Reading symbols from /usr/X11R6/lib/X11/locale/common/xlcDef.so.2...done.
Reading symbols from /usr/libexec/ld-elf.so.1...done.
#0  0x28268e60 in kill () from /usr/lib/libc.so.4
(gdb) bt
#0  0x28268e60 in kill () from /usr/lib/libc.so.4
#1  0x282aa1d1 in abort () from /usr/lib/libc.so.4
#2  0x28285f2b in __assert () from /usr/lib/libc.so.4
#3  0x809103f in BlackboxWindow::constrain (this=0x80e0600, anchor=TopLeft, pw=0x0, ph=0x0) at Window.cc:4232
#4  0x808e119 in BlackboxWindow::configureRequestEvent (this=0x80e0600, cr=0xbfbff8dc) at Window.cc:3080
#5  0x809b4ba in Blackbox::process_event (this=0xbfbff988, e=0xbfbff8dc) at blackbox.cc:269
#6  0x804e4f6 in BaseDisplay::eventLoop (this=0xbfbff988) at BaseDisplay.cc:318
#7  0x80a5f36 in main (argc=3, argv=0xbfbffb20) at main.cc:190

(gdb) bt full
#0  0x28268e60 in kill () from /usr/lib/libc.so.4
No symbol table info available.
#1  0x282aa1d1 in abort () from /usr/lib/libc.so.4
No symbol table info available.
#2  0x28285f2b in __assert () from /usr/lib/libc.so.4
No symbol table info available.
#3  0x809103f in BlackboxWindow::constrain (this=0x80e0600, anchor=TopLeft, pw=0x0, ph=0x0) at Window.cc:4232
        dw = 1
        dh = 1
        base_width = 597
        base_height = 231
        dx = 134860721
        dy = 1
#4  0x808e119 in BlackboxWindow::configureRequestEvent (this=0x80e0600, cr=0xbfbff8dc) at Window.cc:3080
        this = (BlackboxWindow *) 0x80e0600
        cr = (XConfigureRequestEvent *) 0xbfbff8dc
#5  0x809b4ba in Blackbox::process_event (this=0xbfbff988, e=0xbfbff8dc) at blackbox.cc:269
        win = (BlackboxWindow *) 0x0
        slit = (Slit *) 0x0
#6  0x804e4f6 in BaseDisplay::eventLoop (this=0xbfbff988) at BaseDisplay.cc:318
        xfd = 3
#7  0x80a5f36 in main (argc=3, argv=0xbfbffb20) at main.cc:190
        session_display = 0x0
        rc_file = 0xbfbffc39 "/home/roman/.blackbox/blackboxrc"
        menu_file = 0x0
        blackbox = {<BaseDisplay> = {<TimerQueueManager> = {_vptr$ = 0x80abf20}, shape = {extensions = false, event_basep = 134922644, error_basep = -1077937660}, 
    MaskList = {0, 2, 16, 18, 128, 130, 144, 146}, MaskListLength = 8, run_state = RUNNING, display = 0x80b1000, gccache = 0x80bda80, 
    screenInfoList = {<_Vector_base<ScreenInfo,allocator<ScreenInfo> >> = {<_Vector_alloc_base<ScreenInfo,allocator<ScreenInfo>,>> = {_M_start = 0x80b3ab0, 
          _M_finish = 0x80b3adc, _M_end_of_storage = 0x80b3adc}, <No data fields>}, <No data fields>}, 
    timerList = {<priority_queue<BTimer *,vector<BTimer *,allocator<BTimer *> >,TimerLessThan>> = {
        c = {<_Vector_base<BTimer *,allocator<BTimer *> >> = {<_Vector_alloc_base<BTimer *,allocator<BTimer *>,>> = {_M_start = 0x80b0a18, _M_finish = 0x80b0a1c, 
              _M_end_of_storage = 0x80b0a20}, <No data fields>}, <No data fields>}, comp = {<No data fields>}}, <No data fields>}, display_name = 0xbfbffdd6 ":0", 
    application_name = 0xbfbffc10 "/home/roman/openbox-head/bin/openbox", NumLockMask = 16, ScrollLockMask = 128}, <TimeoutHandler> = {_vptr$ = 0x80abefc}, cursor = {
    session = 8388610, move = 8388611, ll_angle = 8388612, lr_angle = 8388613, ul_angle = 8388614, ur_angle = 8388615}, resource = {double_click_interval = 250, 
    style_file = {static npos = 4294967295, static nilRep = {len = 0, res = 0, ref = 122, selfish = false}, dat = 0x80b7b80 "/home/roman/.blackbox/styles/Cynic"}, 
    colors_per_channel = 4, auto_raise_delay = {tv_sec = 0, tv_usec = 250000}, cache_life = 300000, cache_max = 200, titlebar_layout = {static npos = 4294967295, 
      static nilRep = {len = 0, res = 0, ref = 122, selfish = false}, dat = 0x80b3830 "ILMCpenbox/menu"}, mod_mask = 8}, windowSearchList = {
    _M_t = {<_Rb_tree_base<pair<const long unsigned int,BlackboxWindow *>,allocator<BlackboxWindow *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,BlackboxWindow *>,allocator<BlackboxWindow *>,>> = {_M_header = 0x80b7800}, <No data fields>}, _M_node_count = 131, 
      _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, systraySearchList = {
    _M_t = {<_Rb_tree_base<pair<const long unsigned int,BScreen *>,allocator<BScreen *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,BScreen *>,allocator<BScreen *>,allocator<BlackboxWindow *>,>> = {_M_header = 0x80b7800}, <No data fields>}, _M_node_count = 131, 
      _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, systraySearchList = {
    _M_t = {<_Rb_tree_base<pair<const long unsigned int,BScreen *>,allocator<BScreen *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,BScreen *>,allocator<BScreen *>,>> = {_M_header = 0x80b7818}, <No data fields>}, _M_node_count = 0, 
      _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, groupSearchList = {
    _M_t = {<_Rb_tree_base<pair<const long unsigned int,BWindowGroup *>,allocator<BWindowGroup *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,BWindowGroup *>,allocator<BWindowGroup *>,>> = {_M_header = 0x80b7830}, <No data fields>}, _M_node_count = 10, 
      _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, menuSearchList = {
    _M_t = {<_Rb_tree_base<pair<const long unsigned int,Basemenu *>,allocator<Basemenu *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,Basemenu *>,allocator<Basemenu *>,>> = {_M_header = 0x80b7848}, <No data fields>}, _M_node_count = 150, 
      _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, toolbarSearchList = {
    _M_t = {<_Rb_tree_base<pair<const long unsigned int,Toolbar *>,allocator<Toolbar *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,Toolbar *>,allocator<Toolbar *>,>> = {_M_header = 0x80b7860}, <No data fields>}, _M_node_count = 8, 
      _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, slitSearchList = {
    _M_t = {<_Rb_tree_base<pair<const long unsigned int,Slit *>,allocator<Slit *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,Slit *>,allocator<Slit *>,>> = {_M_header = 0x80b7878}, <No data fields>}, _M_node_count = 1, 
      _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, menuTimestamps = {<_List_base<Blackbox::MenuTimestamp *,allocator<Blackbox::MenuTimestamp *> >> = {<_List_alloc_base<Blackbox::MenuTimestamp *,allocator<Blackbox::MenuTimestamp *>,>> = {_M_node = 0x80b3cf0}, <No data fields>}, <No data fields>}, 
  screenList = {<_List_base<BScreen *,allocator<BScreen *> >> = {<_List_alloc_base<BScreen *,allocator<BScreen *>,>> = {
        _M_node = 0x80b79e0}, <No data fields>}, <No data fields>}, active_screen = 0x80be000, focused_window = 0x80e0600, changing_window = 0x0, timer = 0x80c77e0, 
  config = {static _initialized = true, _file = {static npos = 4294967295, static nilRep = {len = 0, res = 0, ref = 122, selfish = false}, 
      dat = 0x80b7b30 "/home/roman/.blackbox/blackboxrc"}, _modified = false, _autosave = true, _database = 0x80b4cc0, _vptr$ = 0x80ab9c4}, xatom = 0x80b0800, 
  no_focus = false, reconfigure_wait = false, reread_menu_wait = false, last_time = 295371656, argv = 0xbfbffb20, menu_file = {static npos = 4294967295, static nilRep = {
      len = 0, res = 0, ref = 122, selfish = false}, dat = 0x80b3af0 "/home/roman/.openbox/menu"}, rc_file = {static npos = 4294967295, static nilRep = {len = 0, res = 0, 
      ref = 122, selfish = false}, dat = 0x80b7b30 "/home/roman/.blackbox/blackboxrc"}}



From ben at orodu.net  Thu Oct 24 10:45:48 2002
From: ben at orodu.net (Ben Jansens)
Date: Thu, 24 Oct 2002 09:45:48 -0500
Subject: [openbox] assertion failed (coredump)
In-Reply-To: <20021024104831.GD29294@freepuppy.bellavista.cz>
References: <20021024104831.GD29294@freepuppy.bellavista.cz>
Message-ID: <20021024144548.GD293@orodu.net>

From what i can tell this is simply a misplaced assert.

What app caused this?

Ben

On Thu, Oct 24, 2002 at 12:48:31PM +0200, Roman Neuhauser wrote:
> here's the promised backtrace.
> 
> i'm not sure it will be usefull, however: i was dumb enough to wipe the
> original bt (not sent yet at that time) *after* updating the sources,
> without knowing the exact age of the original checkout.
> i experimented with probable values for cvs up -D, and this *seems* to
> be the right version. luckily enough, the HEAD was really quiet at that
> time.
> 
> bt doesn't look very useful, so there's bt full as well.
> 
> Blackbox::process_event(): MapRequest for 0x1e0002d
> BlackboxWindow::BlackboxWindow(): creating 0x1e0002d
> BlackboxWindow::mapRequestEvent() for 0x1e0002d
> assertion "dw >= base_width && dh >= base_height" failed: file "Window.cc", line 4232
> Oct 23 16:14:50 freepuppy /kernel: pid 9611 (openbox), uid 1001: exited on signal 6 (core dumped)
> epist: Restarting on request.
> X connection to :0.0 broken (explicit kill or server shutdown).
> Gdk-ERROR **: X connection to :0.0 broken (explicit kill or server shutdown).
> 
> 

