- xmonad - Haskell.org

unclutter
by Andrea Rossato 03 Sep '07

03 Sep '07

Hi, yesterday there has been a complain about the way unclutter (the small utility to hide the pointer after some inactivity) interacts with XMonad. If you run it without the -grab option and you happen to have the pointer over the borders of the screen, it is not possible to change the focus anymore. This is because XMonad uses the mouse to focus windows.[1] This is due to the unclutter default behavior, which uses fake X events to interact with other X clients. So I rewrote unclutter in Haskell. There's a small issue with threads and the timer, but it is usable here. You can grab it with other haskell utilities for xmonad here: darcs get http://gorgias.mine.nu/repos/xmonad-utils/ or get it directly here: http://gorgias.mine.nu/repos/xmonad-utils/src/Hhp.hs together with this: http://gorgias.mine.nu/repos/xmonad-utils/src/Utils.hs This is a collection of utilities I've already written about. I'm going to release them as soon as X11-extras-0.3 will be released. Andrea [1] btw there is another issue with that: open 4 windows and float and resize the last one. Focus the first or the second window and switch to the full layout. Now Enter the floating window with the mouse. Did you notice something on the background? You are on the 3rd window.

2 1

Re: [Haskell-cafe] Hawiki articles
by Andrea Rossato 03 Sep '07

03 Sep '07

On Mon, Sep 03, 2007 at 05:20:22PM +0100, Neil Mitchell wrote: > > Bring back HaWiki! > > I couldn't agree more! We built up an incredible array of articles, by > fantastic authors with stunning content - which we then deleted... I > learnt much from the old wiki, and it would be a shame if others > didn't get that opportunity. I can't say it better! Bring it back please. Andrea

1 0

Switching workspace ...
by kg 03 Sep '07

03 Sep '07

Dear all, I am a beginner of Haskell language. I am a beginnner of xmonad :) I am a Ion3's user who wants leave Ion3 (for many reasons ...) So, I have tested Xmonad. All fonctionnalities are ok execpts workspaces: I follow the guided tour and I try to "use other workspaces" : " xmonad has, by default, 9 virtual workspaces. You can switch between them using *mod-1* to *mod-9*. Switching to workspace 4 *mod-4* we find it empty:" But I don't find a empty workspace. PS: -I have 2 screen -> Xinerama. I can switch between screens (mod-[ew]) -I beleve the shortcuts is "reserved" for xmonad (Nothing is written in xterm) Have you got ideas ? Thx in advance.

5 10

a safer ghci prompt
by Andrea Rossato 03 Sep '07

03 Sep '07

Hi, attached you'll find a safer ghci prompt, that uses an external program to evaluate haskell expression. This is still work in progress, but you can have an idea of what I'd like to do. Feedback highly appreciated! ciao andrea ps: to use: 1. save the attached file as GhcPrompt.hs and follow the install instruction; 2. save and compile the uitlity you'll find at the end of the file. 3. enjoy.

