headers
David Tweed | 2 Sep 2004 10:02
Picon
Picon
Favicon

possible ion3 WFloatWS-workspace bug

This behaviour in the last ion3 snapshot 20040703 looks like a bug 
(haven't tried a later subversion but no changelog entries appear to 
mention this problem):

on a WFloatWS open up an xv on some image. by opening up the `xv controls'
window and clicking on the an image name you can cause xv to (re)load an
image. if you do this when the main xv window has a tab, the image reloads
whilst the overall xv window stays in the same place on the screen (as
expected). if you toggle the window tab on the main xv window to off and
do a series of (re)loads, each load causes the xv window to move up the
screen by an amount that looks to be the tab bar height. The first time,
this might be some sort of adjustment of the geometry to take in to
account the absence of a tab and not worth bothering about but the fact it
happens on each loading seems odd. I don't think this is an xv bug (I've
seen something similar with a different window manager and the wm author
told me it was due to a bug in the gravity handling (?) in the wm, which
he then fixed.)

Anyway, hope the bug report helps,

__cheers, dave____________________________________________
www.inf.ed.ac.uk/people/staff/David_Tweed.html
tel: +44 131 651 3447  fax: +44 131 651 3435
X wrote a book about this, which Y was carrying around for
a long time with little discernible effect -- John Baez
Permalink | Reply | 1 comment |
headers
Tuomo Valkonen | 2 Sep 2004 12:01
Picon
Picon

Re: possible ion3 WFloatWS-workspace bug

On Thu, Sep 02, 2004 at 09:02:41AM +0100, David Tweed wrote:
> I don't think this is an xv bug 

No, it is. I looked at the xv code, and it uses window manager frame
offsets from the time when the window was last reparented. (Applications
are supposed to ask for geometry for the WM frame instead of their own
window.) If you attach the image window to a frame that already has tabs
removed, the frame will not be moved when you load another image (or
resize, or whatever).

--

-- 
Tuomo
Permalink | Reply | 0 comments |
headers
Joe Skinner | 6 Sep 2004 08:03
Picon

compiling from svn seems broken under debian.

hi I have been trying for a while to compile a recent svn checkout under 
debian unstable.

after running configure with the instructions from the readme I get 
errors that look like variables that haven't been defined by the 
system.

ie  <at> DL_LIBS <at> 

is this a known problem or is the use of configure gong to be dropped ?

Joe.

--

-- 

Joseph Skinner               email: crypt <at> ihug.co.nz       

There's no such thing as a wizard
who minds his own business
 - Berengis the Black  Court Mage to the Earls Caeline
Permalink | Reply | 2 comments |
headers
Tuomo Valkonen | 6 Sep 2004 20:37
Picon
Picon

ion-3ds-20040906


This release finally includes a usable (yet still incomplete) version
of the mod_panews workspace module (that has also bored the name 'autows'
and 'rubberws' previously). The final outcome is not exactly what I
initially planned, as those plans turned out not to be that workable.
Instead, what mod_panews does is add to the basic tiled ionws approach
overlappable splits and automatically filled panes that are initially
empty. Each window is classified and assigned to the pane matching that
classification. (The default classifications are T(erminal), V(iewer)
and M(iscellaneous).)

For a better feel for it, try it out yourself and please give feedback.
The overlapping panes may be a bit confusing (this is one place where
true translucency might actually be useful and not just eyecandy) at
first, but you'll get used to them if you use the feature, and e.g.
the Gimp works quite splendidly with the toolboxes in the 'M' pane and
image windows in the 'V' pane --  although the simple initial size-based
classification heuristics don't always get it right and overriding
winprops (setting: panews_classification) should eventually added. 

In addition to the new module, this release adds support for translations
of program messages and the manual page, regardless of whether such is
of any use in a program like Ion or not. (Currently Finnish and Czech
translations are available.) Of course there are some bug fixes and
other minor additions as well, and the ./configure script is also back,
the abandonment of libtool being final now. 

New ChangeLog entries:

2004-09-06:
(Continue reading)

Permalink | Reply | 5 comments |
headers
Tom Payne | 6 Sep 2004 20:44

Re: compiling from svn seems broken under debian.

On Mon, Sep 06, 2004 at 06:03:08PM +1200, Joe Skinner wrote:
> after running configure with the instructions from the readme I get 
> errors that look like variables that haven't been defined by the 
> system.
> 
> ie  <at> DL_LIBS <at> 

This should have been fixed about a week ago. Do:
	svn up
	autoconf
	./configure
and report back if it still doesn't work. Plesae post the output of
./configure when you do so.

--

-- 
Tom
Permalink | Reply | 1 comment |
headers
Joe Skinner | 7 Sep 2004 05:57
Picon

Re: compiling from svn seems broken under debian.