> roman at freepuppy ~ 1003:0 > gdb install/openbox-head/src/openbox openbox.core
> GNU gdb 4.18 (FreeBSD)
> Copyright 1998 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you are
> welcome to change it and/or distribute copies of it under certain conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for details.
> This GDB was configured as "i386-unknown-freebsd"...Deprecated bfd_read called at /usr/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 2627 in elfstab_build_psymtabs
> Deprecated bfd_read called at /usr/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 933 in fill_symbuf
> 
> 
> warning: exec file is newer than core file.
> Core was generated by `openbox'.
> Program terminated with signal 6, Abort trap.
> Reading symbols from /usr/X11R6/lib/libSM.so.6...done.
> Reading symbols from /usr/X11R6/lib/libICE.so.6...done.
> Reading symbols from /usr/X11R6/lib/libX11.so.6...done.
> Reading symbols from /usr/X11R6/lib/libXft.so.1...done.
> Reading symbols from /usr/lib/libxpg4.so.3...done.
> Reading symbols from /usr/lib/libstdc++.so.3...done.
> Reading symbols from /usr/lib/libm.so.2...done.
> Reading symbols from /usr/lib/libc.so.4...done.
> Reading symbols from /usr/X11R6/lib/libXThrStub.so.6...done.
> Reading symbols from /usr/X11R6/lib/libXrender.so.1...done.
> Reading symbols from /usr/X11R6/lib/libXext.so.6...done.
> Reading symbols from /usr/local/lib/libfreetype.so.9...done.
> Reading symbols from /usr/X11R6/lib/X11/locale/common/xlcDef.so.2...done.
> Reading symbols from /usr/libexec/ld-elf.so.1...done.
> #0  0x28268e60 in kill () from /usr/lib/libc.so.4
> (gdb) bt
> #0  0x28268e60 in kill () from /usr/lib/libc.so.4
> #1  0x282aa1d1 in abort () from /usr/lib/libc.so.4
> #2  0x28285f2b in __assert () from /usr/lib/libc.so.4
> #3  0x809103f in BlackboxWindow::constrain (this=0x80e0600, anchor=TopLeft, pw=0x0, ph=0x0) at Window.cc:4232
> #4  0x808e119 in BlackboxWindow::configureRequestEvent (this=0x80e0600, cr=0xbfbff8dc) at Window.cc:3080
> #5  0x809b4ba in Blackbox::process_event (this=0xbfbff988, e=0xbfbff8dc) at blackbox.cc:269
> #6  0x804e4f6 in BaseDisplay::eventLoop (this=0xbfbff988) at BaseDisplay.cc:318
> #7  0x80a5f36 in main (argc=3, argv=0xbfbffb20) at main.cc:190
> 
> (gdb) bt full
> #0  0x28268e60 in kill () from /usr/lib/libc.so.4
> No symbol table info available.
> #1  0x282aa1d1 in abort () from /usr/lib/libc.so.4
> No symbol table info available.
> #2  0x28285f2b in __assert () from /usr/lib/libc.so.4
> No symbol table info available.
> #3  0x809103f in BlackboxWindow::constrain (this=0x80e0600, anchor=TopLeft, pw=0x0, ph=0x0) at Window.cc:4232
>         dw = 1
>         dh = 1
>         base_width = 597
>         base_height = 231
>         dx = 134860721
>         dy = 1
> #4  0x808e119 in BlackboxWindow::configureRequestEvent (this=0x80e0600, cr=0xbfbff8dc) at Window.cc:3080
>         this = (BlackboxWindow *) 0x80e0600
>         cr = (XConfigureRequestEvent *) 0xbfbff8dc
> #5  0x809b4ba in Blackbox::process_event (this=0xbfbff988, e=0xbfbff8dc) at blackbox.cc:269
>         win = (BlackboxWindow *) 0x0
>         slit = (Slit *) 0x0
> #6  0x804e4f6 in BaseDisplay::eventLoop (this=0xbfbff988) at BaseDisplay.cc:318
>         xfd = 3
> #7  0x80a5f36 in main (argc=3, argv=0xbfbffb20) at main.cc:190
>         session_display = 0x0
>         rc_file = 0xbfbffc39 "/home/roman/.blackbox/blackboxrc"
>         menu_file = 0x0
>         blackbox = {<BaseDisplay> = {<TimerQueueManager> = {_vptr$ = 0x80abf20}, shape = {extensions = false, event_basep = 134922644, error_basep = -1077937660}, 
>     MaskList = {0, 2, 16, 18, 128, 130, 144, 146}, MaskListLength = 8, run_state = RUNNING, display = 0x80b1000, gccache = 0x80bda80, 
>     screenInfoList = {<_Vector_base<ScreenInfo,allocator<ScreenInfo> >> = {<_Vector_alloc_base<ScreenInfo,allocator<ScreenInfo>,>> = {_M_start = 0x80b3ab0, 
>           _M_finish = 0x80b3adc, _M_end_of_storage = 0x80b3adc}, <No data fields>}, <No data fields>}, 
>     timerList = {<priority_queue<BTimer *,vector<BTimer *,allocator<BTimer *> >,TimerLessThan>> = {
>         c = {<_Vector_base<BTimer *,allocator<BTimer *> >> = {<_Vector_alloc_base<BTimer *,allocator<BTimer *>,>> = {_M_start = 0x80b0a18, _M_finish = 0x80b0a1c, 
>               _M_end_of_storage = 0x80b0a20}, <No data fields>}, <No data fields>}, comp = {<No data fields>}}, <No data fields>}, display_name = 0xbfbffdd6 ":0", 
>     application_name = 0xbfbffc10 "/home/roman/openbox-head/bin/openbox", NumLockMask = 16, ScrollLockMask = 128}, <TimeoutHandler> = {_vptr$ = 0x80abefc}, cursor = {
>     session = 8388610, move = 8388611, ll_angle = 8388612, lr_angle = 8388613, ul_angle = 8388614, ur_angle = 8388615}, resource = {double_click_interval = 250, 
>     style_file = {static npos = 4294967295, static nilRep = {len = 0, res = 0, ref = 122, selfish = false}, dat = 0x80b7b80 "/home/roman/.blackbox/styles/Cynic"}, 
>     colors_per_channel = 4, auto_raise_delay = {tv_sec = 0, tv_usec = 250000}, cache_life = 300000, cache_max = 200, titlebar_layout = {static npos = 4294967295, 
>       static nilRep = {len = 0, res = 0, ref = 122, selfish = false}, dat = 0x80b3830 "ILMCpenbox/menu"}, mod_mask = 8}, windowSearchList = {
>     _M_t = {<_Rb_tree_base<pair<const long unsigned int,BlackboxWindow *>,allocator<BlackboxWindow *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,BlackboxWindow *>,allocator<BlackboxWindow *>,>> = {_M_header = 0x80b7800}, <No data fields>}, _M_node_count = 131, 
>       _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, systraySearchList = {
>     _M_t = {<_Rb_tree_base<pair<const long unsigned int,BScreen *>,allocator<BScreen *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,BScreen *>,allocator<BScreen *>,allocator<BlackboxWindow *>,>> = {_M_header = 0x80b7800}, <No data fields>}, _M_node_count = 131, 
>       _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, systraySearchList = {
>     _M_t = {<_Rb_tree_base<pair<const long unsigned int,BScreen *>,allocator<BScreen *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,BScreen *>,allocator<BScreen *>,>> = {_M_header = 0x80b7818}, <No data fields>}, _M_node_count = 0, 
>       _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, groupSearchList = {
>     _M_t = {<_Rb_tree_base<pair<const long unsigned int,BWindowGroup *>,allocator<BWindowGroup *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,BWindowGroup *>,allocator<BWindowGroup *>,>> = {_M_header = 0x80b7830}, <No data fields>}, _M_node_count = 10, 
>       _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, menuSearchList = {
>     _M_t = {<_Rb_tree_base<pair<const long unsigned int,Basemenu *>,allocator<Basemenu *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,Basemenu *>,allocator<Basemenu *>,>> = {_M_header = 0x80b7848}, <No data fields>}, _M_node_count = 150, 
>       _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, toolbarSearchList = {
>     _M_t = {<_Rb_tree_base<pair<const long unsigned int,Toolbar *>,allocator<Toolbar *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,Toolbar *>,allocator<Toolbar *>,>> = {_M_header = 0x80b7860}, <No data fields>}, _M_node_count = 8, 
>       _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, slitSearchList = {
>     _M_t = {<_Rb_tree_base<pair<const long unsigned int,Slit *>,allocator<Slit *> >> = {<_Rb_tree_alloc_base<pair<const long unsigned int,Slit *>,allocator<Slit *>,>> = {_M_header = 0x80b7878}, <No data fields>}, _M_node_count = 1, 
>       _M_key_compare = {<binary_function<long unsigned int,long unsigned int,bool>> = {<No data fields>}, <No data fields>}}}, menuTimestamps = {<_List_base<Blackbox::MenuTimestamp *,allocator<Blackbox::MenuTimestamp *> >> = {<_List_alloc_base<Blackbox::MenuTimestamp *,allocator<Blackbox::MenuTimestamp *>,>> = {_M_node = 0x80b3cf0}, <No data fields>}, <No data fields>}, 
>   screenList = {<_List_base<BScreen *,allocator<BScreen *> >> = {<_List_alloc_base<BScreen *,allocator<BScreen *>,>> = {
>         _M_node = 0x80b79e0}, <No data fields>}, <No data fields>}, active_screen = 0x80be000, focused_window = 0x80e0600, changing_window = 0x0, timer = 0x80c77e0, 
>   config = {static _initialized = true, _file = {static npos = 4294967295, static nilRep = {len = 0, res = 0, ref = 122, selfish = false}, 
>       dat = 0x80b7b30 "/home/roman/.blackbox/blackboxrc"}, _modified = false, _autosave = true, _database = 0x80b4cc0, _vptr$ = 0x80ab9c4}, xatom = 0x80b0800, 
>   no_focus = false, reconfigure_wait = false, reread_menu_wait = false, last_time = 295371656, argv = 0xbfbffb20, menu_file = {static npos = 4294967295, static nilRep = {
>       len = 0, res = 0, ref = 122, selfish = false}, dat = 0x80b3af0 "/home/roman/.openbox/menu"}, rc_file = {static npos = 4294967295, static nilRep = {len = 0, res = 0, 
>       ref = 122, selfish = false}, dat = 0x80b7b30 "/home/roman/.blackbox/blackboxrc"}}
> 
> 



-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021024/fb998ce0/attachment.pgp>

From neuhauser at bellavista.cz  Thu Oct 24 11:01:58 2002
From: neuhauser at bellavista.cz (Roman Neuhauser)
Date: Thu, 24 Oct 2002 17:01:58 +0200
Subject: [openbox] assertion failed (coredump)
In-Reply-To: <20021024144548.GD293@orodu.net>
References: <20021024104831.GD29294@freepuppy.bellavista.cz> <20021024144548.GD293@orodu.net>
Message-ID: <20021024150158.GI29294@freepuppy.bellavista.cz>

# ben at orodu.net / 2002-10-24 09:45:48 -0500:
> On Thu, Oct 24, 2002 at 12:48:31PM +0200, Roman Neuhauser wrote:
> > here's the promised backtrace.
> > 
> > i'm not sure it will be usefull, however: i was dumb enough to wipe the
> > original bt (not sent yet at that time) *after* updating the sources,
> > without knowing the exact age of the original checkout.
> > i experimented with probable values for cvs up -D, and this *seems* to
> > be the right version. luckily enough, the HEAD was really quiet at that
> > time.
> 
> From what i can tell this is simply a misplaced assert.
> 
> What app caused this?

    not sure (i was doing two or three things at once), but i *think* ob
    died when:

    1. i went to my online bank, mozilla popped up the msgbox
       "this ssl page contains some non-ssl elements, do you want to
       proceed?"
    2. i clicked the ok button on the msgbox
    3. i went to another workspace

    i'm not sure 2. actually happened, but ISTR mozilla was in the
    process of drawing or destroying a window. :|

-- 
If you cc me or take the list(s) out completely I'll most likely
ignore your message.


From ben at orodu.net  Fri Oct 25 12:53:46 2002
From: ben at orodu.net (Ben Jansens)
Date: Fri, 25 Oct 2002 11:53:46 -0500
Subject: [aconchillo@acm.org: Anaphora is an Epistrophy configurator]
Message-ID: <20021025165346.GE293@orodu.net>

----- Forwarded message from Aleix Conchillo Flaque <aconchillo at acm.org> -----

From: Aleix Conchillo Flaque <aconchillo at acm.org>
Subject: Anaphora is an Epistrophy configurator
To: ben at orodu.net
Delivery-date: Fri, 25 Oct 2002 11:07:11 -0500
X-Spam-Level: 
X-Spam-Status: No, hits=0.0 required=5.0 tests= version=2.20
X-Spam-Level: 

hi,

i'm writing to you because i'd like to contribute in some way to
Openbox. The way i'm trying to help is by making a graphical epistrophy
configurator, is called anaphora. (curiosity: an anaphora is the antonym
of an epistrophy).

how am i doing it?

first of all i have created an XML configuration file for anaphora wich
contains all options listed in 'man epistrc' (i've attached the
file). these options will be presented to the user in a graphical tool,
that will allow the user to save an epistrc file. and of course it will
also allow to configure the grabbed keys.

libraries used:

i've choosen expat (for xml reading) and gtk2 for the gui.

status?

i've already load the xml file and in a few days i will have the gui and
something more. i just work on it at nights, but i'm trying to do my
best.

homepage?

i've created a project in savannah:

http://savannah.nongnu.org/projects/anaphora

well, that's it. i hope it will be useful for many people, at least it
will help me :).

i'll inform you when i have something visible. i hope it will be very
soon.

best regards,

aleix




----- End forwarded message -----


-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021025/7a9f7413/attachment.pgp>

From marius at cs.pdx.edu  Fri Oct 25 13:21:39 2002
From: marius at cs.pdx.edu (Marius Nita)
Date: Fri, 25 Oct 2002 10:21:39 -0700
Subject: [openbox] [aconchillo@acm.org: Anaphora is an Epistrophy configurator]
In-Reply-To: <20021025165346.GE293@orodu.net>
References: <20021025165346.GE293@orodu.net>
Message-ID: <20021025172139.GA8074@cs.pdx.edu>

On Fri, Oct 25, 2002 at 11:53:46AM -0500, Ben Jansens wrote:
> i'm writing to you because i'd like to contribute in some way to
> Openbox. The way i'm trying to help is by making a graphical epistrophy
> configurator, is called anaphora. (curiosity: an anaphora is the antonym
> of an epistrophy).

Wow, this is way cool. The initial architecture sounds awesome, being based on
an xml file... I assume that adding a new option/action would be a matter of
editing this xml file, at least to a large extent. 

Anyway, cool! Keep it up.

marius

> how am i doing it?
> 
> first of all i have created an XML configuration file for anaphora wich
> contains all options listed in 'man epistrc' (i've attached the
> file). these options will be presented to the user in a graphical tool,
> that will allow the user to save an epistrc file. and of course it will
> also allow to configure the grabbed keys.
> 
> libraries used:
> 
> i've choosen expat (for xml reading) and gtk2 for the gui.
> 
> status?
> 
> i've already load the xml file and in a few days i will have the gui and
> something more. i just work on it at nights, but i'm trying to do my
> best.
> 
> homepage?
> 
> i've created a project in savannah:
> 
> http://savannah.nongnu.org/projects/anaphora
> 
> well, that's it. i hope it will be useful for many people, at least it
> will help me :).
> 
> i'll inform you when i have something visible. i hope it will be very
> soon.
> 
> best regards,
> 
> aleix
> 
> 
> 
> 
> ----- End forwarded message -----
> 
> 
> -- 
> I am damn unsatisfied to be killed in this way.




From smoynes at nexus.carleton.ca  Fri Oct 25 21:19:35 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Fri, 25 Oct 2002 21:19:35 -0400
Subject: The shadow boys are breaking all the laws
Message-ID: <20021026011935.GG24275@marilyn>

In attempt to slaughter the menu redraw bug I've been seeing I came up
with a partial solution. It is ugly and probably wrong, but I hope it
gives someone an idea how to fix it for good because I am lost.

