Re: FVWM: FVWM Wiki

2014-04-16 Thread Jaimos F Skriletz

On 04/16/2014 01:40 PM, Thomas Funk wrote:

Hi there!

I am wondering what's happening with the FVWM wiki on 
http://fvwmwiki.xteddy.org/ ?

I remember there was a discussion to move it into FVWM's website or something 
like that.

But now the wiki is gone and never appears anywhere :-/

Does anybody knows about that? Because it would be a pitty if these very well 
information are lost.


The information is not lost, I have a copy of it and need to add it to 
the fvwmforums.org domain but the server didn't have new enough perl for 
the ikiwiki and I wasn't able to get a clean html dump of it.


I tried at http://zensites.net/wiki/fvwmwiki-html/ as a test, but it 
didn't render the html correctly all the way


Then I got distracted from learning ikiwiki

jaimos




Re: FVWM: Forums and Wiki need new maintainers/homes

2013-09-06 Thread Jaimos F Skriletz

On 09/06/2013 12:06 PM, Bert Geens wrote:

Thomas Adam  writes:


Hi Dan,

On Thu, Sep 05, 2013 at 09:58:36PM -0400, Dan Espen wrote:

Thomas Adam  writes:


Hi all,

Given my diminishing work on FVWM, I need to think about handing over the
FVWM Forums [1] and the FVWM Wiki [2].  I cannot devote the time needed for
their maintenance, nor do I want to act as a point of contact for them
anymore.

 From my point of view, I'd have no problem with closing them down.
If you could direct users to the mailing lists, that would be a good
thing.

That'd be the simplest solution, yes.  But I'm unclear how many users of
FVWM and the forums would welcome that.  I was always in favour of closing
it down, but it exists _because_ it's what people wanetd, and still do, from
what I can tell.  They're certainly a lot busier than fvwm@ in terms of
questions.


The barrier to sending questions to mailinglists certainly is a lot
higher for most people than posting to a forum (whether this is is an
imaginary barrier or not is, in the end, irrelevant), which is the
main reason I set up the Fvwm forums back then.

If need be I will take over maintaining the forums again. The software
is relatively low maintenance (certainly compared to the horror that was
phpBB2). It would be nice if there would be some more knowledgeable
moderators, my personal use of Fvwm is, to say the least, pretty basic.

While I don't foresee a repeat of the issue that forced Thomas to take
over maintaining the forum software it would certainly be nice if
hosting could remain somewhere where more people have access to it in
case of emergency. In other words, if Jaimos doesn't mind it would be
nice if the forums could stay where they currently are.


The wiki is less important, but there's a lot of documentation which I feel
should make it in to the man pages somewhere if someone is wanting to do
that.


It would be a shame to lose the information stored in the wiki, I find
it often more to the point that having to distill that same information
form the manpage (assuming it's in there at all).


For the time being I see no reason I cannot keep hosting the forums and 
host the wiki as well. This could change in the future (but I don't 
foresee such an event) and I would prefer to host if someone I have 
known via #fvwm or in the fvwm community for a while took charge of what 
little maintenance there is on the admin side. As for getting some more 
moderators for the forums themselves that could be useful and moderators 
do not need any access to the servers.


I could help Bert with anything on the admin side of the server and 
keeping phpBB3 up and secure, but I too only use the very basic features 
of fvwm and don't want to be a moderator on the forums.


One thing that could be useful on my end to save a small issue we had 
when I changed servers was having control of the domain names. I 
consider them part of the fvwm community but if I end up keep hosting 
the forums and the wiki it just saves time and hassle if I can access 
and control them directly instead of waiting for the owner of them to 
update DNS records and the likes.


jaimos




Re: FVWM: how to? FakeClick for Middle button

2013-01-25 Thread Jaimos F Skriletz

On 01/25/2013 08:23 AM, Spoofing wrote:

sup.

I like to use "Shift + Insert" for paste text from clipboard everywhere,
and for example, when I run Firefox, want to press "Shift + Insert" for
open URL from clipboard, like the mouse middle (button 2) click.

 Key Insert A S FakeClick depth 2 press 2 wait 250 release 2

doesn't work, and also try to change depth, but nothing happens...

how to simulate mouse middle click button with Shift + Insert? :)

First, not all applications like fake (or synthetic) events, and as per 
the man page FakeClick and FakeKeyPress are for debugging fvwm and not 
going to work in all situations. So your approach should be rethought.


