Re: darcs permissions error?

Excerpts from Gwern Branwen's message of Thu Mar 31 09:30:54 -0600 2011:
> On Thu, Mar 31, 2011 at 10:50 AM, Wirt Wolff <wirtwolff@...> wrote:
> > I believe keys are unchanged, so committers can ssh in and remove the
> > lock.
> 
> That doesn't seem right. When I ssh in:
> 
> gwern <at> lun:/srv/code/XMonadContrib$ ls -al
> total 60
> drwxrwsr-x   6 root     XMonadContrib-p  4096 Mar 31 16:24 .
> drwxrwsr-x 388 root     admin           12288 Mar 26 19:26 ..
> -rw-rw-r--   1 sjanssen allan            1472 Oct  6  2007 LICENSE
> -rw-rw-r--   1 aavogt   XMonadContrib-p  3227 Oct 24  2009 README
> -rw-rw-r--   1 sjanssen allan              76 Nov  4  2007 Setup.lhs
> drwxrwsr-x   9 sjanssen allan            4096 Sep 14  2010 XMonad
> drwxrwsr-x   7 sjanssen allan            4096 Jan 19 23:02 _darcs
> drwxrwsr-x   2 sjanssen allan            4096 Nov  8  2009 scripts
> drwxrwsr-x   2 sjanssen allan            4096 Apr 19  2010 tests
> -rw-rw-r--   1 aavogt   XMonadContrib-p 12628 Jan 15 18:10 xmonad-contrib.cabal
> 
> Committers can only modify the files in the XMonadContrib-p group.
> 

Oops, please forgive the misinformation. I've missed a lot of context
having been off IRC till recently, and shouldn't have made assumptions
about this issue.

What attempts have been made to solve the problem? Given that dons and
sjanssen have changed their roles, do they need to be involved in
altering groups and permissions for the repos? Would a proposal formed

Re: darcs permissions error?

Hi,

I've mailed support <at> community.haskell.org about this issue over a month ago 
(February 26), and had dons bring it up with the admins more recently 
(March 13).

Core permissions are fine.

Maybe we should just set up an alternate repo somewhere, but then there is 
the question of updating the website and other documentation that includes 
the repo url.

--
Adam

* On Tuesday, March 29 2011, Brent Yorgey wrote:

>On Tue, Mar 29, 2011 at 05:55:38PM -0400, Gwern Branwen wrote:
>> On Tue, Mar 29, 2011 at 5:43 PM, Brent Yorgey <byorgey <at> seas.upenn.edu> wrote:
>> > I tried pushing a patch to the xmonad-contrib repo today and got the
>> > following error:
>> >
>> >  darcs: takeLock /home/srv/code/XMonadContrib/_darcs/lock in
>> >  /home/srv/code/XMonadContrib: atomic_create
>> >  /home/srv/code/XMonadContrib/_darcs/lock in
>> >  /home/srv/code/XMonadContrib: permission denied (Permission denied)
>> >
>> > Anyone have any ideas about the cause?
>> 
>> Remains from the hack attack of c.h.o. Apparently no one has verified

Re: Issue 436 in xmonad: XMonad.Layout.IM doesn't match UTF8 chars

Comment #1 on issue 436 by evvs...@...: XMonad.Layout.IM
doesn't  
match UTF8 chars
http://code.google.com/p/xmonad/issues/detail?id=436

Latest xmonad package from Arch x86_64 repo have this problem too

Issue 441 in xmonad: XMonad.Util.Run.safeSpawn does not uninstallSignalHandlers

Status: New
Owner: ----

New issue 441 by dylan-go...@...: XMonad.Util.Run.safeSpawn
does not  
uninstallSignalHandlers
http://code.google.com/p/xmonad/issues/detail?id=441

I have run into programs that fail when run by safeSpawn but succeed with  
spawn.  I tracked it down in one (python) and it seems to be due to  
uninstallSignalHandlers.  When run by safeSpawn, the program reports errors  
from wait.

Arguably applications should perhaps be more robust, but it seems worth  
fixing here.

Xmonad-contrib AppLauncher

The XMonad AppLauncher doesn't work properly with filenames with spaces.
It seems to be because it uses the 'spawn' function from core. I
replaced the spawn with safeSpawn from XMonad.Utils.Run for my instance,
but I'm not sure how it would affect the "grand plan". I also can't make
a patch since the tests fail on -fglasgow-exts.

