- xmonad - Haskell.org

mod events not reaching xmonad
by Bas van Dijk 18 Apr '07

18 Apr '07

Hello, I have a problem. Often, after a while of using OpenOffice with xmonad mod key events no longer get to xmonad. When this happens I'm forced to switch to a console and killall xmonad loosing all non-saved work in the process. I've also seen this using other applications like Firefox for example. What can be causing this and do other people have the same problems? regards, Bas van Dijk P.S. I run the latest xmonad

3 4

darcs patch: refactor layout interface.
by David Roundy 17 Apr '07

17 Apr '07

Hi all, Here's an attempt at an interface that allows the definition of custom workspace layouts in Config. I'm not really pleased with it, so you can view it as a discussion piece more than an actual proposal. In particular, it punts on the idea of any sort of mouse interaction with a layout, which almost makes a tabbed layout useless. Still, I think it's an interesting approach to allowing layouts that contain arbitrary sorts of state. David Mon Apr 16 20:48:13 PDT 2007 David Roundy <droundy(a)darcs.net> * refactor layout interface. This change only changes functionality in that the tileFraction is now a per-layout variable. The main point is to add the flexibility to define layouts in the Config file. There's a bit of cheesiness in the type of modifyLayout, since I could figure out a pleasant choice for passing the key-binding information to the Layout, so I just used a string.

1 0

darcs patch: fix typo. (and 1 more)
by David Roundy 16 Apr '07

16 Apr '07

Hi all, Here's a feature I've added (on my quest to make xmonad non-minimal...) to my version of xmonad. It's pretty crude, but I think some sort of "rotating" is in order. This way if I have only three workspaces with anything in them, I don't have to remember their numbers, and don't have to search through all nine options. David Sat Apr 14 22:56:16 PDT 2007 David Roundy <droundy(a)darcs.net> * fix typo. Sat Apr 14 23:30:34 PDT 2007 David Roundy <droundy(a)darcs.net> * add code to rotate through non-empty workspaces. I don't expect that this will actually be accepted, as my code's pretty ugly, but it's a feature I want, and might inspire some ideas.

5 11

another idea for configurable layouts.
by David Roundy 16 Apr '07

16 Apr '07

Hi all, I've had another idea for how we might implement configurable layouts, which would have the advantage (or disadvantage?) of removing the tileFraction from the StackSet, and moving it into the layout itself. I don't have time just yet to hack this up, but thought I'd run the idea by you. The idea is to have a type Layout which holds whatever information is needed to determine how to lay out the windows: data NewLayout = L { doLayout :: Rectangle -> [Window] -> [(Window, Rectangle)] , modifyLayout :: Event -> Maybe NewLayout } The doLayout does the actual laying out (e.g. calls tile or vtile). The second function, modifyLayout would be in charge of updating whatever state there might be (e.g. changing the tileFraction. A very sketchy example would be this sketch of the Tall layout: tall :: Rational -> NewLayout tall tileFraction = L { doLayout = tile tileFraction , modifyLayout _ = Just $ tall (tileFraction*1.5) } Obviously modifyLayout is just wrong, but I don't have all that much time. The point is that it'd be able to update the tileFraction, and by using closures like this, we wouldn't have to hard-code what extra data (e.g. tileFraction) needs to be stored with each workspace. The catch is that I've changed the semantics, as each layout will now have its own tileFraction. I don't think that's a Bad Thing, but you may disagree. The other controversial (or at least it's controversial to my mind) aspect of this is the use of Event in modifyLayout. It's pretty ugly (particularly with regard to key bindings), but I can't see how else to allow a tabbed layout to respond to mouse clicks. I imagine we'd need some sort of translation layer to make keybindings pretty... but I can't solve all the worlds problems overnight. Anyhow, this is the closest I've come to imagining a StackSet data format that will allow for the implementation of clever new layout algorithms, which might need some extra data, and will certainly need to respond to extra events. I should also mention that I imagine mod-space working by storing a list of layouts for each workspace, and rotating between them. Obviously, when you change layout, you'd like to keep the old one so you can go back to it (e.g. so you can switch to fullscreen and back without losing your tileFraction). -- David Roundy http://www.darcs.net

1 1

gmrun and pasting (was Re: [Xmonad] darcs patch: fix typo. (and 1 more))
by Alec Berryman 16 Apr '07

16 Apr '07

Gwern Branwen on 2007-04-15 23:20:24 -0400: > Alec Berryman <alec(a)thened.net> writes: > > > Donald Bruce Stewart on 2007-04-16 09:13:31 +1000: > > > >> droundy: > >> > > >> > After that, I'd be interested in a decent input prompt (which dmenu is > >> > not). Ion's is better, but I'd rather figure out how to make use of an > >> > actual bash shell, as the tab completion of ion's query box is annoying > >> > (but better than nothing). > >> > >> Does anyone know of a 3rd party tool for this? Like dmenu , but for more > >> complex input? > > > > gmrun includes tab-completion, handles URLs, and can open specified > > programs in a shell. > > Does gmrun also handle pasting, either through a mouse or a keybinding > (preferably the latter)? Yes to both, but due to either a bug in gmrun or xmonad, it isn't particularly convenient. If I copy something to the X11 clipboard by selecting it with the mouse, then run gmrun via the default keybinding, I can't paste either with a middle click or with shift-insert. If I switch to another window, I can still paste what I had previously copied, and if I now copy something to the clipboard and switch back to the still-running gmrun I can paste fine.