Second it is not firefox that 'pastes' when you click the middle mouse 
button. It is xorg that intercepts the middle mouse click and then sends 
the resulting paste to the window. So sending the middle mouse button 
click to the root window or the firefox window (or any other window) 
will note generate the paste event. Firefox does not know to paste when 
it receives a middle mouse click.


On the other hand firefox and many applications honor the ctrl-V paste 
button. In my testing I got


FakeKeyPress depth 2 modifiers 8 press v

To work just fine and correctly paste something into chromium (provided 
I copied with ctrl-C from an application first).


So that may work for you, but now you need to get a clipboard manager to 
keep the clipboard and cutbuffer in sync. It looks like something like 
'autocutsel' may do that for you. Though it could be possible that there 
are more advanced clibboard managers and you could use one of those to 
not only keep the two methods of copying/pasting in sync can probably 
set up custom key bindings to send the clibboard to the desired window. 
This is the direction I would look. It looks like clipit may have that 
functionality.


Using FakeKeyPress to make shift-insert work like ctrl-V will only work 
for applications which already honor ctrl-V so you are just changing the 
default. It will have no effect (or even other effects) because each 
application will honor the ctrl-V how ever it deems is correct. So even 
though that works in some applications it may not do what you want.


So I think getting a clipboard manager that has the feature to set up a 
keybinding would be the direction to go (note I have never messed with 
them, so I am unsure what sort of things the clipboard managers can do).


jaimos



Re: FVWM: Debian Unstable now has FVWM 2.6.5

2013-01-23 Thread Jaimos F Skriletz

On 01/23/2013 11:29 AM, E Frank Ball III wrote:

On Wed, Jan 23, 2013 at 01:53:32PM +, Thomas Adam wrote:
  >  Hi,
  >
  >  Just a quick note to all the Debian users who've long since been waiting 
for
  >  Debian to update their version of FVWM, this has finally happened.
  >
  >  I've been helping out the new maintainer and talking to the DD who's
  >  sponsering the maintainer, and we've finally got FVWM 2.6.5 released to
  >  Debian Unstable.
  >
  >  Someone may well backport this; I don't know.


Thank you, thats great.  Any chance it will make it into Wheezy?

Wheezy is frozen so most likely not.

jaimos



Re: FVWM: Smart maximize

2012-11-28 Thread Jaimos F Skriletz

On 11/28/2012 10:17 AM, Bastian wrote:

Have you tried to set the Style MinPlaceOverlap to subjected window and
then call PlaceAgain on it?

This works for me:

DestroyFunc movetofreeplace
AddToFunc movetofreeplace
+ I WindowStyle MinOverlapPlacement
+ I PlaceAgain



First please do not top post (read the rules of the mailing list).

The problem with this method is that it will find the first available 
free (or min overlap) region it can place the window. If there are 
multiple rectangles on the screen that are free of windows it will 
always pick the first one it finds (not the biggest rectangle). So 
though it was a suggestion to look at there is really no nice way I can 
see to mix it to be able to grow a window to fit in the biggest possible 
rectangle it can find.


A FvwmPerl script will end up being the the best way I can think of to 
first find the biggest free rectangle and then to move/resize the window 
to fit in that rectangle. The hard part will be the problem of finding 
the biggest open rectangle (this is a logic/math problem). Once you have 
that it is fairly straight forward to move the window to that rectangle 
and maximize grow grow (or just resize it) to the right size.


jaimos



Re: FVWM: set style by state

2012-11-07 Thread Jaimos F Skriletz

On 11/07/2012 02:07 AM, Bastian wrote:

DestroyFunc F1
AddToFunc F1
+ I All (FvwmConsole)
+ I Move w+10 w+0
+ I Move w+0 w+10
Most likely you only have a single window named FvwmConsole, so this All 
command only gets a single window and the once it has the context of 
that single window it knows which one to move. Try running this function 
if you have two FvwmConsole windows open. See what happens.




DestroyFunc PullTaggedWindows
AddToFunc PullTaggedWindows
+ I All (State $0)
+ I Move w+10 w+0
+ I Move w+0 w+10


What happens here is you are most likely matching multiple windows. You 
aren't telling fvwm to move each window but only to move one window (in 
which you used the conditional to get the correct window context). Since 
fvwm cannot determine which one window you wanted to move to is asking 
you which one to move.