First, to reproduce the bug consistently:
 - open the root menu
 - tear off the workspaces menu
 - on the root menu, highlight the styles entry and move the mouse off
 without losing the highlight
 - drag the torn off menu over the root menu.
 - watch the expose bummy up

After this patch it kind of works. It doesn't work on the left/right
expose. On the right it will lose the bullet, and on the left it will
lose the text. Also, with xft you can really see the flicker, so it
might be prudent to throw it all out and come up with some sort of
futuristic caching mechanism that uses quantum flux capacitors and
emoticon engines or something.

Regards,
Scott

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
? menu-redraw-idea.diff
Index: Basemenu.cc
===================================================================
RCS file: /cvs/cvsroot/openbox/src/Basemenu.cc,v
retrieving revision 1.35
diff -p -u -r1.35 Basemenu.cc
--- Basemenu.cc	2002/10/17 02:37:23	1.35
+++ Basemenu.cc	2002/10/26 00:59:38
@@ -528,6 +528,8 @@ void Basemenu::drawItem(int index, bool 
 
   bool dotext = True, dohilite = True, dosel = True, dooppsel = True;
   const char *text = item->label();
+  cout << text << endl;
+  bool cleared = false;
   const int sbl = index / menu.persub, i = index - (sbl * menu.persub);
   const unsigned int half_w = menu.item_h / 2, quarter_w = menu.item_h / 4;
   int item_x = (sbl * menu.item_w), item_y = (i * menu.item_h);
@@ -579,12 +581,12 @@ void Basemenu::drawItem(int index, bool 
   if (clear) {
     XClearArea(display, menu.frame, item_x, item_y, menu.item_w, menu.item_h,
                False);
+    cout << "clear1"<<endl;
   } else if (! (x == y && y == -1 && w == h && h == 0)) {
     // calculate the which part of the hilite to redraw
     if (! (max(item_x, x) <= min<signed>(item_x + menu.item_w, x + w) &&
            max(item_y, y) <= min<signed>(item_y + menu.item_h, y + h))) {
       dohilite = False;
-    } else {
       hilite_x = max(item_x, x);
       hilite_y = max(item_y, y);
       hilite_w = min(item_x + menu.item_w, x + w) - hilite_x;
@@ -613,10 +615,15 @@ void Basemenu::drawItem(int index, bool 
 
 #ifdef    XFT
   if (dotext)
-    XClearArea(display, menu.frame, text_x, text_y , text_w, text_h, False);
+    {
+      XClearArea(display, menu.frame, text_x, text_y , text_w, text_h, False);
+      cout << "clear2" << endl;
+      cleared = true;
+    }
+  
 #endif // XFT
-
-  if (dohilite && highlight && (menu.hilite_pixmap != ParentRelative)) {
+  cout << dohilite << dotext << highlight << (menu.hilite_pixmap != ParentRelative) << endl;
+  if ( (dohilite || cleared) && highlight && (menu.hilite_pixmap != ParentRelative)) {
     if (menu.hilite_pixmap)
       XCopyArea(display, menu.hilite_pixmap, menu.frame,
                 hipen.gc(), hoff_x, hoff_y,
@@ -626,7 +633,7 @@ void Basemenu::drawItem(int index, bool 
                      hilite_x, hilite_y, hilite_w, hilite_h);
   }
 
-  if (dooppsel && item->isSelected()) {
+  if ( (dooppsel || cleared) && item->isSelected()) {
 #ifdef    BITMAPBUTTONS
     if ( style->tick_image.mask != None) {
       XSetClipOrigin(blackbox->getXDisplay(), pen.gc(),
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021025/cc4920af/attachment.pgp>

From dems at sdimaps.com  Tue Oct 29 10:12:25 2002
From: dems at sdimaps.com (Daniel Ems)
Date: Tue, 29 Oct 2002 10:12:25 -0500 (EST)
Subject: RedHat 8.0
Message-ID: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com>

I apologize if this has been asked before - I couldn't find a way to view 
the mailing list archives.

I can't compile openbox 2.1.2 on RedHat 8.0

The configure works correctly, but make fails.  Here's the error:

g++ -DHAVE_CONFIG_H -I. -I. -I..   -DNDEBUG 
-DLOCALEPATH=\"/usr/local/share/openbox/nls\" 
-DDEFAULTMENU=\"/usr/local/share/openbox/menu\" 
-DDEFAULTSTYLE=\"/usr/local/share/openbox/styles/mbdtex\"  -g -O2  
-I/usr/X11R6/include -Wall -W -c Texture.cc
g++ -DHAVE_CONFIG_H -I. -I. -I..   -DNDEBUG 
-DLOCALEPATH=\"/usr/local/share/openbox/nls\" 
-DDEFAULTMENU=\"/usr/local/share/openbox/menu\" 
-DDEFAULTSTYLE=\"/usr/local/share/openbox/styles/mbdtex\"  -g -O2  
-I/usr/X11R6/include -Wall -W -c Timer.cc
g++ -DHAVE_CONFIG_H -I. -I. -I..   -DNDEBUG 
-DLOCALEPATH=\"/usr/local/share/openbox/nls\" 
-DDEFAULTMENU=\"/usr/local/share/openbox/menu\" 
-DDEFAULTSTYLE=\"/usr/local/share/openbox/styles/mbdtex\"  -g -O2  
-I/usr/X11R6/include -Wall -W -c Toolbar.cc
Toolbar.cc: In constructor `Toolbarmenu::Toolbarmenu(Toolbar*)':
Toolbar.cc:1021: output_operand: `%l' operand isn't a label
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://bugzilla.redhat.com/bugzilla/> for instructions.
make[2]: *** [Toolbar.o] Error 1
make[2]: Leaving directory `/usr/local/src/openbox-2.1.2/src'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/usr/local/src/openbox-2.1.2'
make: *** [all-recursive-am] Error 2


Any thoughts?  Or should I just file a bug report?  What more inforamtion 
do you need?

-- 
Daniel Ems
Database Administrator
Spatial Data Integrations, Inc.



From smoynes at nexus.carleton.ca  Tue Oct 29 11:31:39 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Tue, 29 Oct 2002 11:31:39 -0500
Subject: [openbox] RedHat 8.0
In-Reply-To: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com>
References: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com>
Message-ID: <20021029163139.GA10471@marilyn>

* Daniel Ems (dems at sdimaps.com) wrote:
> I apologize if this has been asked before - I couldn't find a way to view 
> the mailing list archives.
> 
> I can't compile openbox 2.1.2 on RedHat 8.0
> 
> The configure works correctly, but make fails.  Here's the error:
> 
> g++ -DHAVE_CONFIG_H -I. -I. -I..   -DNDEBUG 
> -DLOCALEPATH=\"/usr/local/share/openbox/nls\" 
> -DDEFAULTMENU=\"/usr/local/share/openbox/menu\" 
> -DDEFAULTSTYLE=\"/usr/local/share/openbox/styles/mbdtex\"  -g -O2  
> -I/usr/X11R6/include -Wall -W -c Texture.cc
> g++ -DHAVE_CONFIG_H -I. -I. -I..   -DNDEBUG 
> -DLOCALEPATH=\"/usr/local/share/openbox/nls\" 
> -DDEFAULTMENU=\"/usr/local/share/openbox/menu\" 
> -DDEFAULTSTYLE=\"/usr/local/share/openbox/styles/mbdtex\"  -g -O2  
> -I/usr/X11R6/include -Wall -W -c Timer.cc
> g++ -DHAVE_CONFIG_H -I. -I. -I..   -DNDEBUG 
> -DLOCALEPATH=\"/usr/local/share/openbox/nls\" 
> -DDEFAULTMENU=\"/usr/local/share/openbox/menu\" 
> -DDEFAULTSTYLE=\"/usr/local/share/openbox/styles/mbdtex\"  -g -O2  
> -I/usr/X11R6/include -Wall -W -c Toolbar.cc
> Toolbar.cc: In constructor `Toolbarmenu::Toolbarmenu(Toolbar*)':
> Toolbar.cc:1021: output_operand: `%l' operand isn't a label
> Please submit a full bug report,
> with preprocessed source if appropriate.
> See <URL:http://bugzilla.redhat.com/bugzilla/> for instructions.
> make[2]: *** [Toolbar.o] Error 1
> make[2]: Leaving directory `/usr/local/src/openbox-2.1.2/src'
> make[1]: *** [all-recursive] Error 1
> make[1]: Leaving directory `/usr/local/src/openbox-2.1.2'
> make: *** [all-recursive-am] Error 2
> 
> 
> Any thoughts?  Or should I just file a bug report?  What more inforamtion 
> do you need?
> 

Looks like a compiler error to me. Fill out a bug report with Redhat I
assume. However, it won't help much as we've had many bug reports from
people compiling on Redhat systems, and none of them have been helpful
enough to get it debugged properly.

Scott.
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021029/f067f7c8/attachment.pgp>

From john at hl.lu.se  Tue Oct 29 11:40:13 2002
From: john at hl.lu.se (John Eckerdal)
Date: Tue, 29 Oct 2002 17:40:13 +0100 (CET)
Subject: [openbox] RedHat 8.0
In-Reply-To: <20021029163139.GA10471@marilyn>
Message-ID: <Pine.LNX.4.44.0210291736210.26004-100000@maestro.hl.lu.se>

On Tue, 29 Oct 2002, Scott Moynes wrote:

> Looks like a compiler error to me. Fill out a bug report with Redhat I
> assume. However, it won't help much as we've had many bug reports from
> people compiling on Redhat systems, and none of them have been helpful
> enough to get it debugged properly.

A CVS version compiled on RH 8.0 some days ago worked fine here.

The problem is that you have to manually edit the Makefiles after
configure has created them - otherwise Xft support is missing.
Other than that there are no bugs, except maybe that menu bug that
someone mentioned awhile ago.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From dems at sdimaps.com  Tue Oct 29 10:48:55 2002
From: dems at sdimaps.com (Daniel Ems)
Date: Tue, 29 Oct 2002 10:48:55 -0500 (EST)
Subject: [openbox] RedHat 8.0
In-Reply-To: <Pine.LNX.4.44.0210291736210.26004-100000@maestro.hl.lu.se>
Message-ID: <Pine.LNX.4.44.0210291046550.31741-100000@gatekeeper.sdimaps.com>

On Tue, 29 Oct 2002, John Eckerdal wrote:

> On Tue, 29 Oct 2002, Scott Moynes wrote:
> 
> > Looks like a compiler error to me. Fill out a bug report with Redhat I
> > assume. However, it won't help much as we've had many bug reports from
> > people compiling on Redhat systems, and none of them have been helpful
> > enough to get it debugged properly.
> 
> A CVS version compiled on RH 8.0 some days ago worked fine here.
> 
> The problem is that you have to manually edit the Makefiles after
> configure has created them - otherwise Xft support is missing.
> Other than that there are no bugs, except maybe that menu bug that
> someone mentioned awhile ago.
> 
> 

Thanks for the quick responses.  How complicated is editing the Makefiles 
- I don't have much experience with them.  Could you tell me the steps?  
If I don't make the changes, what will I be missing?

-- 
Daniel Ems
Database Administrator
Spatial Data Integrations, Inc.



From smoynes at nexus.carleton.ca  Tue Oct 29 12:02:09 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Tue, 29 Oct 2002 12:02:09 -0500
Subject: [openbox] RedHat 8.0
In-Reply-To: <Pine.LNX.4.44.0210291736210.26004-100000@maestro.hl.lu.se>
References: <20021029163139.GA10471@marilyn> <Pine.LNX.4.44.0210291736210.26004-100000@maestro.hl.lu.se>
Message-ID: <20021029170209.GA10661@marilyn>

* John Eckerdal (john at hl.lu.se) wrote:
> A CVS version compiled on RH 8.0 some days ago worked fine here.
> 
> The problem is that you have to manually edit the Makefiles after
> configure has created them - otherwise Xft support is missing.
> Other than that there are no bugs, except maybe that menu bug that
> someone mentioned awhile ago.
> 

I've had contrary reports. I believe Redhat has shipped an unreleased
version of Xfree that uses Xft2, and this has broken the drop shadow
fonts.

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021029/5aef071f/attachment.pgp>

From john at hl.lu.se  Tue Oct 29 11:56:01 2002
From: john at hl.lu.se (John Eckerdal)
Date: Tue, 29 Oct 2002 17:56:01 +0100 (CET)
Subject: [openbox] RedHat 8.0
In-Reply-To: <Pine.LNX.4.44.0210291046550.31741-100000@gatekeeper.sdimaps.com>
Message-ID: <Pine.LNX.4.44.0210291747470.26127-100000@maestro.hl.lu.se>

On Tue, 29 Oct 2002, Daniel Ems wrote:

> On Tue, 29 Oct 2002, John Eckerdal wrote:
> 
> > On Tue, 29 Oct 2002, Scott Moynes wrote:
> > 
> > > Looks like a compiler error to me. Fill out a bug report with Redhat I
> > > assume. However, it won't help much as we've had many bug reports from
> > > people compiling on Redhat systems, and none of them have been helpful
> > > enough to get it debugged properly.
> > 
> > A CVS version compiled on RH 8.0 some days ago worked fine here.
> > 
> > The problem is that you have to manually edit the Makefiles after
> > configure has created them - otherwise Xft support is missing.
> > Other than that there are no bugs, except maybe that menu bug that
> > someone mentioned awhile ago.
> > 
> > 
> 
> Thanks for the quick responses.  How complicated is editing the Makefiles 
> - I don't have much experience with them.  Could you tell me the steps?  
> If I don't make the changes, what will I be missing?