The XMonad AppLauncher doesn't work properly with filenames with spaces.
It seems to be because it uses the 'spawn' function from core. I
replaced the spawn with safeSpawn from XMonad.Utils.Run for my instance,
but I'm not sure how it would affect the "grand plan". I also can't make
a patch since the tests fail on -fglasgow-exts.

Re: Issue 441 in xmonad: XMonad.Util.Run.safeSpawn does not uninstallSignalHandlers

Comment #1 on issue 441 by gwe...@...:
XMonad.Util.Run.safeSpawn does  
not uninstallSignalHandlers
http://code.google.com/p/xmonad/issues/detail?id=441

Have you tested that adding uninstalSignalHandlers fixes spawn?

Re: Issue 441 in xmonad: XMonad.Util.Run.safeSpawn does not uninstallSignalHandlers

Comment #2 on issue 441 by dylan-go...@...:
XMonad.Util.Run.safeSpawn  
does not uninstallSignalHandlers
http://code.google.com/p/xmonad/issues/detail?id=441

Yes, the program runs successfully using:

   io $ forkProcess $ do
    uninstallSignalHandlers
    _ <- createSession
    executeFile prog True args Nothing

But fails without the uninstallSignalHandlers and createSession.  (I  
haven't tried them separately, but based on the error, the former seems  
like the culprit.)

Full example of IndependentScreens

<div><p>Hi Xmonad list..<br><br>Am a new user of Xmonad coming from awesome due to having to use Gnome at work and Xmonads Gnome integration is real nice..<br><br>Have one problem though , Multiple Screens , can't wrap my head around how xmonad handles that default and want to go for the way I'm used to have it work with separate workspaces for each screen..<br>
Found the IndependentScreens thing in the documentation but can't figure out how to get it in the configuration so needs some help..<br><br>Could someone give me a working example of a Gnome enables xmonad config that uses IndependentScreens?<br><br>Regards<br>/Ola Karlsson<br></p></div>

Issue 442 in xmonad: --recompile should support arguments to GHC

Status: New
Owner: ----

New issue 442 by k04j...@...: --recompile should support
arguments to  
GHC
http://code.google.com/p/xmonad/issues/detail?id=442

What steps will reproduce the problem?
1. In your xmonad.hs, add an import to a missing module
2. Try to recompile
3. Watch xmonad present you a message suggesting you pass the -v argument  
to see what folders it's searching. Try passing -v to xmonad --recompile  
and get the same error.

What is the expected output? What do you see instead?

What I'd like to see is for --recompile to forward any arguments after it  
to the compiler, or for there to be a separate --args argument if that's  
preferable.

What version of the product are you using? On what operating system?
0.9, ubuntu linux

Re: Full example of IndependentScreens

>    Have one problem though , Multiple Screens , can't wrap my head around how
>    xmonad handles that default and want to go for the way I'm used to have it
>    work with separate workspaces for each screen..

In Xmonad, there are N workspaces (you can define as many as you like).
Most people have 9 of them, because that's the default config. I
personally have 22 workspaces (hotkeyed to modm + 1-9, F1-F12). Each
workspace can have N windows in it. Anyway, each screen (monitor) can
view any 1 workspace at a time. That's all there's to it...

>    Found the IndependentScreens thing in the documentation but can't figure
>    out how to get it in the configuration so needs some help..

The "Usage" section at
http://xmonad.org/xmonad-docs/xmonad-contrib/XMonad-Layout-IndependentScreens.html
seems to have everything you're looking for. After editing your
xmonad.hs, just run "ghci" on it: ghci xmonad.hs. If you get any errors,
then your config is messed up.

>    Could someone give me a working example of a Gnome enables xmonad config
>    that uses IndependentScreens?

A quick google search got me this:
http://www.haskell.org/pipermail/xmonad/2009-December/009466.html
Maybe you should ask the user there about it...

Personally though, I'd just stick with the default. There are lots of
neat contrib modules out there (e.g., CycleWS) that rely on the default
setup (1 set of workspaces). You'd have to hack all of those modules to
get them working with your IndependentScreens setup.

-Linus