Hence + I All (State $0) CustomMoveFunction

where the CustomMoveFunction does your move. That command will run that 
function on each and every window it matches (instead of just trying to 
get the window context for the rest of the function).


jaimos



Re: FVWM: set style by state

2012-11-05 Thread Jaimos F Skriletz

On 11/05/2012 09:04 AM, Bastian wrote:

On 05Nov12 08:40 -0700, Jaimos F Skriletz wrote:

So you could do something like this

AddToFunc TagWindow
+ I Pick
+ I TestRc (Error) Break
+ I State $0 1
+ I WindowStyle BorderColorSet foo, HilightBorderColorset bar

If you have a way to toggle the state of the window (so you can turn
off State 10 in your case, you will have to do something similar and
reverse the colorset change on the border color)

Thank you for this comprehensive instructions.
Using WindowStyle does the job.

Off topic: Using conditional commands (Pick, All ,...) without a command
in the same line does not work for me.



In general that is correct, you need All (conditions) Action, Next 
(conditions) Action, etc


The idea is things like Raise, Move, WindowStyle need to know what 
window you want to do it for and the conditionals find the right window. 
ThisWindow is a useful one to use once you already have the window 
chosen/context is known.


I am under the impression that Pick is slightly different in once you 
pick the window the context is known so you only have to pick the window 
once and then run multiple things on that window. That example I gave 
you comes directally from the man page for FVWM 2.6.5 so if it doesn't 
work (I haven't had time to test it) it is a bug with the man page. In 
that case you may want to do something like this instead


AddToFunc ToggleWindow
+ I Pick (Conditions) CustomStateFunction

AddToFunc CustomStateFunction
+ I State ...
+ I WindowStyle ...
+ I Move ...

and so forth. The reason why you want to do that is you don't want the 
function to pause and have you pick the window for each action in your 
function.


The other option is to use the ThisWindow conditional, so + I ThisWindow 
(conditions) Action. I will admit I'm a bit unsure as to the proper 
method, but that should be enough to play with to get it to work as 
expected.


jaimos




Re: FVWM: set style by state

2012-11-05 Thread Jaimos F Skriletz

On 11/05/2012 07:57 AM, Bastian wrote:

State is for conditional expressions.
We need more information about what you are trying to do.

More in detail:

Mouse 2 W 4 Function TagWindow 10
Mouse 2 R 4 Function PullTaggedWindows 10

DestroyFunc TagWindow
AddToFunc TagWindow
+ I Pick State $0 1


When you tag your window you can do other things to it. One useful tool 
is WindowStyle. It will change the Style of the window in question. So 
just like your TagWindow function picks a window to set to State 10, you 
can use that to set the WindowStyle to change the border color (or do 
other options) on that window.


So you could do something like this

AddToFunc TagWindow
+ I Pick
+ I TestRc (Error) Break
+ I State $0 1
+ I WindowStyle BorderColorSet foo, HilightBorderColorset bar

You can of course add other options/actions to that window you picked 
(note the second line is just to stop the function if a window wasn't 
sucessfuly picked). You might also prefer to use + I Pick (conditions) 
and put some sane conditions so you don't accidently pick a window you 
don't want to (like transient windows, pannels, etc)


If you have a way to toggle the state of the window (so you can turn off 
State 10 in your case, you will have to do something similar and reverse 
the colorset change on the border color)


jaimos



Re: FVWM: can not unlock "date time" tool

2012-10-18 Thread Jaimos F Skriletz



On 10/18/2012 09:47 AM, Roger Campbell wrote:


 Some where things went bad and I was unable to set the background
 and the "date&   time" tool any more. I can bring the tool up but I
 can not unlock the tool.


 I still can not unlock the "date and time" tool. The error in
 ~/.xsession-errors is :

 (gnome-control-center:2202): Gtk-WARNING **: Error acquiring \
 permission: Failed to acquire permission for action-id \
 org.gnome.controlcenter.datetime.configure



It appears that you are trying to use the gnome date/time tool. Not sure
if you have some policykit or issue with that but are you running a
gnome pannel in fvwm? Fvwm does not come with a date/time tool. So
support for that tool is not what this mailing list is for. So the issue
you are having is with the gnome tool you are trying to use.