Xft support. No anti-alias on fonts.

To do this:

1. ./configure
2. edit src/Makefile
3. Locate openbox_LDFLAGS and append `xft-config --libs`
4. Locate CXXFLAGS and append `xft-config --cflags`
5. edit config.h
6. Locate /* Enable support of the Xft extension */
7. change the line below that one to #define XFT 1
8. Run make


Note that in step 3 & 4 is is note the back-quote not the ' one.

Also note that xlsfonts in the util directory will fail. This does not 
matter I think.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From john at hl.lu.se  Tue Oct 29 11:57:57 2002
From: john at hl.lu.se (John Eckerdal)
Date: Tue, 29 Oct 2002 17:57:57 +0100 (CET)
Subject: [openbox] RedHat 8.0
In-Reply-To: <20021029170209.GA10661@marilyn>
Message-ID: <Pine.LNX.4.44.0210291756110.26127-100000@maestro.hl.lu.se>

On Tue, 29 Oct 2002, Scott Moynes wrote:

> * John Eckerdal (john at hl.lu.se) wrote:
> > A CVS version compiled on RH 8.0 some days ago worked fine here.
> > 
> > The problem is that you have to manually edit the Makefiles after
> > configure has created them - otherwise Xft support is missing.
> > Other than that there are no bugs, except maybe that menu bug that
> > someone mentioned awhile ago.
> > 
> 
> I've had contrary reports. I believe Redhat has shipped an unreleased
> version of Xfree that uses Xft2, and this has broken the drop shadow
> fonts.

Yep. RH ships Xft2.

And dropshadows do not work. It compiles OK though so it's probably
something different with Xft2.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From awol at home.nl  Tue Oct 29 14:51:00 2002
From: awol at home.nl (Alexander Volovics)
Date: Tue, 29 Oct 2002 20:51:00 +0100
Subject: [openbox] RedHat 8.0
In-Reply-To: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com>
References: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com>
Message-ID: <20021029195100.GA1017@home.nl>

On Tue, Oct 29, 2002 at 10:12:25AM -0500, Daniel Ems wrote:

> I can't compile openbox 2.1.2 on RedHat 8.0
> The configure works correctly, but make fails.  Here's the error:

Openbox-2.1.2 does compile on RH-8.0. Some time ago I stated the
same problem om this list and was helped by John Eckerdal.

As you can't access the archives I'll give a summary:

The problem exists because RH-8.0 has changed from Xft (=Xft1)
to Xft2/fontconfig.

Not only have the locations of the relevant libs been changed 
(for example "libXft.so..." is no longer to be found in /usr/X11R6/lib/
 but in /usr/lib/ and "Xft.h" is no longer to be found in
 /usr/X11R6/include/X11/Xft but in /usr/include/Xft2/X11/)
process but also "Xft.h" from Xft-devel in RH-8.0 contains some
other 'structures' than the old Xft.h from XFree86-devel-4.2.0.

The solution is to use the shell script 'xft-config' from Xft-devel
to display the necessary CPPFLAGS, CXXFLAGS and LDFLAGS, namely:

$ xft-config --libs
-lXft2 -lfreetype -lfontconfig -L/usr/X11R6/lib -lXrender
$ xft-config --cflags
-I/usr/include/Xft2 -I/usr/include/freetype2 -I/usr/include -I/usr/X11R6/include

John Eckerdal's solutions was to
1) run ./configure
2) edit 'config.h' and change /*enable support of the Xft extension*/
   from /* #undef XFT */ to #define XFT 1
3) edit Makefile in /src and add:
   `xft-config --libs` to the line "openbox_LDFLAGS ="
   `xft-config --cflags` to the end of the line "CXXFLAGS = -g...."
4) edit Makefile in /util and add:
   `xft-config --cflags` to the end of the line "CXXFLAGS = -g...."
   I also added `xft-config --libs` to the lines with
   "xftlsfonts_LDFLAGS =" and "bsetroot_LDFLAGS ="
   (I don't know if these last two are strictly necessay. xftlsfonts
    refuses to compile anyway)
5) do make and make install

REMARK the double `` in the above lines with xft-config!

This compiles openbox, epist, bsetroot and they then work OK under
RH-8.0 and you can use TT fonts in the styles.
(however "dropShadowFonts" does not seem to work).

There are probably more elegant ways to do this but this works.
However make still seems to grumble about some of the 'structures'
in Xft.h not being recognized. I have not noticed any lack of
functionality in openbox though.

It would be very nice if the developers could look into adding
Xft2/fontconfig support for openbox.
Xft2/fontconfig makes the handling of freetype fonts so much more
elegant and easy under Linux. It is going to be the future anyway.

You can download and install the necessary xft/fontconfig libs under your
own distribution by downloading fcpackage.2_0.tar.gz from fontconfig.org.

Alexander
    



From smoynes at nexus.carleton.ca  Tue Oct 29 15:11:26 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Tue, 29 Oct 2002 15:11:26 -0500
Subject: [openbox] RedHat 8.0
In-Reply-To: <20021029195100.GA1017@home.nl>
References: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com> <20021029195100.GA1017@home.nl>
Message-ID: <20021029201125.GB11487@marilyn>

* Alexander Volovics (awol at home.nl) wrote:
> It would be very nice if the developers could look into adding
> Xft2/fontconfig support for openbox.
> Xft2/fontconfig makes the handling of freetype fonts so much more
> elegant and easy under Linux. It is going to be the future anyway.

Except it hasn't been part of a proper release from Xfree86. The API
is still changing.
 
> You can download and install the necessary xft/fontconfig libs under your
> own distribution by downloading fcpackage.2_0.tar.gz from fontconfig.org.

And it doesn't compile on my system, and I'm not going to jump through
hoops when so few systems have it installed. If someone comes up with
a proper patch that works, I'll apply it, but I don't have the
inclination to try to fix more Redhatisms.

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021029/0876124d/attachment.pgp>

From ben at orodu.net  Tue Oct 29 15:11:09 2002
From: ben at orodu.net (Ben Jansens)
Date: Tue, 29 Oct 2002 14:11:09 -0600
Subject: [openbox] RedHat 8.0
In-Reply-To: <20021029201125.GB11487@marilyn>
References: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com> <20021029195100.GA1017@home.nl> <20021029201125.GB11487@marilyn>
Message-ID: <20021029201109.GA16327@orodu.net>

On Tue, Oct 29, 2002 at 03:11:26PM -0500, Scott Moynes wrote:
> * Alexander Volovics (awol at home.nl) wrote:
> > It would be very nice if the developers could look into adding
> > Xft2/fontconfig support for openbox.
> > Xft2/fontconfig makes the handling of freetype fonts so much more
> > elegant and easy under Linux. It is going to be the future anyway.
> 
> Except it hasn't been part of a proper release from Xfree86. The API
> is still changing.

Xft2 is being separated out from XFree. Thus making it more portable
etc etc, and good.
  
> > You can download and install the necessary xft/fontconfig libs under your
> > own distribution by downloading fcpackage.2_0.tar.gz from fontconfig.org.
> 
> And it doesn't compile on my system, and I'm not going to jump through
> hoops when so few systems have it installed. If someone comes up with
> a proper patch that works, I'll apply it, but I don't have the
> inclination to try to fix more Redhatisms.

It doesn't compile? Tried reading the instructions? :) Compiled fine
on my debian box.

Xft 2 does have a first release version out, with its 'finalized'
API.

Switching Openbox to it would definately be nice. Problem is that so
few systems have it yet. And I'm not fluent enough in autoconf to make
openbox use either at the moment.

Ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021029/214a25f0/attachment.pgp>

From dems at sdimaps.com  Tue Oct 29 14:26:19 2002
From: dems at sdimaps.com (Daniel Ems)
Date: Tue, 29 Oct 2002 14:26:19 -0500 (EST)
Subject: [openbox] RedHat 8.0
In-Reply-To: <20021029195100.GA1017@home.nl>
Message-ID: <Pine.LNX.4.44.0210291421580.7818-100000@gatekeeper.sdimaps.com>


Thank you again for the quick help!

I was able to compile and install version 2.2.0 on RH 8.0.  It worked, but
the fonts were literally unreadably small - they appeared as a thin line.  
I'm sure this is part of the same problem.  So, I tried the steps below
with version 2.1.2 and got the following error:

cc1plus: warning: changing search order for system directory 
"/usr/include"
cc1plus: warning:   as it has already been specified as a non-system 
directory
xftlsfonts.cc: In function `int main(int, char**)':
xftlsfonts.cc:83: `FcListFontsPatternObjects' undeclared (first use this 
   function)
xftlsfonts.cc:83: (Each undeclared identifier is reported only once for 
each 
   function it appears in.)
xftlsfonts.cc:93: invalid use of undefined type `struct _FcPattern'
/usr/include/fontconfig/fontconfig.h:164: forward declaration of `struct 
   _FcPattern'
xftlsfonts.cc:99: invalid use of undefined type `struct _FcPattern'
/usr/include/fontconfig/fontconfig.h:164: forward declaration of `struct 
   _FcPattern'
xftlsfonts.cc:101: `XftValueList' undeclared (first use this function)
xftlsfonts.cc:101: `vallist' undeclared (first use this function)
xftlsfonts.cc:101: invalid use of undefined type `struct _FcPattern'
/usr/include/fontconfig/fontconfig.h:164: forward declaration of `struct 
   _FcPattern'




On Tue, 29 Oct 2002, Alexander Volovics wrote:

> On Tue, Oct 29, 2002 at 10:12:25AM -0500, Daniel Ems wrote:
> 
> > I can't compile openbox 2.1.2 on RedHat 8.0
> > The configure works correctly, but make fails.  Here's the error:
> 
> Openbox-2.1.2 does compile on RH-8.0. Some time ago I stated the
> same problem om this list and was helped by John Eckerdal.
> 
> As you can't access the archives I'll give a summary:
> 
> The problem exists because RH-8.0 has changed from Xft (=Xft1)
> to Xft2/fontconfig.
> 
> Not only have the locations of the relevant libs been changed 
> (for example "libXft.so..." is no longer to be found in /usr/X11R6/lib/
>  but in /usr/lib/ and "Xft.h" is no longer to be found in
>  /usr/X11R6/include/X11/Xft but in /usr/include/Xft2/X11/)
> process but also "Xft.h" from Xft-devel in RH-8.0 contains some
> other 'structures' than the old Xft.h from XFree86-devel-4.2.0.
> 
> The solution is to use the shell script 'xft-config' from Xft-devel
> to display the necessary CPPFLAGS, CXXFLAGS and LDFLAGS, namely:
> 
> $ xft-config --libs
> -lXft2 -lfreetype -lfontconfig -L/usr/X11R6/lib -lXrender
> $ xft-config --cflags
> -I/usr/include/Xft2 -I/usr/include/freetype2 -I/usr/include -I/usr/X11R6/include
> 
> John Eckerdal's solutions was to
> 1) run ./configure
> 2) edit 'config.h' and change /*enable support of the Xft extension*/
>    from /* #undef XFT */ to #define XFT 1
> 3) edit Makefile in /src and add:
>    `xft-config --libs` to the line "openbox_LDFLAGS ="
>    `xft-config --cflags` to the end of the line "CXXFLAGS = -g...."
> 4) edit Makefile in /util and add:
>    `xft-config --cflags` to the end of the line "CXXFLAGS = -g...."
>    I also added `xft-config --libs` to the lines with
>    "xftlsfonts_LDFLAGS =" and "bsetroot_LDFLAGS ="
>    (I don't know if these last two are strictly necessay. xftlsfonts
>     refuses to compile anyway)
> 5) do make and make install
> 
> REMARK the double `` in the above lines with xft-config!
> 
> This compiles openbox, epist, bsetroot and they then work OK under
> RH-8.0 and you can use TT fonts in the styles.
> (however "dropShadowFonts" does not seem to work).
> 
> There are probably more elegant ways to do this but this works.
> However make still seems to grumble about some of the 'structures'
> in Xft.h not being recognized. I have not noticed any lack of
> functionality in openbox though.
> 
> It would be very nice if the developers could look into adding
> Xft2/fontconfig support for openbox.
> Xft2/fontconfig makes the handling of freetype fonts so much more
> elegant and easy under Linux. It is going to be the future anyway.
> 
> You can download and install the necessary xft/fontconfig libs under your
> own distribution by downloading fcpackage.2_0.tar.gz from fontconfig.org.
> 
> Alexander
>     
> 

-- 
Daniel Ems
Database Administrator
Spatial Data Integrations, Inc.



From smoynes at nexus.carleton.ca  Tue Oct 29 15:30:09 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Tue, 29 Oct 2002 15:30:09 -0500
Subject: [openbox] RedHat 8.0
In-Reply-To: <20021029201109.GA16327@orodu.net>
References: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com> <20021029195100.GA1017@home.nl> <20021029201125.GB11487@marilyn> <20021029201109.GA16327@orodu.net>
Message-ID: <20021029203009.GA11608@marilyn>

* Ben Jansens (ben at orodu.net) wrote:
> It doesn't compile? Tried reading the instructions? :) Compiled fine
> on my debian box.

It compiled on your 4.0 Xfree86?

> Xft 2 does have a first release version out, with its 'finalized'
> API.

