April 2010 - xmonad - Haskell.org

Issue 340 in xmonad: Error In Passing Clicks in Multi-Head non-Xinerama Setup
by codesite-noreply＠google.com 12 Aug '12

12 Aug '12

Status: New Owner: ---- New issue 340 by llcawthorne: Error In Passing Clicks in Multi-Head non-Xinerama Setup http://code.google.com/p/xmonad/issues/detail?id=340 What steps will reproduce the problem? 1. Set 'Option "Xinerama" "0"' in xorg.conf in a multi-head non-xrandr setup. 2. Run xmonad on the primary display 3. Run another copy of xmonad, or another WM on the secondary display. What is the expected output? What do you see instead? I should be able to re-enter a window managed by XMonad and click on things after leaving it. There are two different setups that respond slightly differently. 1) Xmonad on DISPLAY 0.0 and No WM on DISPLAY 0.1 (or Openbox). I can leave the XMonad display perfectly every time and click on things on the secondary display. The problem pops up when XMonad needs to re-accept control of the pointer. I come back to my main screen and clicks are not accepted by whatever window I entered into. Buttons highlight like I am over them; links get little boxes around them, but clicks do not register. The problem can be remedied by moving my mouse over any other window on the same display, and things "catch up". Ex: DISPLAY=0.0. I usually run Firefox, and three terminals. DISPLAY=0.1. I usually have Thunderbird open. If I go to Thunderbird, I never have any problem getting there (as long as I am not running XMonad on display 0.1 also). If I come back from thunderbird to Firefox (which is in the master tile), I cannot successfully click anything. If I move my mouse over to a terminal (with or without clicking on it) and then back to Firefox, I can click things fine. 2) Running xmonad on Display 0.0 and Display 0.1 Same as above, except I have problems moving to each display. Previous Example but: If I leave Display 0.0 and go to Thunderbird, I cannot click anything. Once again, messages shadow when the mouse is over them, as do buttons in Thunderbird. If I hit Mod+Shift+Return at this point, the terminal will open on Display 0.0, even though I am on Display 0.1 with my pointer. Now, if I move my pointer back to Display 0.0, I cannot click on Firefox and if I hit Mod+Shift+Return the terminal will open on Display 0.1 even though the mouse is on Display 0.0. Everything is back to normal if I mouse over a terminal and back to Firefox. (This was especially annoying until I figured out what was going on.. Now I run with a clock open on Display 0.1 so I have a second window to mouse over on the way to read my e-mail, because otherwise I had no way to make Thunderbird start receiving clicks). Things work the same with application other than Firefox/Thunderbird termimal in both examples, they are just what I tend to have open. What version of the product are you using? On what operating system? Xmonad 0.8 or Xmonad 0.9. Gentoo Linux or ArchLinux. Are you using an xmonad.hs? Please attach it and the output of "xmonad --recompile". Problem exists with default xmonad.hs or with a custom xmonad.hs (or with a randomly grabbed off the website xmonad.hs). Please provide any additional information below. I can set "Xinerama" "1" and there is no problem; I just don't particularly care for Xinerama. A) It turns off compositing. B) I cannot (at least don't know how to) launch apps to other screens from a prompt (ie 'DISPLAY=:0.2 mplayer -fs South.Park.S01312.avi'). Also, I generally use three monitors, but I can recreate the problem with the secondary card and tertiary monitor removed from my configuration, so it has not been mentioned here. (Things get even stranger when I try running xmonad on all three displays without Xinerama). If you need more info than my wordy description, I have a forum post on the ArchLinux boards and sent a message to the mailing list a bit back relating to this problem. (Note that the first few messages of the posting were before I figured out clearly what was happening). http://bbs.archlinux.org/viewtopic.php?id=83789 -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 3

Re: Issue 276 in xmonad: switching keyboard focus to a wine app is odd
by codesite-noreply＠google.com 29 Feb '12

29 Feb '12

Comment #2 on issue 276 by byorgey: switching keyboard focus to a wine app is odd http://code.google.com/p/xmonad/issues/detail?id=276 What is it about the way xmonad and wine interact that cause this? Obviously keyboard focus works fine for (almost) all other applications, so what's different about wine? -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 4

Re: Issue 339 in xmonad: Support for toggling totem movie player fullscreen
by codesite-noreply＠google.com 03 Jan '12