My suggestion would be to use a far simpler tool such as FvwmScript
(which getts the output of date from the cli), xclock, or many other
simple clocks. You can then choose run this clock as its own window and
position it somewhere or swallow it into a pannel such as FvwmButtons.

If you want realtime up to the second xclock would be the best choice. I
use fvwm script myself but my clock can be up to 15seconds off the real
time as I don't refresh it every second.

If you want to configure xclock's background read the manpage for the
options/resources to set the colors. If you decided to use FvwmScript,
you should have been shiped with a sample script for the date-time and
you just need to configure a Colorset for it.

jaimos






Re: FVWM: $PATH in Test x

2012-05-01 Thread Jaimos F Skriletz

On 05/01/2012 12:53 PM, Dominique Michel wrote:

A very common way to write $HOME in a path is with ~

With "PATH="~/bin:$PATH" into ~/.bash_profile, I get only the first
menu line on the screen.
With "PATH="/home/dom/bin:$PATH" into ~/.bash_profile, I get the 2
menu lines on the screen.

It look like the "Test (x xdradio) ..." work only in the first case.
Is it a bug in fvwm or somewhere else, or some obscure and wanted stuff
I am not aware of?
It is an issue with the shell, and how it expands special characters. 
When the shell encounters an Env Variable or a special character like ~, 
*, !, etc it exapnds it based on the rules of the character.


So when you have PATH="~/bin/:$PATH" because of the way the shell works 
it will expand $PATH correctly but it will not mess with the character 
"~" because it is inside quotes. What you should do it type this in a shell


echo $PATH

and you will see that your PATH still contains the ~ and it wasn't 
expanded to your home directally like it should have been. What this 
does is confuse things that use the PATH variable. Some may expand the ~ 
but my guess is most won't. It is preferable to do


PATH="$HOME/bin:$PATH" to ensure everything gets expanded correctly by 
the shell.


Another issue is putting a * or a ? in there would also cause issues 
because the shell is instructed to not expand things because of being 
between quotes.


jaimos



Re: FVWM: limiting window movement to working area

2012-04-12 Thread Jaimos F Skriletz

On 04/12/2012 09:23 AM, Thomas Adam wrote:
I toyed with the idea of having wall-boundaries where windows couldn't 
be moved, but due to the way one can configure their DesktopSize, and 
flip between pages when resizing/moving windows, etc., soon came to 
the realisation it becomes impossible to give a realistic configuration.


Therefore, you won't ever get hard-walled areas honouring the EWMH 
Working Area for interactive window moves. As for your current 
problem, just use Layers, perhaps? -- Thomas Adam 


Would it be possible and useful to set an invisiable boundary that could 
be used with SnapAttraction or EdgeMoveDelay/EdgeMoveResitance over the 
EWMH boundaries. This way things can still be moved anywhere but there 
is a bit of resistance or snaping to the borders (even if there isn't a 
pannel there to snap to)? I am thinking allowing the EWMH bounderies to 
behave more like the edge of the pages when moving things around, and 
something that is configurable to the amount of delay/resistance needed?


jaimos



Re: FVWM: A small issue with SetEnv

2012-03-12 Thread Jaimos F Skriletz

On 03/12/2012 11:47 AM, Roman Grazhdan wrote:

It's about FvwmPiazza again. At cruncbanglinux forums Thomas kindly
proposed to use

ModulePath +:$./modules


There is an extra dollar sign in your ModulePath (I think that is 
causing your issues):


Also you have using ./modules this is a relative path, but this could 
cause issues depending on the cwd at the time that path is used, best to 
use full paths.


From the fvwm man page

  The ModulePath may contain environment variables such as 
$HOME (or
  ${HOME}).  Further, a '+' in the path is expanded to the 
previous
  value of the path, allowing easy appending or prepending 
to the path.


  For example:

  ModulePath ${HOME}/lib/fvwm/modules:+

So I think you want ModulePath +:${HOME}/modules/ Though I think the 
example in the manpage is better since it makes your custom path be 
first on the module list over the default (This is useful if you want to 
patch/use a custom fvwm module that already has the same name in the 
full path).




Re: FVWM: confused about rounded borders

2012-03-03 Thread Jaimos F Skriletz

On 03/03/2012 12:06 AM, Globe Trotter wrote:

Hi,

Looking at a few themes, there seems to be talk of "patched fvwms" to get rounded borders 
on windows. Where does one get these "patched fvwms"?