Funny, I see changes as recent as 2 weeks ago. Not real big ones, but
from what I've seen keithp saying, the API could still change
radically. The code in Xfree86 4.2.1, from what I can tell from CVS
isn't Xft2.

> Switching Openbox to it would definately be nice. Problem is that so
> few systems have it yet. And I'm not fluent enough in autoconf to make
> openbox use either at the moment.

If someone really wants it, they can send me a patch, like I said. 

Oh yeah, Redhat's Xft2 also has memory leaks in it, so those who care
should watch for errata.
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021029/e3f119ba/attachment.pgp>

From smoynes at nexus.carleton.ca  Tue Oct 29 15:33:00 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Tue, 29 Oct 2002 15:33:00 -0500
Subject: [openbox] RedHat 8.0
In-Reply-To: <Pine.LNX.4.44.0210291421580.7818-100000@gatekeeper.sdimaps.com>
References: <20021029195100.GA1017@home.nl> <Pine.LNX.4.44.0210291421580.7818-100000@gatekeeper.sdimaps.com>
Message-ID: <20021029203300.GB11608@marilyn>

* Daniel Ems (dems at sdimaps.com) wrote:
> 
> Thank you again for the quick help!
> 
> I was able to compile and install version 2.2.0 on RH 8.0.  It worked, but
> the fonts were literally unreadably small - they appeared as a thin line.  
> I'm sure this is part of the same problem.  So, I tried the steps below
> with version 2.1.2 and got the following error:

xftlsfonts won't build, but that isn't critical. Removing it from the
all: target in the Makefile in util should skip. (Untested and may be
a member of the Axis of Evil)


-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021029/74508852/attachment.pgp>

From ben at orodu.net  Tue Oct 29 17:05:22 2002
From: ben at orodu.net (Ben Jansens)
Date: Tue, 29 Oct 2002 16:05:22 -0600
Subject: [openbox] RedHat 8.0
In-Reply-To: <20021029203009.GA11608@marilyn>
References: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com> <20021029195100.GA1017@home.nl> <20021029201125.GB11487@marilyn> <20021029201109.GA16327@orodu.net> <20021029203009.GA11608@marilyn>
Message-ID: <20021029220522.GB16327@orodu.net>

On Tue, Oct 29, 2002 at 03:30:09PM -0500, Scott Moynes wrote:
> * Ben Jansens (ben at orodu.net) wrote:
> > It doesn't compile? Tried reading the instructions? :) Compiled fine
> > on my debian box.
> 
> It compiled on your 4.0 Xfree86?

XFree86 Version 4.2.1 (Debian 4.2.1-3 20021016191246
branden at deadbeast.net) / X Window System

Standard debian install, and yep, compiled just fine.

> > Xft 2 does have a first release version out, with its 'finalized'
> > API.
> 
> Funny, I see changes as recent as 2 weeks ago. Not real big ones, but
> from what I've seen keithp saying, the API could still change
> radically. The code in Xfree86 4.2.1, from what I can tell from CVS
> isn't Xft2.

Of course it can..just like the API for any software. Afaik, the code isnt
going to be in Xfree86 at all anymore. So you arent going to see it
there. Its sharing the Xfree cvs site, but thats about it.
 
> > Switching Openbox to it would definately be nice. Problem is that so
> > few systems have it yet. And I'm not fluent enough in autoconf to make
> > openbox use either at the moment.
> 
> If someone really wants it, they can send me a patch, like I said. 
> 
> Oh yeah, Redhat's Xft2 also has memory leaks in it, so those who care
> should watch for errata.

Ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021029/ef343e56/attachment.pgp>

From smoynes at nexus.carleton.ca  Tue Oct 29 22:23:40 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Tue, 29 Oct 2002 22:23:40 -0500
Subject: [openbox] RedHat 8.0
In-Reply-To: <20021029220522.GB16327@orodu.net>
References: <Pine.LNX.4.44.0210290958120.31741-100000@gatekeeper.sdimaps.com> <20021029195100.GA1017@home.nl> <20021029201125.GB11487@marilyn> <20021029201109.GA16327@orodu.net> <20021029203009.GA11608@marilyn> <20021029220522.GB16327@orodu.net>
Message-ID: <20021030032340.GE23295@marilyn>

* Ben Jansens (ben at orodu.net) wrote:
> XFree86 Version 4.2.1 (Debian 4.2.1-3 20021016191246
> branden at deadbeast.net) / X Window System
> 
> Standard debian install, and yep, compiled just fine.

I have whatever is in stable. 4.1 of some sort I guess.

> Of course it can..just like the API for any software. Afaik, the code isnt
> going to be in Xfree86 at all anymore. So you arent going to see it
> there. Its sharing the Xfree cvs site, but thats about it.

Bummy.
In conclusion, send patches or http://icculus.org/openbox/send_pls.php

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021029/30fd02c1/attachment.pgp>

From ajar at katanalynx.dyndns.org  Wed Oct 30 05:12:17 2002
From: ajar at katanalynx.dyndns.org (Ava Arachne Jarvis)
Date: Wed, 30 Oct 2002 04:12:17 -0600
Subject: Patch to fix Epist's buggy 2D workspace
Message-ID: <20021030101217.GD2892@dan0032.urh.uiuc.edu>

The patch is attached.  This fixes three bugs:

  - nextWorkspaceRow went up instead of down, and prevWorkspaceRow went
    down instead of up. (simple change to parser.cc)

  - changeWorkspaceHorz had some issues with the "left-wards" case, 
    caused by calling changeWorkspace once in a special case, then 
    once more at the end of the function.  Oops.  (action.cc)

  - changeWorkspaceVert is conceptually a call to cycleWorkspace with 
    appropriate parameters.  cycleWorkspace seemed to take care of 
    wrapping as well, so I just changed changeWorkspaceVert to call 
    cycleWorkspace....  

Patched against the latest CVS.

-- 
| Women are wiser than men because they know less and understand more.
| 		-- Stephens
-------------- next part --------------
? epist-2dworkspace.patch
Index: util/epist/parser.cc
===================================================================
RCS file: /cvs/cvsroot/openbox/util/epist/parser.cc,v
retrieving revision 1.11
diff -p -u -r1.11 parser.cc
--- util/epist/parser.cc	2002/10/14 12:54:46	1.11
+++ util/epist/parser.cc	2002/10/30 10:02:56
@@ -115,8 +115,8 @@ void parser::setAction(string act)
     { "changeworkspace", Action::changeWorkspace },
     { "nextworkspace", Action::nextWorkspace },
     { "prevworkspace", Action::prevWorkspace },
-    { "nextworkspacerow", Action::upWorkspace },
-    { "prevworkspacerow", Action::downWorkspace },
+    { "nextworkspacerow", Action::downWorkspace },
+    { "prevworkspacerow", Action::upWorkspace },
     { "prevworkspacecolumn", Action::leftWorkspace },
     { "nextworkspacecolumn", Action::rightWorkspace },
     { "nextscreen", Action::nextScreen },
Index: util/epist/screen.cc
===================================================================
RCS file: /cvs/cvsroot/openbox/util/epist/screen.cc,v
retrieving revision 1.83
diff -p -u -r1.83 screen.cc
--- util/epist/screen.cc	2002/10/14 02:50:44	1.83
+++ util/epist/screen.cc	2002/10/30 10:02:56
@@ -706,8 +706,6 @@ void screen::changeWorkspaceVert(const i
   assert(_managed);
   int width = 0;
   int num_desktops = (signed)_num_desktops;
-  int active_desktop = (signed)_active_desktop;
-  int wnum = 0;
 
   _config->getValue(Config::workspaceColumns, width);
 
@@ -715,23 +713,7 @@ void screen::changeWorkspaceVert(const i
     return;
 
   // a cookie to the person that makes this pretty
-  if (num < 0) {
-    wnum = active_desktop - width;
-    if (wnum < 0) {
-      wnum = num_desktops/width * width + active_desktop;
-      if (wnum >= num_desktops)
-        wnum = num_desktops - 1;
-    }
-  }
-  else {
-    wnum = active_desktop + width;
-    if (wnum >= num_desktops) {
-      wnum = (active_desktop + width) % num_desktops - 1;
-      if (wnum < 0)
-        wnum = 0;
-    }
-  }
-  changeWorkspace(wnum);
+  cycleWorkspace(num > 0, width, true);
 }
 
 void screen::changeWorkspaceHorz(const int num) const {
@@ -748,7 +730,7 @@ void screen::changeWorkspaceHorz(const i
 
   if (num < 0) {
     if (active_desktop % width != 0)
-      changeWorkspace(active_desktop - 1);
+      wnum = active_desktop - 1;
     else {
       wnum = active_desktop + width - 1;
       if (wnum >= num_desktops)
@@ -764,6 +746,7 @@ void screen::changeWorkspaceHorz(const i
     else
       wnum = active_desktop - width + 1;
   }
+
   changeWorkspace(wnum);
 }
 

From smoynes at nexus.carleton.ca  Wed Oct 30 07:58:44 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Wed, 30 Oct 2002 07:58:44 -0500
Subject: [openbox] Patch to fix Epist's buggy 2D workspace
In-Reply-To: <20021030101217.GD2892@dan0032.urh.uiuc.edu>
References: <20021030101217.GD2892@dan0032.urh.uiuc.edu>
Message-ID: <20021030125843.GK23295@marilyn>

* Ava Arachne Jarvis (ajar at katanalynx.dyndns.org) wrote:
> The patch is attached.  This fixes three bugs:
> 
>   - nextWorkspaceRow went up instead of down, and prevWorkspaceRow went
>     down instead of up. (simple change to parser.cc)
> 
>   - changeWorkspaceHorz had some issues with the "left-wards" case, 
>     caused by calling changeWorkspace once in a special case, then 
>     once more at the end of the function.  Oops.  (action.cc)
> 
>   - changeWorkspaceVert is conceptually a call to cycleWorkspace with 
>     appropriate parameters.  cycleWorkspace seemed to take care of 
>     wrapping as well, so I just changed changeWorkspaceVert to call 
>     cycleWorkspace....  
> 
> Patched against the latest CVS.
> 

The behaviour you are looking for is possible using the numeric
argument to the cycleWorkspace action. The behaviour that currently
exists is to have some sane cycling when using a non-square workspace
grid. ie, when on workspace 3 of a grid with 5 workspaces and 3 in the
first row going to the next row would put you on workspace 5.
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021030/13910a38/attachment.pgp>

From neuhauser at bellavista.cz  Wed Oct 30 07:59:56 2002
From: neuhauser at bellavista.cz (Roman Neuhauser)
Date: Wed, 30 Oct 2002 13:59:56 +0100
Subject: [openbox] Patch to fix Epist's buggy 2D workspace
In-Reply-To: <20021030125843.GK23295@marilyn>
References: <20021030101217.GD2892@dan0032.urh.uiuc.edu> <20021030125843.GK23295@marilyn>
Message-ID: <20021030125956.GN18364@freepuppy.bellavista.cz>

# smoynes at nexus.carleton.ca / 2002-10-30 07:58:44 -0500:
> * Ava Arachne Jarvis (ajar at katanalynx.dyndns.org) wrote:
> > The patch is attached.  This fixes three bugs:
> > 
> >   - nextWorkspaceRow went up instead of down, and prevWorkspaceRow went
> >     down instead of up. (simple change to parser.cc)
> > 
> >   - changeWorkspaceHorz had some issues with the "left-wards" case, 
> >     caused by calling changeWorkspace once in a special case, then 
> >     once more at the end of the function.  Oops.  (action.cc)
> > 
> >   - changeWorkspaceVert is conceptually a call to cycleWorkspace with 
> >     appropriate parameters.  cycleWorkspace seemed to take care of 
> >     wrapping as well, so I just changed changeWorkspaceVert to call 
> >     cycleWorkspace....  
> > 
> > Patched against the latest CVS.
> 
> The behaviour you are looking for is possible using the numeric
> argument to the cycleWorkspace action. The behaviour that currently
> exists is to have some sane cycling when using a non-square workspace
> grid. ie, when on workspace 3 of a grid with 5 workspaces and 3 in the
> first row going to the next row would put you on workspace 5.

    umm, who in their right mind would have workspaces laid out like
    that? what does one achieve by punching holes into the ws grid?

-- 
If you cc me or take the list(s) out completely I'll most likely ignore
your message.      see http://www.eyrie.org./~eagle/faqs/questions.html


From smoynes at nexus.carleton.ca  Wed Oct 30 08:55:19 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Wed, 30 Oct 2002 08:55:19 -0500
Subject: [openbox] Patch to fix Epist's buggy 2D workspace
In-Reply-To: <20021030125956.GN18364@freepuppy.bellavista.cz>
References: <20021030101217.GD2892@dan0032.urh.uiuc.edu> <20021030125843.GK23295@marilyn> <20021030125956.GN18364@freepuppy.bellavista.cz>
Message-ID: <20021030135519.GA15652@marilyn>

* Roman Neuhauser (neuhauser at bellavista.cz) wrote:
> # smoynes at nexus.carleton.ca / 2002-10-30 07:58:44 -0500:
>     umm, who in their right mind would have workspaces laid out like
>     that? what does one achieve by punching holes into the ws grid?

I think it makes more sense then wrapping around, otherwise it isn't
really a "next row" action. It becomes a "next row unless there isn't
enough workspaces, then 2 rows please." However, both behaviours are
possible, so I don't see it as a big deal.