03 Jan '12

Updates: Summary: Support for toggling totem movie player fullscreen Status: Accepted Labels: -Priority-Low Priority-Medium Comment #5 on issue 339 by wirtwo...@gmail.com: Support for toggling totem movie player fullscreen http://code.google.com/p/xmonad/issues/detail?id=339 Nice work audunskaugen. Thanks for researching this and providing the _NET_WM_STATE_FULLSCREEN event hook (totemFullscreen.hs sample xmonad.hs). IMO, this will be good to get into contrib soon to make it available for more usage, testing and further development. There are some aspects to improve so if anyone wants to take on ownership and submit a patch addressing them that would be great. Otherwise if no action in a week or so I'll send a patch to add it to EwmhDesktops as is, initially as a standalone 'fullscreenEventHook'. Eventually, though, I think it would be good to include an improved version in the standard ewmhDesktopsEventHook, plus having it available by itself. (sjanssen feedback on IRC when asked for thoughts as to where to put this: "enabling the full screen event hook in EWMH would be fine, and would probably make sense.") As is it gives more function for totem users and surely some other similar apps too than the current (0.9) situation, but also with some additional work it seems like it can cover uses of the doFullFloat manageHooks with isFullscreen and the fullscreen query alexey attached to http://code.google.com/p/xmonad/issues/detail?id=333 It seems it covers tiled to fullscreen and back usage plus allowing multihead users to focus other screens without affecting fullscreened apps. In my quick tests it works for totem, flash, mplayer, mplayer-plugin, gnome-mplayer plugin. (This is with all fullscreen manageHooks disabled.) It doesn't seem to help with ooimpress unfortunately, nor evince. Areas to address: * Toggling seems less smooth than it should be. In my tests it appears that the window untiles, goes to hinted float size, then finally fullscreen although it's hard to tell as it's very rapid. There is some glitch anyway that I didn't research yet as to cause. Also other tiled windows from the same workspace look briefly as if they are composited with the fullscreen app since they change size to fill the space as the fullscreen app floats. Not sure how much improvement can practically be made here, but hopefully can something close to the nice mapping with fullscreen manageHook. * Whether a window starts out tiled or floating, once toggled fullscreen and back it restores to tiled state. My reading of wm-spec and probably most users expectations is that floats return to float instead. I don't think PositionStore would be needed for this but didn't look into it yet either. audunskaugen, thanks again. If you'd like please send a darcs patch to X.Hooks.EwmhDesktops, or if you want to be credited by some other name you are welcome to email me privately with your preference and I'll include in any patch I prepare. -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 12

Issue 363 in xmonad: non-positive argument error on NVidia Xinerama set
by codesite-noreply＠google.com 14 Dec '11

14 Dec '11

Status: New Owner: ---- New issue 363 by seanerussell: non-positive argument error on NVidia Xinerama set http://code.google.com/p/xmonad/issues/detail?id=363 What steps will reproduce the problem? 1. 2. 3. What is the expected output? What do you see instead? What version of the product are you using? On what operating system? Are you using an xmonad.hs? Please attach it and the output of "xmonad --recompile". Please provide any additional information below. -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 10

Re: Issue 126 in xmonad: xmonad doesn't warn when xmonad executable cannot be found
by codesite-noreply＠google.com 12 Nov '11

12 Nov '11

Updates: Summary: xmonad doesn't warn when xmonad executable cannot be found Status: Accepted Labels: Usability Comment #6 on issue 126 by vogt.adam: xmonad doesn't warn when xmonad executable cannot be found http://code.google.com/p/xmonad/issues/detail?id=126 In #xmonad today, jeffwheeler had some trouble getting his config to reload. It seems like this problem could have been avoided by adding a bit of logic to the default config: Currently M-q is bound to: spawn "xmonad --recompile && xmonad --restart" This could instead be something like: spawn $ "which xmonad && { xmonad --recompile && xmonad --restart } || " ++ "xmessage xmonad not in \$PATH: \"$PATH\"" But maybe this logic should be implemented in haskell, or is the /bin/sh syntax used here going to work everywhere? -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 3

Issue 372 in xmonad: Changes made to defaultConfig do not take on recompile and restart.
by codesite-noreply＠google.com 05 Sep '11