Archlinux seems to have the patches floating still and that is the place 
I would look. Seems they have done a bit to ensure they still apply to 
2.6.x. I would only take the rounded corners one, they have a bunch of 
them, some which may not be relevant any more. But this is of course 
unsupported, have fun.


jaimos





Re: FVWM: Touch Screens

2012-02-24 Thread Jaimos F Skriletz

On 02/24/2012 10:43 AM, Chris Siebenmann wrote:

| One important aspect of project life cycles is the ability
| how projects keep in step with general technical development.
|
| It may be the case that five or ten years from now a lot of computers
| will be equipped with touch screens.
|
| It could be a good idea to think about how FVWM would cope
| with touch screens in the future. Decisions should be made:
| Is FVWM supposed to support touch screens sometime or not?
| If touch screen support is a good idea, how are the steps to
| achieve that goal?

  My personal opinion is that we don't currently know what the right
window manager interface is for decent-sized touchscreen displays
(displays that are not going to be single application/window at a
time). Thus I feel that doing anything for touchscreens in FVWM right
now is highly premature; any work on touchscreen window managers is
likely to be highly experimental for years to come.



There was a great multi touch screen demo I saw a few years ago I can't 
seem to find but here is something similar, and I would love to be 
running fvwm on one of those. Though I do agree that for the standard 
consumer such things are a ways off, it still dosen't hurt to start 
thinking about such things now. Though I don't think it is a major 
project for GSOC at this time.


Yes I would love to be running FVWM on something like this

http://www.youtube.com/watch?v=JfFwgPuEdSk

jaimos




Re: FVWM: GSoC 2012: Project ideas

2012-02-24 Thread Jaimos F Skriletz

On 02/24/2012 07:56 AM, msib...@crosswire.com wrote:

Ok,

I give. There is no better way than FVWM.

Thanks!



That was not the point, it is just another extreme stance. FVWM is good 
at what it does, dosen't mean it is the best, as this is usually 
personal opinion. A rewrite of FVWM would only waste the time of the 
developers that could be spent on projects that actually add 
functionality to improving FVWM.


jaimos





Re: FVWM: GSoC 2012: Project ideas

2012-02-19 Thread Jaimos F Skriletz

On 02/17/2012 03:49 PM, Dan Espen wrote:

Thomas Adam  writes:


On Fri, Feb 17, 2012 at 02:10:54PM -0700, Jaimos Skriletz wrote:

2) Rewrite the menu syntax and redsign the object, there was some talk
about this on the mailing list years ago now and some good ideas for

And the link to that thread is?


http://www.mail-archive.com/fvwm-workers@lists.math.uh.edu/msg07167.html

Hmm.  This is terrible.  If only because menus are treated here as some kind
of alien-citizen and if this suggestion was to ever go through they should
be treated the same way in FVWM like any other entity.

Seems to be a lot of change with no new function.

As I understand it, Fvwm can be accepting commands from many sources at
the same time.  Various modules can send commands to Fvwm, each is
sent one at a time.  I don't believe "+" is handled as it should be
to account for this.

This new design would make that worse.

That was something I ran across years ago and saw it as thinking about a 
slightly more versatile syntax, though it is the functionality in the 
long run that is desired. I like how the idea was to give each menu item 
a more detailed list of properties, to allow for a larger variety in 
menu design. Such as being able to apply the different MenuStyles, 
Actions, etc to an individual item as opposed to the whole menu. And I 
think the current menu syntax is slightly limited in the ability to 
allow the above in a nice way.


Though a slight modification of the current syntax to be more like 
FvwmButtons (item1, item2, item3) as another option. Something like


AddToMenu MenuName (Item="xTerm",Hotkey="t",Action (Mouse 1) Exec xterm, 
Action (Mouse 3) Exec xterm -g 80x60,Colorset=,Icon=,Frame=,etc)


The change of syntax was more along the thoughts of in order to improve 
functionality an improved config syntax could go hand in hand. It was 
just a suggestion, but after thomas mentioned that adding most (if not 
all) the functionality is on the TODO list and not a large project, 
there are plenty of other ideas to work with for GSoC.


jaimos




Re: FVWM: Taking a break...

2009-09-24 Thread Jaimos F Skriletz

Thomas Adam wrote:

Hi all,

It umm, well it seems I can't quite deal with this how I'd like, so
for the benefit of the project, I think I will take a short break -- I
appreciate "sticks and stones" and all that, but when someone has
repeatedly wished you would drop dead, it is a little dis-heartening.