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021030/da8125e0/attachment.pgp>

From john at hl.lu.se  Wed Oct 30 10:04:46 2002
From: john at hl.lu.se (John Eckerdal)
Date: Wed, 30 Oct 2002 16:04:46 +0100 (CET)
Subject: Xft-2 configure.in patch
Message-ID: <Pine.LNX.4.44.0210301557210.813-200000@maestro.hl.lu.se>


This patch adds support for Xft-2 during configure time. Works for me(tm).
Note that I can not test if this breaks compilation where earlier versions
of Xft is used. I do not fully understand how configure is supposed to 
work internally so I may very well have missed something.

It is against the current cvs.

I did not write this stuff. This is a bad merge of what I found in the
waima configure.in - I do not use waima, never had. It just popped up
on freshmeat when searching for Xft (I was looking for the Xft homepage).

Note that pkg-config is used to determine where Xft is placed. I think
that pkg-config is mandatory for the new Xft but I am not sure. If I
read the stuff correct it should still check for older/other versions 
using another way.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e

-------------- next part --------------
? autom4te-2.53.cache
? yourchanges.patch
Index: configure.in
===================================================================
RCS file: /cvs/cvsroot/openbox/configure.in,v
retrieving revision 1.38
diff -p -u -r1.38 configure.in
--- configure.in	2002/10/19 07:20:36	1.38
+++ configure.in	2002/10/30 15:00:12
@@ -143,30 +143,72 @@ AC_ARG_ENABLE(
   xft, [  --enable-xft            enable support of the Xft extension [default=yes]])
 : ${enableval="yes"}
 if test "$enableval" = "yes"; then
-  AC_MSG_RESULT([yes])
+   AC_MSG_RESULT([yes])
+   xftsucceeded=no
+        AC_PATH_PROG(PKG_CONFIG, pkg-config, no)
+   if test "x$PKG_CONFIG" != "xno" ; then
+     AC_MSG_CHECKING(for Xft)
+     if $PKG_CONFIG --exists "xft" ; then
+          AC_MSG_RESULT(yes)
+          AC_MSG_CHECKING(XFT_CFLAGS)
+          Xft_flags=`$PKG_CONFIG --cflags "xft"`
+          AC_MSG_RESULT($Xft_flags)
+          AC_MSG_CHECKING(XFT_LIBS)
+          Xft_lib=`$PKG_CONFIG --libs "xft"`
+           AC_MSG_RESULT($Xft_lib)
+           xftsucceeded=yes
+          use_xft="yes"
+          XFT="-DXFT"
+          CPPFLAGS="$CPPFLAGS $Xft_flags"
+     else
+          AC_MSG_RESULT(no)
+     fi
+   fi
 
-  AC_CHECK_LIB(Xft, XftFontOpenXlfd,
-    AC_MSG_CHECKING([for X11/Xft/Xft.h])
-    AC_TRY_LINK(
-#include <X11/Xlib.h>
-#include <X11/Xft/Xft.h>
-, XftFont foo,
+   if test $xftsucceeded = no; then
+     dnl Checks for freetype/freetype.h
+     have_freetype="false"
+     AC_PATH_PROG(FREETYPE_CONFIG, freetype-config, no)
+     if test "x$FREETYPE_CONFIG" != "xno" ; then
+       FREETYPE_CFLAGS=`$FREETYPE_CONFIG --cflags`
+       save_cppflags="$CPPFLAGS"
+       CPPFLAGS="$CPPFLAGS $FREETYPE_CFLAGS"
+     fi
+     AC_MSG_CHECKING([for freetype/freetype.h])
+     AC_TRY_COMPILE(
+#include <freetype/freetype.h>
+      ,,
       AC_MSG_RESULT([yes])
-      XFT="yes"
-      LIBS="$LIBS -lXft"
+      dnl Checks for Xft
 
-      dnl Check for utf8 support in the Xft library
-      AC_CHECK_LIB(Xft, XftDrawStringUtf8,
-        XFT_UTF8="yes",
+      AC_CHECK_LIB(Xft, XftFontOpen,
+        AC_MSG_CHECKING([for X11/Xft/Xft.h])
+        AC_TRY_COMPILE(
+#include <X11/Xlib.h>
+#include <X11/Xft/Xft.h>
+        ,,
+        AC_MSG_RESULT([yes])
+        use_xft="yes"
+        XFT="-DXFT"; Xft_lib="-lXft",
+        AC_MSG_RESULT([no])
+        AC_MSG_RESULT([** WARNING: X11/Xft/Xft.h is missing])
+       AC_MSG_RESULT([** WARNING: Disabled Xft library support])
       ),
-
+      AC_MSG_RESULT([** WARNING: Could not find XftFontOpen in -lXft])
+      AC_MSG_RESULT([** WARNING: Disabled Xft library support]))
+      ,
       AC_MSG_RESULT([no])
-    )
-  )
-
+      AC_MSG_RESULT([** WARNING: freetype/freetype.h is missing])
+      AC_MSG_RESULT([** WARNING: Disabled Xft library support])
+      CPPFLAGS="$save_cppflags")
+   fi
 else
   AC_MSG_RESULT([no])
 fi
+AC_SUBST(XFT)
+
+LIBS="$LIBS $Xft_lib"
+
 if test "$XFT" = "yes"; then
   AC_DEFINE(XFT,1,[Enable support of the Xft extension])
   if test "$XFT_UTF8" = "yes"; then

From john at hl.lu.se  Wed Oct 30 10:08:13 2002
From: john at hl.lu.se (John Eckerdal)
Date: Wed, 30 Oct 2002 16:08:13 +0100 (CET)
Subject: [openbox] Xft-2 configure.in patch
In-Reply-To: <Pine.LNX.4.44.0210301557210.813-200000@maestro.hl.lu.se>
Message-ID: <Pine.LNX.4.44.0210301605430.965-100000@maestro.hl.lu.se>


Replying to oneself is allways a bad sign, but here I go.

Addendum:

This will atleast compile xftlsfonts. Can't test if it works just right
now (not at currently at home).

This is my very first patch so there most certainly will be problems
with it (did follow the directions on the homepage though).

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From smoynes at nexus.carleton.ca  Wed Oct 30 10:18:04 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Wed, 30 Oct 2002 10:18:04 -0500
Subject: [openbox] Xft-2 configure.in patch
In-Reply-To: <Pine.LNX.4.44.0210301605430.965-100000@maestro.hl.lu.se>
References: <Pine.LNX.4.44.0210301557210.813-200000@maestro.hl.lu.se> <Pine.LNX.4.44.0210301605430.965-100000@maestro.hl.lu.se>
Message-ID: <20021030151804.GA15987@marilyn>

* John Eckerdal (john at hl.lu.se) wrote:
> 
> Replying to oneself is allways a bad sign, but here I go.
> 
> Addendum:
> 
> This will atleast compile xftlsfonts. Can't test if it works just right
> now (not at currently at home).
> 
> This is my very first patch so there most certainly will be problems
> with it (did follow the directions on the homepage though).


Thanks! I'll try this out tonight.

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021030/3812c15f/attachment.pgp>

From ajar at katanalynx.dyndns.org  Wed Oct 30 11:46:00 2002
From: ajar at katanalynx.dyndns.org (Ava Arachne Jarvis)
Date: Wed, 30 Oct 2002 10:46:00 -0600
Subject: [openbox] Patch to fix Epist's buggy 2D workspace
In-Reply-To: <20021030125843.GK23295@marilyn>
References: <20021030101217.GD2892@dan0032.urh.uiuc.edu> <20021030125843.GK23295@marilyn>
Message-ID: <20021030164600.GA22408@dan0032.urh.uiuc.edu>

[Scott Moynes - Wed, 30 Oct 2002 06:51:30 AM CST]
> The behaviour you are looking for is possible using the numeric
> argument to the cycleWorkspace action. 

You can do cycling through rows with cycleWorkspace, but I think that
unless you allow arithmetic operations in the conf file, you can not
cycle through columns (e.g., remain on the same row).


> The behaviour that currently exists is to have some sane cycling when
> using a non-square workspace grid. ie, when on workspace 3 of a grid
> with 5 workspaces and 3 in the first row going to the next row would
> put you on workspace 5.

No, the behaviour that currently exists is buggy.  

1. Looking at the patch:

   -    { "nextworkspacerow", Action::upWorkspace },
   -    { "prevworkspacerow", Action::downWorkspace },
   +    { "nextworkspacerow", Action::downWorkspace },
   +    { "prevworkspacerow", Action::upWorkspace },

  Mixed up Up and Down.  Nothing to do with cycling yet.

2. The second one, I agree, is probably not the same as cycleWorkspace 
   with a dud last row.

3. The third one, however, is a very obvious bugfix for the case where 
   one is cycling through columns leftwards on a full row:

      if (num < 0) {
        if (active_desktop % width != 0)
   -      changeWorkspace(active_desktop - 1);   <-- called correctly
   +      wnum = active_desktop - 1;             <-- should have been this
        else {
          wnum = active_desktop + width - 1;
          if (wnum >= num_desktops)
   @@ -761,6 +743,7 @@
        else
          wnum = active_desktop - width + 1;
      }
   +
      changeWorkspace(wnum);                 <-- called 2nd time with unitialized 
                                                 value in original version
    }



-- 
| Swap read error.  You lose your mind.


From smoynes at nexus.carleton.ca  Wed Oct 30 12:07:00 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Wed, 30 Oct 2002 12:07:00 -0500
Subject: [openbox] Patch to fix Epist's buggy 2D workspace
In-Reply-To: <20021030164600.GA22408@dan0032.urh.uiuc.edu>
References: <20021030101217.GD2892@dan0032.urh.uiuc.edu> <20021030125843.GK23295@marilyn> <20021030164600.GA22408@dan0032.urh.uiuc.edu>
Message-ID: <20021030170700.GA16396@marilyn>

* Ava Arachne Jarvis (ajar at katanalynx.dyndns.org) wrote:
> You can do cycling through rows with cycleWorkspace, but I think that
> unless you allow arithmetic operations in the conf file, you can not
> cycle through columns (e.g., remain on the same row).

How is the cycle column not the behaviour you expect, other than the
bug fix below?

> 1. Looking at the patch:
> 
>    -    { "nextworkspacerow", Action::upWorkspace },
>    -    { "prevworkspacerow", Action::downWorkspace },
>    +    { "nextworkspacerow", Action::downWorkspace },
>    +    { "prevworkspacerow", Action::upWorkspace },

This appears to be purely semantic. The interpretation here is based
on the numbering of the workspaces, and I think matches the layout of
bbpager. However, this brings up an interesting aside: the grid stuff
has to support the recent NETWM revision that specifies this sort of
behaviour. What pager are you using? I'm not clear why this isn't
expected.

> 3. The third one, however, is a very obvious bugfix for the case where 
>    one is cycling through columns leftwards on a full row:
> 
>       if (num < 0) {
>         if (active_desktop % width != 0)
>    -      changeWorkspace(active_desktop - 1);   <-- called correctly
>    +      wnum = active_desktop - 1;             <-- should have been this
>         else {
>           wnum = active_desktop + width - 1;
>           if (wnum >= num_desktops)
>    @@ -761,6 +743,7 @@
>         else
>           wnum = active_desktop - width + 1;
>       }
>    +
>       changeWorkspace(wnum);                 <-- called 2nd time with unitialized 
>                                                  value in original version
>     }

I'm sorry, I should have reserved comment on this until I could get
home and test it properly. Also, until I wasn't so drunk. That's
pretty obvious.

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021030/a681bc44/attachment.pgp>

From ajar at katanalynx.dyndns.org  Wed Oct 30 12:06:37 2002
From: ajar at katanalynx.dyndns.org (Ava Arachne Jarvis)
Date: Wed, 30 Oct 2002 11:06:37 -0600
Subject: [openbox] Patch to fix Epist's buggy 2D workspace
In-Reply-To: <20021030170700.GA16396@marilyn>
References: <20021030101217.GD2892@dan0032.urh.uiuc.edu> <20021030125843.GK23295@marilyn> <20021030164600.GA22408@dan0032.urh.uiuc.edu> <20021030170700.GA16396@marilyn>
Message-ID: <20021030170637.GA22605@dan0032.urh.uiuc.edu>

[Scott Moynes - Wed, 30 Oct 2002 10:59:45 AM CST]
> How is the cycle column not the behaviour you expect, other than the
> bug fix below?

See below.  :)  I didn't expect to keep jumping to workspace 0 is all.
I should have been more specific.


> >    -    { "nextworkspacerow", Action::upWorkspace },
> >    -    { "prevworkspacerow", Action::downWorkspace },
> >    +    { "nextworkspacerow", Action::downWorkspace },
> >    +    { "prevworkspacerow", Action::upWorkspace },
> 
> This appears to be purely semantic. The interpretation here is based
> on the numbering of the workspaces, and I think matches the layout of
> bbpager. However, this brings up an interesting aside: the grid stuff
> has to support the recent NETWM revision that specifies this sort of
> behaviour. What pager are you using? I'm not clear why this isn't
> expected.

I'm using bbpager, actually.

I expect the gridding to go like something this (workspace numbers for 
a 3-grid workspace):

    0 1 2   
    3 4 5 
    6 7 8

and to me, previousRow should go down in number and nextRow should go up
in number.