1 3

XSelection.hs - Unicode problems
by Gwern Branwen 02 Sep '07

02 Sep '07

So I was working on splitting out Andrea's hxsel into something useful for one's Config.hs, and I'm basically done, and it seems to work well. However, while it works fine for your basic bread and butter ASCII characters, I noticed that it does terrible things to more exotic phrases involving Unicode, such as "Henri Poincaré". I borrowed some code from utf-string, and that improved it a little bit - "Henri Poincaré" now becomes "Henri Poincarý" which is still better than "Henri Poincar\245" or whatever. As far as I can tell, whenever a String involving UTF-8 stuff leaves the Haskell environment, it gets messed up. This is a little hard to test since things get messed up even when I test them in GHCi. :) But I'm sure it has to be something to do with Haskell, since I know I can copy and paste such strings with no problems using the mouse, and I know that my shell isn't the problem and nor are the Surfraw programs I use to pass them to Firefox (and Firefox obviously has no problems handling those characters). utf-string does have a special IO module which can print out and receive UTF-8 strings, but it's limited to things like 'putStr' and 'getContents'. There doesn't seem to be anything that would be useful for fixing 'spawn', which seems to be the specific function that needs fixing, since it isn't using any of them and executeFile doesn't seem to use any of them either. spawn x = io $ do pid <- forkProcess $ do forkProcess (createSession >> executeFile "/bin/sh" False ["-c", x] Nothing) exitWith ExitSuccess getProcessStatus True False pid return () Any ideas on how to fix this? ----------------------------------------------------------------------------- -- | -- Module : XMonadContrib.XSelection -- Copyright : (C) 2007 Andrea Rossato -- License : BSD3 -- -- Maintainer : andrea.rossato(a)unibz.it -- Stability : unstable -- Portability : unportable -- -- A module for accessing and manipulating the X Window mouse selection (used in copy and pasting). -- ----------------------------------------------------------------------------- {- $usage Add 'import XMonadContrib.XSelection' to the top of Config.hs Then make use of getSelection or promptSelection as needed. TODO: * Fix Unicode handling. Currently it's still better than calling 'chr' to translate to ASCII, though. As near as I can tell, the mangling happens when the String is outputted somewhere, such as via promptSelection's passing through the shell, or GHCi printing to the terminal. utf-string has IO functions which can fix this, though I do not know have to use them here. * Add a 'putSelection' to allow modification of the selection. * Possibly add some more elaborate functionality: Emacs' registers are nice. -} module XMonadContrib.XSelection (getSelection, promptSelection) where -- getSelection's imports: import Graphics.X11.Xlib (allocaXEvent, createSimpleWindow, defaultScreen, destroyWindow, internAtom, nextEvent, openDisplay, rootWindow, selectionNotify, sTRING) import Graphics.X11.Xlib.Extras (currentTime, ev_event_type, getEvent, getWindowProperty8, xConvertSelection) import Data.Maybe (fromMaybe) import Data.Char (chr) -- promptSelection's imports: import XMonad (io, spawn, X ()) -- decode's imports import Foreign -- (Word8(), (.&.), shiftL, (.|.)) -- | Returns a String corresponding to the current mouse selection in X; if there is none, an empty string is returned.. Note that this is -- only reliable for ASCII text and currently mangles/escapes more complex UTF-8 characters. getSelection :: IO String getSelection = do dpy <- openDisplay "" let dflt = defaultScreen dpy rootw <- rootWindow dpy dflt win <- createSimpleWindow dpy rootw 0 0 200 100 0 0 0 p <- internAtom dpy "PRIMARY" True clp <- internAtom dpy "BLITZ_SEL_STRING" False xConvertSelection dpy p sTRING clp win currentTime allocaXEvent $ \e -> do nextEvent dpy e ev <- getEvent e if ev_event_type ev == selectionNotify then do res <- getWindowProperty8 dpy clp win return $ decode . fromMaybe [] $ res else destroyWindow dpy win >> return "" -- | A wrapper around getSelection. Makes it convenient to run a program with the current selection as an argument. This is convenient -- for handling URLs, in particular. For example, in your Config.hs you could bind a key to 'promptSelection "firefox"'; this would allow you to -- highlight a URL string and then immediately open it up in Firefox. promptSelection :: String -> X () promptSelection app = spawn . ((app ++ " ") ++) =<< io getSelection {- UTF-8 decoding for internal use in getSelection. This code is totally stolen from Eric Mertens's utf-string library <http://code.haskell.org/utf8-string/> (version 0.1), which fortunately is BSD-3 licensed, so I can just copy it into this BSD-3 licensed module. I guess it'd be better to just import Codec.Binary.UTF8.String (decode), but then users of this would need to install it, and Xmonad has enough dependencies as it is. -} decode :: [Word8] -> String decode [ ] = "" decode (c:cs) | c < 0x80 = chr (fromEnum c) : decode cs | c < 0xc0 = replacement_character : decode cs | c < 0xe0 = multi_byte 1 0x1f 0x80 | c < 0xf0 = multi_byte 2 0xf 0x800 | c < 0xf8 = multi_byte 3 0x7 0x10000 | c < 0xfc = multi_byte 4 0x3 0x200000 | c < 0xfe = multi_byte 5 0x1 0x4000000 | otherwise = replacement_character : decode cs where replacement_character :: Char replacement_character = '\xfffd' multi_byte :: Int -> Word8 -> Int -> [Char] multi_byte i mask overlong = aux i cs (fromEnum (c .&. mask)) where aux 0 rs acc | overlong <= acc && acc <= 0x10ffff && (acc < 0xd800 || 0xdfff < acc) && (acc < 0xfffe || 0xffff < acc) = chr acc : decode rs | otherwise = replacement_character : decode rs aux n (r:rs) acc | r .&. 0xc0 == 0x80 = aux (n-1) rs $ shiftL acc 6 .|. fromEnum (r .&. 0x3f) aux _ rs _ = replacement_character : decode rs -- gwern DDR&E E911 BCCI State EO PRIME iButton WID OAU Flintlock

3 3

darcs patch: XPrompt: a very long string in the completion list can...
by Andrea Rossato 30 Aug '07

30 Aug '07

Hi, I discovered this by evaluating [1..100] in the ghc prompt. andrea Thu Aug 30 16:15:24 CEST 2007 Andrea Rossato <andrea.rossato(a)unibz.it> * XPrompt: a very long string in the completion list can lead to a division by zero

2 1

darcs patch: XPrompt: just code formatting (and 3 more)
by Andrea Rossato 30 Aug '07