05 Sep '11

Status: New Owner: ---- New issue 372 by aditya.siram: Changes made to defaultConfig do not take on recompile and restart. http://code.google.com/p/xmonad/issues/detail?id=372 What steps will reproduce the problem? 1. change any defaultConfig setting, in my case change terminal = "gnome-termimal" to terminal = "xterm", recompile (xmonad --recompile) and restart(xmonad --restart) What is the expected output? What do you see instead? mod-shift-return should now pull up an xterm. Instead it still opens a gnome-terminal. What version of the product are you using? On what operating system? 0.9.1, Ubuntu Karmic Are you using an xmonad.hs? Please attach it and the output of "xmonad --recompile". config file is attached. Recompile step has no output. Attachments: xmonad.hs 2.0 KB -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 1

Issue 328 in xmonad: RunOrRaise and dbus(?)
by codesite-noreply＠google.com 05 Sep '11

05 Sep '11

Status: New Owner: ---- New issue 328 by grapescan: RunOrRaise and dbus(?) http://code.google.com/p/xmonad/issues/detail?id=328 What steps will reproduce the problem? 1. Set some bindings like this: ("M-t", runOrRaise "transmission" (className =? "Transmission")) or ("M-w", runOrRaise "firefox" (className =? "Shiretoko")) 2. Try to run these programs with created shortcuts 3. Try to run these programs without created shortcuts (i.e. from terminal or gmrun) What is the expected output? What do you see instead? Starting from terminal or gmrun looks fine while using shortcuts causes strange behavior. Firefox complains about connecting to session: "/usr/bin/dbus-launch terminated abnormally without any error message". Transmission doesn't want to open torrents from Firefox (torrents opened with Transmission from web-browser). It says something like "Transmission is already started." What version of the product are you using? On what operating system? I use actual version of ArchLinux. I checked dbus packages - they seem to be OK. Packages xmonad-darcs and xmonad-contrib-darcs built a few weeks ago. Attachments: xmonad.hs 5.3 KB -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 3

Re: Issue 322 in xmonad: Background shown when workspace switches
by codesite-noreply＠google.com 17 Aug '11

17 Aug '11

Comment #2 on issue 322 by sgsoftware: Background shown when workspace switches http://code.google.com/p/xmonad/issues/detail?id=322 I am not sure that this is strictly an X server problem since I've got similar behavior after upgrading xmonad from version 0.8.1 to version 0.9. Apparently something has changed in between and now xmonad draws background so that it is very noticeable (and quite annoying). In the previous version of xmonad switching between two full-screen windows was immediate and I could see no background. -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 6

Issue 325 in xmonad: Layout.Spacing breaks Layout.WindowNavigation
by codesite-noreply＠google.com 15 Aug '11

15 Aug '11

Status: New Owner: ---- New issue 325 by OrbisVicis: Layout.Spacing breaks Layout.WindowNavigation http://code.google.com/p/xmonad/issues/detail?id=325 Removing the spacing modifier in the layoutHook or setting "spacing 0" fixes the problem. Symptoms: While Go or Swap *might* initially work, they very quickly lock up and do nothing. -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 9

Issue 373 in xmonad: Hlintify XMonad & XMonadContrib
by codesite-noreply＠google.com 11 Aug '11

11 Aug '11

Status: Accepted Owner: gwern0 Labels: Type-Enhancement Priority-Low Component-Contrib Maintainability Component-Core New issue 373 by gwern0: Hlintify XMonad & XMonadContrib http://code.google.com/p/xmonad/issues/detail?id=373 Hlint (http://community.haskell.org/~ndm/hlint/) is a code-style tool that flags verbose, useless, or otherwise bad bits of code. It would be good if XMC had clean code. (Hlint strives for 'correctness'; none of the changes ought to affect the semantics of the code.) XMonad-core generates roughly 31 suggestions and XMC roughly 468. Even excluding debatable suggestions like eta-reduction, this is a lot of small bad things. Fortunately, they are all generally very easy to fix and understand why the code was flagged, making this a good beginner project. Attached are the 2 HTML reports hlint generates (obviously 31+468 messages is a bit much to paste in). Attachments: core.html 38.5 KB xmc.html 528 KB -- You received this message because you are listed in the owner or CC fields of this issue, or because you starred this issue. You may adjust your issue notification preferences at: http://code.google.com/hosting/settings

1 2