> I'm sorry, I should have reserved comment on this until I could get
> home and test it properly. Also, until I wasn't so drunk. That's
> pretty obvious.

It's cool.  


-- 
| Remember, UNIX spelled backwards is XINU.
| 		-- Mt.


From ajar at katanalynx.dyndns.org  Wed Oct 30 12:08:29 2002
From: ajar at katanalynx.dyndns.org (Ava Arachne Jarvis)
Date: Wed, 30 Oct 2002 11:08:29 -0600
Subject: [openbox] Patch to fix Epist's buggy 2D workspace
In-Reply-To: <20021030170637.GA22605@dan0032.urh.uiuc.edu>
References: <20021030101217.GD2892@dan0032.urh.uiuc.edu> <20021030125843.GK23295@marilyn> <20021030164600.GA22408@dan0032.urh.uiuc.edu> <20021030170700.GA16396@marilyn> <20021030170637.GA22605@dan0032.urh.uiuc.edu>
Message-ID: <20021030170829.GB22605@dan0032.urh.uiuc.edu>

[Ava Arachne Jarvis - Wed, 30 Oct 2002 11:06:38 AM CST]
> I expect the gridding to go like something this (workspace numbers for 
> a 3-grid workspace):
> 
>     0 1 2   
>     3 4 5 
>     6 7 8
> 
> and to me, previousRow should go down in number and nextRow should go up
> in number.

I should have said "previous Row should decrease in number and nextRow 
should increase in number".... 

(up!  down!  sideways!  sidereal!  widdershins!)

-- 
 Generated by Signify v1.07.  For this and more, visit http://www.debian.org/.


From john at hl.lu.se  Thu Oct 31 11:42:29 2002
From: john at hl.lu.se (John Eckerdal)
Date: Thu, 31 Oct 2002 17:42:29 +0100 (CET)
Subject: Openbox bootstap (and configuration fails)
Message-ID: <Pine.LNX.4.44.0210311737520.9555-100000@maestro.hl.lu.se>


Todays (as of now) CVS fails to bootstrap (and configure).
These are the versions of the tools I currently have installed.

automake-1.6.3-1
autoconf-2.53-8
libtool-1.4.2-12.2
libtool-libs-1.4.2-12.2

And below is what happens. Is there anything I can do to make it
bootstrap again? 

---
./bootstrap
gives:

+ aclocal
aclocal: configure.ac: 5: macro `AM_PATH_CHECK' not found in library
+ autoheader
autoheader: error: AC_CONFIG_HEADERS not found in configure.ac
+ autoconf
configure.ac:2: error: possibly undefined macro: AM_INIT_AUTOMAKE
configure.ac:5: error: possibly undefined macro: AM_PATH_CHECK
configure.ac:47: error: possibly undefined macro: AM_MAINTAINER_MODE
configure.ac:167: error: possibly undefined macro: AM_CONFIG_HEADER
+ automake --foreign --include-deps --add-missing
configure.ac:2: no proper implementation of AM_INIT_AUTOMAKE was found,
configure.ac:2: probably because aclocal.m4 is missing...
configure.ac:2: You should run aclocal to create this file, then
configure.ac:2: run automake again.
configure.ac:167: required file `./config.h.in' not found
src/Makefile.am: installing `./depcomp'
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
automake: util/epist/Makefile.am: C source seen but `CC' not defined in `configure.ac'
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
automake: util/epist/Makefile.am: C source seen but `CC' not defined in `configure.ac'
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
/usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL

You are now ready to run ./configure
enjoy!

---
And then
./configure
gives:

./configure: line 1194: syntax error near unexpected token `openbox,'
./configure: line 1194: `AM_INIT_AUTOMAKE(openbox, 2.90.0, no-define)'


-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From f12.hometower at attbi.com  Thu Oct 31 11:39:22 2002
From: f12.hometower at attbi.com (Paul J)
Date: 31 Oct 2002 11:39:22 -0500
Subject: My theme
In-Reply-To: <Pine.LNX.4.44.0210311737520.9555-100000@maestro.hl.lu.se>
References: <Pine.LNX.4.44.0210311737520.9555-100000@maestro.hl.lu.se>
Message-ID: <1036082362.7333.4.camel@Tower>

Hope you enjoy it.  

-- 
    ____/   ___
   |_   \__'  _\      (Paul J. - f12.hometower at attbi.com)
   `-(*)----(*)'	 My PGP PublicKey - 0x990CE306
-------------- next part --------------
! myst for blackbox.

! Miscellaneous settings...
style.name:	Myst
style.author:	Paul
style.date:	Fri, Oct. 25 2002
style.comments:	Enjoy please. :)		
		 

rootCommand:	bsetroot -mod 4 4 -bg rgb:10/18/20 -fg rgb:30/38/40

! ***** toolbar *****
toolbar:				raised gradient vertical
     	toolbar.color:			#8B99AA
     	toolbar.colorTo:			#424951

toolbar.label:				parentrelative
     	toolbar.label.textColor:		#000000

toolbar.windowLabel:            	flatborder gradient vertical
     	toolbar.windowLabel.color:		#8B99AA
     	toolbar.windowLabel.colorTo:	#000000
     	toolbar.windowLabel.textColor:	#efefef
	toolbar.windowLabel.borderColor:	#000000

toolbar.clock:				parentrelative
     	toolbar.clock.textColor:		#000000

toolbar.button:				parentrelative
   	 toolbar.button.picColor:		#cccccc

toolbar.button.pressed:			flat gradient vertical
    	 toolbar.button.pressed.color:	#424951
    	 toolbar.button.pressed.colorTo:	#42495

toolbar.button:	flatborder gradient vertical
     	toolbar.button.color:	#8B99AA
	toolbar.button.colorTo: #000000
	toolbar.button.borderColor: #424951

toolbar.justify:	center

! Menu settings...
menu.frame:	Flat Gradient Diagonal
menu.frame.color:	#cecece
menu.frame.colorTo:	#8d8d8d
menu.frame.textColor:	#000000
menu.frame.justify:	Right

menu.title:	Raised Bevel1 Gradient vertical
menu.title.color:	#8B99AA
menu.title.colorTo:	#1F2125
menu.title.textColor:	#c0c0c0
menu.title.justify:	Center

menu.hilite:	Raised Bevel1 Gradient Diagonal
menu.hilite.color:	#8B99AA
menu.hilite.colorTo:	#000000
menu.hilite.textColor:	#c0c0c0

menu.bullet:	Diamond
menu.bullet.position:	Right
menu.frame.disableColor:	#6d6d6d

! Window settings...
window.button.focus:	parentrelative
window.button.focus.color:	#ffffff
window.button.focus.colorTo:	#ffffff
window.button.focus.picColor:	#000000

window.button.unfocus:	parentrelative
window.button.unfocus.color:	#ffffff
window.button.unfocus.colorTo:	#ffffff
window.button.unfocus.picColor:	#000000

window.grip.focus:	Flat Gradient Vertical
window.grip.focus.color:	#8B99AA
window.grip.focus.colorTo:	#6d6d6d

window.grip.unfocus:	Flat Gradient Vertical
window.grip.unfocus.color:	#8B99AA
window.grip.unfocus.colorTo:	#404040

window.handle.focus:	Sunken Bevel1 Gradient Vertical
window.handle.focus.color:	#6d6d6d
window.handle.focus.colorTo:	#8B99AA

window.handle.unfocus:	Sunken Bevel1 Gradient Vertical
window.handle.unfocus.color:	#000000
window.handle.unfocus.colorTo:	#202020

window.label.focus:	parentrelative
window.label.focus.color:	#ffffff
window.label.focus.colorTo:	#ffffff
window.label.focus.textColor:	#000000
window.label.focus.justify:	Center

window.justify:	Center
window.label.unfocus:	parentrelative
window.label.unfocus.color:	#ffffff
window.label.unfocus.colorTo:	#ffffff
window.label.unfocus.textColor:	#000000

window.title.focus:	Flat Gradient Vertical
window.title.focus.color:	#8B99AA
window.title.focus.colorTo:	 #41474F

window.title.unfocus:	Flat Gradient Vertical
window.title.unfocus.color:	#898989
window.title.unfocus.colorTo:	#41474F

window.button.pressed:	Flat Gradient Vertical
window.button.pressed.color:	#2F3339
window.button.pressed.colorTo:	#8B99AA

window.frame.focusColor:	#ffffff
window.frame.unfocusColor:	#ffffff
handleWidth:	4
frameWidth:	0
bevelWidth:	1
borderWidth:	1
borderColor:	#000000

toolbar*xft.font: verdana
toolbar*xft.size: 8

window*xft.font: -schumacher-clean-medium-r-normal-*-*-120-*-*-c-*-iso646.1991-irv
window*xft.size: 20

menu*xft.font:		verdana
menu*xft.size:		8

-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: This is a digitally signed message part
URL: <http://icculus.org/pipermail/openbox/attachments/20021031/cb48d77b/attachment.pgp>

From David.Wooldridge at waymaker.co.uk  Thu Oct 31 12:03:42 2002
From: David.Wooldridge at waymaker.co.uk (David Wooldridge)
Date: Thu, 31 Oct 2002 17:03:42 -0000
Subject: Website screenshots
Message-ID: <004b01c280ff$7794ee20$097816ac@CHEWS002>

Just incase you want any more screenshots for the Openbox website. Keep up
the good work people :)

David
-------------- next part --------------
A non-text attachment was scrubbed...
Name: openbox4.png
Type: image/png
Size: 960520 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021031/46616d50/attachment.png>

From ben at orodu.net  Thu Oct 31 12:16:53 2002
From: ben at orodu.net (Ben Jansens)
Date: Thu, 31 Oct 2002 11:16:53 -0600
Subject: [openbox] Openbox bootstap (and configuration fails)
In-Reply-To: <Pine.LNX.4.44.0210311737520.9555-100000@maestro.hl.lu.se>
References: <Pine.LNX.4.44.0210311737520.9555-100000@maestro.hl.lu.se>
Message-ID: <20021031171653.GE19731@orodu.net>

On Thu, Oct 31, 2002 at 05:42:29PM +0100, John Eckerdal wrote:
> 
> Todays (as of now) CVS fails to bootstrap (and configure).
> These are the versions of the tools I currently have installed.
> 
> automake-1.6.3-1
> autoconf-2.53-8
> libtool-1.4.2-12.2
> libtool-libs-1.4.2-12.2

aclocal (GNU automake) 1.6.3
automake (GNU automake) 1.6.3
autoconf (GNU Autoconf) 2.54
ltmain.sh (GNU libtool) 1.4.3 (1.922.2.111 2002/10/23 02:54:36)

And bootstrap works here. If you really wanna, try get a fresh
checkout from cvs, or rm'ing config.*

I'm not seeing any errors at all..

CVS is going to not be worth running for some time anyways. I'd
recommend not bothering, and sticking to releases at least for now,
heh. Lots of construction work :)

> And below is what happens. Is there anything I can do to make it
> bootstrap again? 
> 
> ./bootstrap
> gives:
> 
> + aclocal
> aclocal: configure.ac: 5: macro `AM_PATH_CHECK' not found in library
> + autoheader
> autoheader: error: AC_CONFIG_HEADERS not found in configure.ac
> + autoconf
> configure.ac:2: error: possibly undefined macro: AM_INIT_AUTOMAKE
> configure.ac:5: error: possibly undefined macro: AM_PATH_CHECK
> configure.ac:47: error: possibly undefined macro: AM_MAINTAINER_MODE
> configure.ac:167: error: possibly undefined macro: AM_CONFIG_HEADER
> + automake --foreign --include-deps --add-missing
> configure.ac:2: no proper implementation of AM_INIT_AUTOMAKE was found,
> configure.ac:2: probably because aclocal.m4 is missing...
> configure.ac:2: You should run aclocal to create this file, then
> configure.ac:2: run automake again.
> configure.ac:167: required file `./config.h.in' not found
> src/Makefile.am: installing `./depcomp'
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
> automake: util/epist/Makefile.am: C source seen but `CC' not defined in `configure.ac'
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
> automake: util/epist/Makefile.am: C source seen but `CC' not defined in `configure.ac'
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/depend2.am: AMDEP does not appear in AM_CONDITIONAL
> /usr/share/automake-1.6/am/lang-compile.am: AMDEP does not appear in AM_CONDITIONAL
> 
> You are now ready to run ./configure
> enjoy!
> 
> And then
> ./configure
> gives:
> 
> ./configure: line 1194: syntax error near unexpected token `openbox,'
> ./configure: line 1194: `AM_INIT_AUTOMAKE(openbox, 2.90.0, no-define)'
> 
> 

xOr
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021031/47187ae0/attachment.pgp>

From smoynes at nexus.carleton.ca  Thu Oct 31 12:38:05 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Thu, 31 Oct 2002 12:38:05 -0500
Subject: [openbox] Openbox bootstap (and configuration fails)
In-Reply-To: <20021031171653.GE19731@orodu.net>
References: <Pine.LNX.4.44.0210311737520.9555-100000@maestro.hl.lu.se> <20021031171653.GE19731@orodu.net>
Message-ID: <20021031173805.GB22646@marilyn>

* Ben Jansens (ben at orodu.net) wrote:
> CVS is going to not be worth running for some time anyways. I'd
> recommend not bothering, and sticking to releases at least for now,
> heh. Lots of construction work :)

FWIW, I have a few things on my machine that I am trying to finish
up. If anyone is interested, I'll make patches against the latest
release.
Off the top of my head, I have:
- I have some fixes for XFT font redrawing that corrupts the menu.
- Some epist fixes reported here earlier. (If there are any math goons
that can make the grid wrapping work with an uneven number of
workspaces, I'd love you long time)
- Some more shadow styling stuff.
- A fix to make the menuFile: in the rc actually be used.