1 0

The kill client / mod-shift-c issue
by dons＠cse.unsw.edu.au 15 Apr '07

15 Apr '07

Ok. After investing some more, and still not obviously seeing what's wrong, I find that: * sometimes, after running xmonad for a long time, I can't kill xclock (for example), or any other app that uses the wmdelete protocol (requiring us to send delete notification). * quitting out completely, and logging back in, with a new build, and everything works. * restarting, and it still works. Also, the case when we need to send this event is still spotted correctly,so the bug is in the X11-extras code I suspect. I'll look into this some more if I can work out how to reproduce it. David, can you pull all the latest patches, rebuild and install xmonad. Quit X, log back in, then check if you can kill xclock? -- Don

1 0

how to install?
by David Roundy 15 Apr '07

15 Apr '07

Hi all, I made a somewhat desultory attempt at compiling xmonad, and am sad to admit that I've failed, mostly because I've never learned to use cabal. Yes, I'm terribly old-fashioned and out-of-date... In the X11-extras directory I run: ./Setup.lhs configure --prefix=/home/droundy ./Setup.lhs build ./Setup.lhs install and install fails because of lack of permissions. It seems that I could instead use ./Setup.lhs copy, but then when I configure xmonad, cabal fails to find X11-extras. Any suggestions? Also, the configure step in X11-extras doesn't seem to bother checking whether the xinerama library is present. Is this a bug in X11-extras, or in cabal? I certainly would hope that cabal would have the functionality to check for presence of libraries... -- David Roundy http://www.darcs.net

6 12

how amenable to making xmonad less minimalistic?
by David Roundy 15 Apr '07

15 Apr '07

Hi all, I've just gotten xmonad running (in a separate X server, for simplicity), and it looks like a nice start, and the code looks pretty elegant, but I'm wondering how amenable it's going to be to added complexity? I'm a current ion3 user who's mostly content, but a bit frustrated by its limitations, and the complexity of configuring it using lua. On the other hand, I've also gotten my setup working pretty much as I like now, so it's not really worthwhile to switch. The features I most like in ion3 that are missing in xmonad are the (very crude) ability to configure the layout based on the windows themselves and the tabs. And what I *really* like (which took a fair amount of sweat and pain) is the ability to set up a single command that prompts me for a directory, and then creates a new workspace with the layout I want, and in it launches several programs (typically an emacs, a couple of terminals and an xclock, but only one terminal on my laptop), with the chosen directory set as the working directory for each program, and any future programs launched in that workspace also has the same working directory. All of this would be much easier to code in Haskell, provided the framework is there. So I'm just wondering if there will be hooks to provide some of this. LayoutDesc, for example, seems to be written in a sufficiently minimalistic way that adding a new layout would seem somewhat like surgery. I imagine we could have a design something like data LayoutDesc = LayoutDesc { !layout :: Rectangle -> [Window] -> [(Window, Rectangle)] } which would have the advantage of allowing commands to introduce arbitrary bindings. The catch is that we've lost the ability to tune the tileFraction, so I've not come up with a real solution. But I'd like to be able to play with new layouts without hacking directly in Operations.hs, and we really shouldn't need to add new parameters to LayoutDesc for every degree of freedom we might want to give our layout. i.e. there's got to be a better way! My first goal (but not most important) would be to ensure that my xclocks always remain square. This would avoid much of the insanity that comes with the current inflexible layout of xmonad (i.e. either the xclock takes up much of the screen and is very distorted, or all but one of my windows is a small square of wasted space). -- David Roundy http://www.darcs.net

2 1

darcs patch: a hypothetical layout approach.
by David Roundy 15 Apr '07

15 Apr '07

Hi all, Here's my sketch of how we might make layouts configurable. It defines a rather silly layout, bound to mod-shift-space. I'm not totally happy with this option, as custom layouts are left out of the cycling with mod-return, and I suspect some sort of a Data.Map.Map from (a modified) Layout to layout functions would be better, so that "builtin" layouts could be treated on a more equal footing with "custom" layouts. David (who once again is spending too much time customizing his window manager...) Sun Apr 15 08:55:17 PDT 2007 David Roundy <droundy(a)darcs.net> * a hypothetical layout approach.

2 1

Redefining modMask
by Bas van Dijk 13 Apr '07

13 Apr '07

Hello again, I would like to assign my modMask to the windows key on my keyboard. Using the xorg tool 'xev' I discovered that the keycode of my windows key is 115. However, setting modMask = 115 in Config.hs doesn't work. Do I have to use a key symbol from Graphics.X11.Types for that? If so, which one? Thanks, Bas van Dijk

3 3