So I will take a break and possibly come back.

-- Thomas Adam

  
Best of luck to you Thomas. Over the years I have grown to appreciate 
and enjoy your insights and how much you have helped the FVWM community. 
I know you will find a place in which you will be truly appreciated for 
your generosity and ability. Here is wishing to see you again.


jaimos



Re: FVWM: starting up applications

2009-09-16 Thread Jaimos F Skriletz

Darren Upsolla wrote:

From: des...@verizon.net
Date: Wed, 16 Sep 2009 20:33:29 -0400

Generally you should use InitFunction to start applications:

DestroyFunc InitFunction
AddToFunc InitFunction
+ I Exec exec xsetroot -solid cyan
+ I Exec exec xterm
+ I Exec exec firefox



oh - but that post says:

"*INITFUNCTION IS CALLED ONCE -- AT INIT TIME WHEN FVWM FIRST STARTS.
STARTFUNCTION IS CALLED AT INIT AND RESTART.*

So you can see now that you only need one function for your purposes,
as you can add checks to StartFunction to tell it to do one thing if
FVWM is initialising, and another if it's not (i.e., presumably at
restart)."

maybe thats why i am confused lol! why is this not in the manpage then?

  
Yes InitFunction, StartFunction, RestartFunction were the older way of 
doing it. Now due to Test () this can all be done in a single function 
StartFunction. I think it makes things a bit easier to locate and keep 
track of. I am unsure if the InitFunction and RestartFunction will be 
depreciated in the future. The trend I see is to use only a 
StartFunction and Test ().


This is in the man page:


  You do not need to define all special functions if some are empty.
  Also note, all these special functions may be emulated now using
  StartFunction and ExitFunction, like this:

  DestroyFunc StartFunction
  AddToFunc StartFunction
  + I Test (Init) Module FvwmBanner
  + I Module FvwmPager * *
  + I Test (Restart) Beep

  DestroyFunc ExitFunction
  AddToFunc ExitFunction
  + I Test (Quit) Echo Bye-bye
  + I KillModule MyBuggyModule
  + I Test (ToRestart) Beep

jaimos



Re: FVWM: starting up applications

2009-09-16 Thread Jaimos F Skriletz

Darren Upsolla wrote:

hi


  

Date: Wed, 16 Sep 2009 17:38:21 -0300
Subject: Re: FVWM: starting up applications
From: joqu...@gmail.com
To: abovefrombe...@hotmail.com

init



sorry what does this mean?

Darren
  


The response was to use the 'initfunction' as your requested. Though it 
is better to put all startup commands into a single function. This 
function is called the StartFunction


DestroyFunc StartFunction
AddToFunc StartFunction
+ I 
+ I 
+ I Test (Init) 
+ I Test (Restart) 

StartFunction is ran after fvwm starts up and reads though your config. 
It will run every time fvwm is started. Some things like FvwmModules 
will need to be started every time while others you may only want to run 
on the initial start up or on a restart. The Test () option says to only 
run that action under particular conditions. With this you can customize 
any number of fvwm  to run at start up and separate them into 
various test conditions. i.e. Screensavers, wallpaper settings only need 
to be ran once, while FvwmModules will need to be run each time you restart.


jaimos




Re: FVWM: thomas adam ill

2009-09-13 Thread Jaimos F Skriletz

Dominik Vogt wrote:

On Sun, Sep 13, 2009 at 10:31:02AM -0400, des...@verizon.net wrote:
  

Darren Upsolla  writes:


it has been brought to me attention thomas adam is seriously ill -
just thought you should know

darren
  

Very sad news indeed.

His contributions have been much appreciated.



Yes, very sad.  My best wishes to him, I hope he gets better.

  
I talked to Thomas today, this is just a bad rumor. I'm just letting 
everyone know he said he is fine and to disregard this bad rumor before 
it goes any further.




Re: FVWM: fvwm crash when setting the mini-icon

2009-08-18 Thread Jaimos F Skriletz

Gautam Iyer wrote:

On Tue, Aug 18, 2009 at 08:20:59PM -0600, Jaimos F Skriletz wrote:

  

My work has OpenSuse 11.1 systems (64-bit), and I notice that Fvwm
(2.6.26 and 2.6.27) crashes when I do the following

Pick WindowStyle MiniIcon vim.png

(basically set any icon which is not the window's current icon, and
locatable by Fvwm in the image path).