That's all I can think of now. Most of them are in various states of
incompleteness but if people show an interest, I'll try to polish them
off.

-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021031/a58e218d/attachment.pgp>

From john at hl.lu.se  Thu Oct 31 14:33:34 2002
From: john at hl.lu.se (John Eckerdal)
Date: Thu, 31 Oct 2002 20:33:34 +0100 (CET)
Subject: [openbox] Openbox bootstap (and configuration fails)
In-Reply-To: <20021031171653.GE19731@orodu.net>
Message-ID: <Pine.LNX.4.44.0210312032240.10499-100000@maestro.hl.lu.se>

On Thu, 31 Oct 2002, Ben Jansens wrote:

> aclocal (GNU automake) 1.6.3
> automake (GNU automake) 1.6.3
> autoconf (GNU Autoconf) 2.54
> ltmain.sh (GNU libtool) 1.4.3 (1.922.2.111 2002/10/23 02:54:36)
> 
> And bootstrap works here. If you really wanna, try get a fresh
> checkout from cvs, or rm'ing config.*

It is a fresh tree, no config.* and similar files present.
Justa cvs co openbox and then ./bootstrap

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From john at hl.lu.se  Thu Oct 31 14:35:05 2002
From: john at hl.lu.se (John Eckerdal)
Date: Thu, 31 Oct 2002 20:35:05 +0100 (CET)
Subject: [openbox] Openbox bootstap (and configuration fails)
In-Reply-To: <20021031173805.GB22646@marilyn>
Message-ID: <Pine.LNX.4.44.0210312034350.10499-100000@maestro.hl.lu.se>

On Thu, 31 Oct 2002, Scott Moynes wrote:

> * Ben Jansens (ben at orodu.net) wrote:
> > CVS is going to not be worth running for some time anyways. I'd
> > recommend not bothering, and sticking to releases at least for now,
> > heh. Lots of construction work :)
> 
> FWIW, I have a few things on my machine that I am trying to finish
> up. If anyone is interested, I'll make patches against the latest
> release.
> Off the top of my head, I have:
> - I have some fixes for XFT font redrawing that corrupts the menu.
> - Some epist fixes reported here earlier. (If there are any math goons
> that can make the grid wrapping work with an uneven number of
> workspaces, I'd love you long time)
> - Some more shadow styling stuff.
> - A fix to make the menuFile: in the rc actually be used.
> 
> That's all I can think of now. Most of them are in various states of
> incompleteness but if people show an interest, I'll try to polish them
> off.

Could be interresting to test since drop shadows do not work at all here.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From ben at orodu.net  Thu Oct 31 14:46:27 2002
From: ben at orodu.net (Ben Jansens)
Date: Thu, 31 Oct 2002 13:46:27 -0600
Subject: [openbox] Openbox bootstap (and configuration fails)
In-Reply-To: <Pine.LNX.4.44.0210312032240.10499-100000@maestro.hl.lu.se>
References: <20021031171653.GE19731@orodu.net> <Pine.LNX.4.44.0210312032240.10499-100000@maestro.hl.lu.se>
Message-ID: <20021031194627.GH19731@orodu.net>

On Thu, Oct 31, 2002 at 08:33:34PM +0100, John Eckerdal wrote:
> On Thu, 31 Oct 2002, Ben Jansens wrote:
> 
> > aclocal (GNU automake) 1.6.3
> > automake (GNU automake) 1.6.3
> > autoconf (GNU Autoconf) 2.54
> > ltmain.sh (GNU libtool) 1.4.3 (1.922.2.111 2002/10/23 02:54:36)
> > 
> > And bootstrap works here. If you really wanna, try get a fresh
> > checkout from cvs, or rm'ing config.*
> 
> It is a fresh tree, no config.* and similar files present.
> Justa cvs co openbox and then ./bootstrap

strange...

% cvs -dxor at cvs.icculus.org:/cvs/cvsroot co openbox
% cd openbox
% bootstrap
+ aclocal
+ autoheader
autoheader: `config.h.in' is created
+ autoconf
+ automake --foreign --include-deps --add-missing
src/Makefile.am: installing `./depcomp'

You are now ready to run ./configure
enjoy!
 

I'd say it seems something is wrong on your end, and I don't know
what. :\

Ben
-- 
I am damn unsatisfied to be killed in this way.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 232 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021031/917d8b3b/attachment.pgp>

From vR at movingparts.net  Thu Oct 31 14:50:57 2002
From: vR at movingparts.net (Jason 'vanRijn' Kasper)
Date: 31 Oct 2002 14:50:57 -0500
Subject: [openbox] Openbox bootstap (and configuration fails)
In-Reply-To: <20021031194627.GH19731@orodu.net>
References: <20021031171653.GE19731@orodu.net>
	<Pine.LNX.4.44.0210312032240.10499-100000@maestro.hl.lu.se> 
	<20021031194627.GH19731@orodu.net>
Message-ID: <1036093857.19490.104.camel@tobias>

On Thu, 2002-10-31 at 14:46, Ben Jansens wrote:
> On Thu, Oct 31, 2002 at 08:33:34PM +0100, John Eckerdal wrote:
> > On Thu, 31 Oct 2002, Ben Jansens wrote:
> > 
> > > aclocal (GNU automake) 1.6.3
> > > automake (GNU automake) 1.6.3
> > > autoconf (GNU Autoconf) 2.54
> > > ltmain.sh (GNU libtool) 1.4.3 (1.922.2.111 2002/10/23 02:54:36)
> > > 
> > > And bootstrap works here. If you really wanna, try get a fresh
> > > checkout from cvs, or rm'ing config.*
> > 
> > It is a fresh tree, no config.* and similar files present.
> > Justa cvs co openbox and then ./bootstrap
> 
> strange...
> 
> % cvs -dxor at cvs.icculus.org:/cvs/cvsroot co openbox
> % cd openbox
> % bootstrap
> + aclocal
> + autoheader
> autoheader: `config.h.in' is created
> + autoconf
> + automake --foreign --include-deps --add-missing
> src/Makefile.am: installing `./depcomp'
> 
> You are now ready to run ./configure
> enjoy!
>  
> 
> I'd say it seems something is wrong on your end, and I don't know
> what. :\

How about the versions of the auto* tools he's running?

> 
> Ben
> -- 
> I am damn unsatisfied to be killed in this way.
-- 

,-----------------------------------------------------------------//
| Jason 'vanRijn' Kasper ::  Numbers 6:22-26 
 `
 | All brontosauruses are thin at one end, much MUCH thicker 
 | in the middle, and then thin again at the far end.  That is 
 | the theory that I have and which is mine, and what it is too.  
 ,
| bash$ :(){ :|:&};:
`----------------------//


From smoynes at nexus.carleton.ca  Thu Oct 31 15:01:32 2002
From: smoynes at nexus.carleton.ca (Scott Moynes)
Date: Thu, 31 Oct 2002 15:01:32 -0500
Subject: [openbox] Openbox bootstap (and configuration fails)
In-Reply-To: <Pine.LNX.4.44.0210312034350.10499-100000@maestro.hl.lu.se>
References: <20021031173805.GB22646@marilyn> <Pine.LNX.4.44.0210312034350.10499-100000@maestro.hl.lu.se>
Message-ID: <20021031200132.GA23263@marilyn>

* John Eckerdal (john at hl.lu.se) wrote:
> On Thu, 31 Oct 2002, Scott Moynes wrote:
> > FWIW, I have a few things on my machine that I am trying to finish
> > up. If anyone is interested, I'll make patches against the latest
> > release.
> > Off the top of my head, I have:
> > - I have some fixes for XFT font redrawing that corrupts the menu.
> > - Some epist fixes reported here earlier. (If there are any math goons
> > that can make the grid wrapping work with an uneven number of
> > workspaces, I'd love you long time)
> > - Some more shadow styling stuff.
> > - A fix to make the menuFile: in the rc actually be used.
> > 
> > That's all I can think of now. Most of them are in various states of
> > incompleteness but if people show an interest, I'll try to polish them
> > off.
> 
> Could be interresting to test since drop shadows do not work at all here.
> 

I'll publish what I have, however, I doubt they'll work. This code
uses xft1, and since I can't compile xft2 on my machine, I have no way
of working on that bug.
-- 
Scott Moynes http://www.icculus.org/openbox/
"Computer science is as much about computers
as astronomy is about telescopes." -- Dijkstra
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 240 bytes
Desc: not available
URL: <http://icculus.org/pipermail/openbox/attachments/20021031/01ad54fd/attachment.pgp>

From john at hl.lu.se  Thu Oct 31 14:54:07 2002
From: john at hl.lu.se (John Eckerdal)
Date: Thu, 31 Oct 2002 20:54:07 +0100 (CET)
Subject: [openbox] Openbox bootstap (and configuration fails)
In-Reply-To: <20021031194627.GH19731@orodu.net>
Message-ID: <Pine.LNX.4.44.0210312052550.10630-100000@maestro.hl.lu.se>

On Thu, 31 Oct 2002, Ben Jansens wrote:

> On Thu, Oct 31, 2002 at 08:33:34PM +0100, John Eckerdal wrote:
> > On Thu, 31 Oct 2002, Ben Jansens wrote:
> > 
> > > aclocal (GNU automake) 1.6.3
> > > automake (GNU automake) 1.6.3
> > > autoconf (GNU Autoconf) 2.54
> > > ltmain.sh (GNU libtool) 1.4.3 (1.922.2.111 2002/10/23 02:54:36)
> > > 
> > > And bootstrap works here. If you really wanna, try get a fresh
> > > checkout from cvs, or rm'ing config.*
> > 
> > It is a fresh tree, no config.* and similar files present.
> > Justa cvs co openbox and then ./bootstrap
> 
> strange...
> 
> % cvs -dxor at cvs.icculus.org:/cvs/cvsroot co openbox
> % cd openbox
> % bootstrap
> + aclocal
> + autoheader
> autoheader: `config.h.in' is created
> + autoconf
> + automake --foreign --include-deps --add-missing
> src/Makefile.am: installing `./depcomp'
> 
> You are now ready to run ./configure
> enjoy!
>  
> 
> I'd say it seems something is wrong on your end, and I don't know
> what. :\

My libtool version is one minor version below yours, same with
autoconf.

This is the first package that has barfed in a very long time.

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From john at hl.lu.se  Thu Oct 31 14:55:21 2002
From: john at hl.lu.se (John Eckerdal)
Date: Thu, 31 Oct 2002 20:55:21 +0100 (CET)
Subject: [openbox] Openbox bootstap (and configuration fails)
In-Reply-To: <20021031200132.GA23263@marilyn>
Message-ID: <Pine.LNX.4.44.0210312055090.10658-100000@maestro.hl.lu.se>

On Thu, 31 Oct 2002, Scott Moynes wrote:

> I'll publish what I have, however, I doubt they'll work. This code
> uses xft1, and since I can't compile xft2 on my machine, I have no way
> of working on that bug.

Sound like a challange then. ;)
 

-- 
John Eckerdal                        the Assembly Gems page
                               http://www.df.lth.se/~john_e



From tr at rebel.net.au  Thu Oct 31 21:41:09 2002
From: tr at rebel.net.au (Tim Riley)
Date: 01 Nov 2002 13:11:09 +1030
Subject: [openbox] Website screenshots
In-Reply-To: <004b01c280ff$7794ee20$097816ac@CHEWS002>
References: <004b01c280ff$7794ee20$097816ac@CHEWS002>
Message-ID: <1036118472.353.4.camel@lili>

On Fri, 2002-11-01 at 03:33, David Wooldridge wrote:
> Just incase you want any more screenshots for the Openbox website. Keep up
> the good work people :)

Dude, perhaps you should consider those in Third World Australia before
sending image attatchments of one megabyte in size.  A link to a URL on
the web would be more than adequate.

Yours in bandwidth conservation,

Tim


-- 
tim riley | tr at rebel.net.au | nolite te bastardes carborundorum



From marius at cs.pdx.edu  Thu Oct 31 23:30:54 2002
From: marius at cs.pdx.edu (Marius Nita)
Date: Thu, 31 Oct 2002 20:30:54 -0800
Subject: events
Message-ID: <20021101043054.GA8411@cs.pdx.edu>

some thoughts on event handling:

- have a OEvent base class
- have OXEvent and OSignalEvent derived from this class

we have an event queue in the wm. when an X event comes into the event_loop,
it's wrapped into a OXEvent and stuck at the back of the queue. Signal
handlers to the same thing: wrap the signal into OSignalEvent and stick it at
the back of the event queue. if we need other types of events, they'll be easy
to incorporate into the scheme of things.

this way we handle signals and events in the same manner.

there's a OEventHandler base class which all event-handling entities extend.
this is like ob3's XEventHandler, except it deals with signal events too.

actual event handling from the queue is a matter of poo->handleEvent(event),
where event is treated as an OEvent opaquely.

anyway, something to think about
marius

PS: let's discuss the toolkit, and how we plan to handle screen info and
colormaps now that basedisplay will go away. (yay)