30 Aug '07

Hi, this is a set of patches to XPrompt: there is David's fix. The rest is just code formatting (new lines removal, vertical alignment), and nothing more. Andrea Wed Aug 22 21:32:20 CEST 2007 Andrea Rossato <andrea.rossato(a)unibz.it> * XPrompt: just code formatting Mon Aug 27 20:58:58 CEST 2007 David Roundy <droundy(a)darcs.net> * fix bug leading to early exit in XPrompt. Mon Aug 27 21:18:30 CEST 2007 David Roundy <droundy(a)darcs.net> * make code more compact in XPrompt. Tue Aug 28 10:12:35 CEST 2007 Andrea Rossato <andrea.rossato(a)unibz.it> * XPrompt.hs: remove debugging bits

2 1

a ghci prompt for xmonad
by Andrea Rossato 30 Aug '07

30 Aug '07

Hi, I wanted to write this prompt from the very beginning. A prompt to evaluate Haskell expressions... At first I couldn't get hs-plugins to compile. Then I could compile hs-plugin but I thought it was quite a difficult requirement for such a prompt. I knew about that mysterious ghc-api, something to come in the future... NO! GHC-API is there since 6.5, and can be easily used. So, here's the prompt. It's just the a proof of concept, right now. It uses the completions window to display the result and the ghc-api is linked from XMonad itself, which makes the xmonad binary a huge file. A lot of memory consumption is the obvious consequence. I have a small program that can evaluate expressions from outside of xmonad, but I'm not able to run it interactively from xmonad: http://www.haskell.org/pipermail/haskell-cafe/2007-August/031167.html I know how to compile Haskell code at runtime but I don't know how to run a process interactively...:( If you can help.... Follow the instruction of the attached modules (to be saved in XMonadContrib. BE CAREFUL and edit ghcPath to fit your system environment otherwise the prompt and XMonad will crash!!! (the is one of the many reasons I want an external application to run the statements!) Hope you'll enjoy. Andrea ps: no error reports for the time being. I repeat it: this is just a proof of concept. We need the external evaluator to work interactively with the prompt.

1 1

I don't notice when someone talks to me via Gaim
by Kai Grossjohann 29 Aug '07

29 Aug '07

I use Gaim for instant messaging. I use guifications so that I am notified when someone sends me a message. But this is not enough for me. I think what I want is this: - All Gaim windows go to a specific workspace. - Whenever someone talks to me via Gaim, then I get a notification about this that is unobtrusive but not transient. I am requesting the first one because sometimes I lose a chat window (in the sense of forgetting where it is). That is, I go to the buddy list, double click on a contact and nothing happens. That's because the chat window for that contact is already open. But I don't know where the chat window is. In one case, it was one of five windows in a "full" layout workspace. For the second one, I am not sure what to do. Perhaps the workspace name in the workspace list should flash? Or I should get an indicator in the status bar that a Gaim message is waiting for me? Kai

5 6

help with layout modifiers
by Andrea Rossato 29 Aug '07

29 Aug '07

Hi, this is something I'm not able to solve without some help, I think. Suppose I want to implement the mouse focus with a layout modifier: now XMobar changes the focused window when you enter it with the pointer, and this is done in Main.handle. It would be nice to do that at the layout level, with modifyLayout returning a new layout when a crossing event is generated. In order not to bother layout writers, it should be possible to implement the mouse focus as a layout modifier, which add a hook to a specific layout event handler. While this approach (code below) works in general, it doesn't work with layouts that add other X event hooks, like Tabbed, or DragPane. The problem is that, when I modify the windows stack in the event handler below, I don't have a clear idea of how the XState should be modified. I think I just need some direction...;-) To test the code below you need to comment out from Main.hs this function: handle e@(CrossingEvent {ev_window = w, ev_event_type = t}) | t == enterNotify && ev_mode e == notifyNormal && ev_detail e /= notifyInferior = focus w And this is the example code. You use it like this, in your Config.hs: defaultLayouts = [ mouseFocus $ dragUpDownPane "dd" 0.1 0.5 , mouseFocus $ tabbed shrinkText defaultTConf , mouseFocus tiled ] mouseFocus :: Layout a -> Layout a mouseFocus cl@(Layout {doLayout = dl , modifyLayout = ml}) = Layout {doLayout = doLay , modifyLayout = modLay} where doLay r s = dl r s modLay sm | Just e <- fromMessage sm = do handle_event e ml sm -- right?? | otherwise = ml sm handle_event e@(CrossingEvent {ev_window = w, ev_event_type = t}) | t == enterNotify && ev_mode e == notifyNormal && ev_detail e /= notifyInferior = do focus w -- what should I do to XState after this? return () handle_event _ = return () Thanks for your kind attention. Andrea

2 4