On Tue, 07 Sep 2004 06:44, Tom Payne wrote:
> On Mon, Sep 06, 2004 at 06:03:08PM +1200, Joe Skinner wrote:
> > after running configure with the instructions from the readme I get
> > errors that look like variables that haven't been defined by the
> > system.
> >
> > ie  <at> DL_LIBS <at> 
>
> This should have been fixed about a week ago. Do:
>  svn up
>  autoconf
>  ./configure
> and report back if it still doesn't work. Plesae post the output of
> ./configure when you do so.

Ah that is the problem. I missed out the autoconf step.

Now working fine.

Joe.

--

-- 

Joseph Skinner               email: crypt <at> ihug.co.nz       

There's no such thing as a wizard
who minds his own business
 - Berengis the Black  Court Mage to the Earls Caeline
(Continue reading)

Permalink | Reply | 0 comments |
headers
David Tweed | 7 Sep 2004 23:55
Picon
Picon
Favicon

Re: ion-3ds-20040906

On Mon, 6 Sep 2004, Tuomo Valkonen wrote:

> 
> This release finally includes a usable (yet still incomplete) version
> of the mod_panews workspace module (that has also bored the name 'autows'
> and 'rubberws' previously). The final outcome is not exactly what I
> initially planned, as those plans turned out not to be that workable.
> Instead, what mod_panews does is add to the basic tiled ionws approach
> overlappable splits and automatically filled panes that are initially
> empty. Each window is classified and assigned to the pane matching that
> classification. (The default classifications are T(erminal), V(iewer)
> and M(iscellaneous).)

Well done for a very useful feature.

I've been messing about with it, and it's quite nice for working with the
insight debugger (the application with `satellite windows' I use most).
One thing that comes up is that there are some applications where the
satellite windows have a `correct' size, where being bigger than this
doesn't mean they are any more useful (think the various gimp tool windows
which don't show more/bigger tool icons if you make them bigger) and some
applications where, if space is available, making satellite windows bigger
improves usability.  An example of these are the windows you get with
insight where the windows contain lists, eg, stack window contains a list
of stack frames, variables window contains a list of local variables. If
these are stacked vertically in a pane and there's more space available
then expanding them equally so they fill the extra space means more is
visible in each window without needing to scroll.

As a wishlist idea, it would be nice to have some setting for a WPaneWS
(Continue reading)

Permalink | Reply | 0 comments |
headers
Ted Zlatanov | 8 Sep 2004 16:38
X-Face
Favicon
Gravatar

Re: ion-3ds-20040906

On Mon, 6 Sep 2004, tuomov <at> modeemi.cs.tut.fi wrote:

> The overlapping panes may be a bit confusing (this is one place where
> true translucency might actually be useful and not just eyecandy) at
> first

I'm sure you've seen the announcement, but the new X.org 6.8.0 release
this morning has true translucency support.  I hope Ion can use it.

Thanks for the great software
Ted
Permalink | Reply | 2 comments |
headers
Tuomo Valkonen | 8 Sep 2004 17:53
Picon
Picon

Re: ion-3ds-20040906

On Wed, Sep 08, 2004 at 10:38:04AM -0400, Ted Zlatanov wrote:
> I'm sure you've seen the announcement, but the new X.org 6.8.0 release
> this morning has true translucency support.  I hope Ion can use it.

AFAIK you need to either implement a composition manager yourself
(certainly out of the scope of Ion), or talk to another one to
actually make any use of the composition extension. Are there any
general-purpose composition managers available, or are they all
some quick&dirty hacks on popular wm:s?

--

-- 
Tuomo
Permalink | Reply | 1 comment |
headers
Jon Åslund | 8 Sep 2004 17:36
Picon
Picon
Favicon

KP_Add without numlock

for ion2 in global_bindings I entered:

 kpress("KP_Subtract", make_exec_fn("amixer -q set Master 2-")),
 kpress("KP_Add", make_exec_fn("amixer -q set Master 2+")),
 kpress("KP_Enter", make_exec_fn("amixer -q set Master toggle")),

I wanted to be able to raise/lower and mute/unmute the volume by
pressing - + and enter on the numeric keyboard, when numlock is
inactive. However I noticed that these keys are called the same with
or without numlock active (according to the program xev, which I use
to find out key names). I changed to the keys 4,5,6 (KP_Left,
KP_Begin, KP_Right) because I noticed that they are called KP_4, KP_5,
KP_6 with numlock active. However, that didn't help either. Ion steals
these keypresses regardless. Is it possible to somehow specify
bindings with or without numlock? I rarely use the numeric keyboard
for entering numbers anyway, so it's not very important, but it would
be nice.

--

-- 
Jon Åslund
Permalink | Reply | 1 comment |

Gmane