- xmonad - Haskell.org

Further xmodmap and numlock woes
by Thomas Løcke 21 Mar '13

21 Mar '13

A while ago I wrote a post to this list about the problems that arose when I disabled numlock with xmodmap -e "clear mod2": http://www.haskell.org/pipermail/xmonad/2013-February/013493.html I ended up "fixing" the problem in the legacy application, so I didn't have to muck about with xmodmap. Or so I thought. Today it reared its ugly head again, once and for all telling me that I seriously need to _completely_ disable the numlock key. My first step was to upgrade xmonad to 0.11 and then I tried xmodmap -e "clear mod2" again and of course it still didn't work: My mod key was now non-functional again. /sigh Next step was to dig out the original suggestion from Adam: handleEventHook = ( \e -> case e of MappingNotifyEvent {} -> return (All False) _ -> mempty) I added the above to my xmonad.hs file and then it failed with this: Not in scope: data constructor `All' Not in scope: `mempty' I suspect I need to import something, but what? Here's my xmonad.hs file: https://gist.github.com/ThomasLocke/5196294 Am I missing some crucial parts? :o) Thomas Løcke ps. I found xmonad at github: https://github.com/xmonad/xmonad is this an official repository?

4 7

My hack to allow disabling of numlock without messing with my xmonad mod key
by Thomas Løcke 21 Mar '13

21 Mar '13

In short: I've thrown a script together that executes the xmodmap command needed to completely disable to numlock key. What this script does is make sure that xmonad is up and running _before_ xmodmap is called. Combined with Adam's addition to my handleEventHook in xmonad.hs I've now managed to squash the numlock key into oblivion. Mission accomplished! I can't say I like this solution, but at least it gets the job done. I would still very much prefer if this could've been fixed in xmonad.hs, or if xmonad would just let me do my xmodmap'ing without destroying my mod key, but as that does not seem to be possible, I'll settle for second best. So if anybody else out there is having this problem, a bit of BASH will solve it. Just make sure you have this in your handleEventHook: ( \e -> case e of MappingNotifyEvent {} -> return (All False) _ -> mempty) and that your xmodmap calls are done _after_ xmonad is fully loaded. I call my script from .xinitrc just before I start xmonad. The script is backgrounded and it simply loops until it finds a running xmonad instance. :o) Thomas Løcke

4 10

Re: [xmonad] Ncurses and drop-down menus
by Matthew Hague 20 Mar '13

20 Mar '13

> Really strange! You should post your config somewhere (say, > hpaste.org) Are you using an urgency hook? > ~d Aha! -- i tracked it down (always the way once you've decided to post). The problem is in a loghook i set (i'd tried simply removing the entire loghook and the problem continued, but removing the offending function fixes it, it turns out). My config is available here https://github.com/yourealwaysbe/xmonad-config/blob/master/xmonad.hs in particular i have a loghook: myLogHook = floatsAvoidStruts <+> raiseFocused <+> hideFloats where raiseFocused :: X () raiseFocused = do disp <- asks display mw <- gets (W.peek . windowset) whenJust mw (io . (raiseWindow disp)) which is to raise the focussed window above the others, because i have a free-form messy workspace which sometimes has a few tiled windows combined with a few floating ones. So if a tiled window is focussed, all the floats get lowered (so i can see the tiled layer), and if a float is focussed, it's raised above everything else (so i can see it). I know this isn't really in the spirit of xmonad, but hey... I guess ncmpcpp triggers the loghook when the play status updates, so then raiseFocused raises the main GUI app window (with focus) above the drop-down menu window (which i guess doesn't have focus). Sorry for the noise, i'll work on writing a better raiseFocused :) Best, Matt

1 0

Ncurses and drop-down menus
by Matthew Hague 19 Mar '13

19 Mar '13

Hi, I've noticed a rather bizarre interaction between ncmpcpp (using ncurses) and drop down menus in applications such as Firefox, Evince and VLC (hence i don't think it's a specific gtk/qt issue). When playing a track, there is a little progress indicator at the bottom of ncmpcpp. If i have a drop-down menu open in a GUI app (Firefox/Evince/VLC) and the indicator updates, then the menu window disappears. Even if ncmpcpp is not on the same workspace as the GUI app. Since ncmpcpp's progress meter updates about every second, this means i can't use drop down menus while listening to music... It does not happen in fluxbox/openbox or awesome. Has anyone else noticed this / have any ideas how to fix it? Thanks, Matt

3 2

Issue 534 in xmonad: X11 version 1.6.1.1 breaks the build
by codesite-noreply＠google.com 18 Mar '13