This leads to an immediate crash, with no error messages. The last
couple of lines in my .xsession-errors are:

XIO:  fatal IO error 11 (Resource temporarily unavailable) on X server ":0.0"
  after 7141 requests (7000 known processed) with 0 events remaining.
XIO:  fatal IO error 11 (Resource temporarily unavailable) on X server ":0.0"
  after 102 requests (102 known processed) with 0 events remaining.

The same happens when I run Fvwm under VNC, as opposed to Fvwm under X
directly (so I'm guessing it's not server dependent). Once fvwm crashes
because of this error, it reload it (via commandline) without restarting
the VNC server causes fvwm to segfault.

I should also add that on my Gentoo laptop, I don't see the above error.

  
I checked to see if I had the issue and it does not crash my system at  
all. I'm running debian (sid) amd64 bit. I am not sure what the issue  
could be. My fvwm version is a cvs (2.5.28) from a month ago.


fvwm --version
fvwm 2.5.28 (from cvs) compiled on Jul 12 2009 at 20:59:11
with support for: ReadLine, RPlay, Stroke, XPM, PNG, SVG, Shape, XShm,  
SM, Bidi text, Xinerama, XRender, XCursor, XFT, NLS


I'd suggest first trying the current cvs version and seeing if it
works  there. If you still are getting crashes in the current cvs
version then  provide some more info like compile options from fvwn
--version to help  debug where the issue is.



Ok, I just checked with the CVS version. (I also noticed that the error
message I get after resetting the MiniIcon is a friendly "Segmentation
fault"). So I can confirm the crash on 2.5.26, 27 and 28CVS. Here's my
version info:

fvwm 2.5.28 (from cvs) compiled on Aug 18 2009 at 22:33:07 with
support for: ReadLine, Stroke, XPM, PNG, SVG, Shape, XShm, SM, Bidi
text, Xinerama, XRender, XCursor, XFT, NLS

I myself am baffled by this, since on my Gentoo box, no matter what I do
I can't crash it. Any suggestions?

GI

  
Is it only with .png files, try to use a xpm or a svg icon? I'd assume 
you'd still crash but only thing I can think of. I would also suggest to 
see if you can set the MiniIcon though a command like


Style "appname" MiniIcon file.png

Just thinking of things that could narrow down where the issue is.

jaimos



Re: FVWM: fvwm crash when setting the mini-icon

2009-08-18 Thread Jaimos F Skriletz

Gautam Iyer wrote:

Hi All,

My work has OpenSuse 11.1 systems (64-bit), and I notice that Fvwm
(2.6.26 and 2.6.27) crashes when I do the following

Pick WindowStyle MiniIcon vim.png

(basically set any icon which is not the window's current icon, and
locatable by Fvwm in the image path).

This leads to an immediate crash, with no error messages. The last
couple of lines in my .xsession-errors are:

XIO:  fatal IO error 11 (Resource temporarily unavailable) on X server ":0.0"
  after 7141 requests (7000 known processed) with 0 events remaining.
XIO:  fatal IO error 11 (Resource temporarily unavailable) on X server ":0.0"
  after 102 requests (102 known processed) with 0 events remaining.

The same happens when I run Fvwm under VNC, as opposed to Fvwm under X
directly (so I'm guessing it's not server dependent). Once fvwm crashes
because of this error, it reload it (via commandline) without restarting
the VNC server causes fvwm to segfault.

I should also add that on my Gentoo laptop, I don't see the above error.

How do I fix this?

Thanks,

GI

  
I checked to see if I had the issue and it does not crash my system at 
all. I'm running debian (sid) amd64 bit. I am not sure what the issue 
could be. My fvwm version is a cvs (2.5.28) from a month ago.


fvwm --version
fvwm 2.5.28 (from cvs) compiled on Jul 12 2009 at 20:59:11
with support for: ReadLine, RPlay, Stroke, XPM, PNG, SVG, Shape, XShm, 
SM, Bidi text, Xinerama, XRender, XCursor, XFT, NLS


I'd suggest first trying the current cvs version and seeing if it works 
there. If you still are getting crashes in the current cvs version then 
provide some more info like compile options from fvwn --version to help 
debug where the issue is.


jaimos