18 Mar '13

Status: New Owner: ---- New issue 534 by sim...(a)cryp.to: X11 version 1.6.1.1 breaks the build http://code.google.com/p/xmonad/issues/detail?id=534 xmonad fails to compile against the latest version of the X11 library with the following error: XMonad/Hooks/DebugEvents.hs:215:61: Ambiguous occurrence `getAtomName' It could refer to either `XMonad.Hooks.DebugEvents.getAtomName', defined at XMonad/Hooks/DebugEvents.hs:57:1 or `XMonad.getAtomName', imported from `XMonad' at XMonad/Hooks/DebugEvents.hs:(24,1)-(26,62) (and originally defined in `Graphics.X11.Xlib.Atom') -- You received this message because this project is configured to send all issue notifications to this address. You may adjust your notification preferences at: https://code.google.com/hosting/settings

1 7

darcs patch: gnome3Config added
by woozletoff＠gmail.com 18 Mar '13

18 Mar '13

1 patch for repository http://code.haskell.org/XMonadContrib: Gnome 3 doesn't support the "gnome-session-save --kill" command to logout, so it might be reasonable to add a "gnome3Config" which defines the logout binding. It doesn't destroy backwards compability, so there should be no harm done. Mon Mar 18 14:42:06 CET 2013 woozletoff(a)gmail.com * gnome3Config added

2 1

Issue 500 in xmonad: Actions.UpdatePointer prevents Chromium save bookmark dialog from appearing.
by codesite-noreply＠google.com 13 Mar '13

13 Mar '13

Status: New Owner: ---- New issue 500 by ancientpc: Actions.UpdatePointer prevents Chromium save bookmark dialog from appearing. http://code.google.com/p/xmonad/issues/detail?id=500 What steps will reproduce the problem? 1. Xmonad 0.9 2. logHook = updatePointer (Nearest) 3. Open Chromium, try to bookmark a page (either using ^D or clicking on the star). What is the expected output? What do you see instead? expected: Chromium bookmark dialog box actual: The star turns yellow, but no dialog box.

1 3

darcs patch: XMonad.Config.Prime, a do-notation for config
by me＠twifkak.com 13 Mar '13

13 Mar '13

Hello, folks, It's me again. First off, I see that you released 0.11 in Dec. Nice! Second, I took some time to take another crack at this config thing. (Previous thread: http://www.haskell.org/pipermail/xmonad/2012-September/013045.html.) Because of the problems you guys mentioned with Template Haskell, I: 1. replaced the fclabels dependency with the lighter data-accessor. 2. replaced the StateT with a fake monad, so the type of the layoutHook can be changed without that existential mess. The upside is an even cleaner syntax. The downside is: 1. You have to come to terms with having RebindableSyntax in your xmonad.hs. 2. Error messages are uglier (thanks, in part, to the really long layout types). Docs at: http://twifkak.com/xmonad-junk/XMonad-Config-Prime2.html Thoughts? Devin 1 patch for repository http://code.haskell.org/XMonadContrib: Mon Mar 11 02:46:17 PDT 2013 me(a)twifkak.com * XMonad.Config.Prime, a do-notation for config Note that the use of RebindableSyntax is because of the need to vary the layoutHook type throughout the config. The alternative, using the existential Layout type, was rejected because it required TemplateHaskell in order to look nice, and TemplateHaskell is not portable.

3 2

Moving windows to the workspace of their parent window
by Mina Nagy Zaki 10 Mar '13

10 Mar '13

I'd like to write a managehook that shifts windows to the workspace of their parent, without having to restrict that workspace (ie, without predefining a specific workspace for these windows). For example if I run skype on one workspace, when I get a call and answer it, I'd like the call window to automatically shift to the same workspace as skype. How would I go about this? -- Mina Nagy Zaki

3 3

Darcs patch to expose arbitrary events to event hooks
by Marshall Lochbaum 02 Mar '13

02 Mar '13

Last year this set of patches was submitted to the xmonad mailing list. It allows the user to set which types of events will be passed to event hooks by moving the two variables rootMask and clientMask from being hard-coded constants to configurable parameters. http://www.haskell.org/pipermail/xmonad/2012-January/012297.html This message got no replies when it was sent, but I have found it very useful. I have attached a few scripts that I have used with the patch: these allow me to spawn things in the root window using a single keypress without modifiers, to map caps lock to a "repeat last key" key that makes typing words with double letters a bit easier, and to implement a "show desktop" key that hides all windows when pressed and returns them when released. Since rootMask and clientMask are not modified by default, this won't break any current xmonad behavior. It also only impacts event hooks, so there is no chance of damaging other parts of xmonad with incorrect configuration--the affected event hooks simply won't receive some events. Can someone apply these changes to xmonad? I have attached a patch which merges the two from the original message (there's really no reason to apply one and not the other) and works with the latest darcs source. If these are added I will also clean up and formally submit the other scripts. Marshall

3 3