- xmonad - Haskell.org

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

darcs patch: fix bug leading to early exit in XPrompt.
by David Roundy 29 Aug '07

29 Aug '07

I think perhaps this bug only manifests itself when working with tabbed. David Mon Aug 27 14:58:58 EDT 2007 David Roundy <droundy(a)darcs.net> * fix bug leading to early exit in XPrompt.

3 6

darcs patch: make code more compact in XPrompt.
by David Roundy 28 Aug '07

28 Aug '07

This should definitely not be applied unless it's approved by Andrea. It's a side-effect of my debugging attempt which resulted in the previously-sent bug fix for XPrompt. I compactified the code so I could read it (since I can't read what I can't see, so a function that doesn't fit on the screen is pretty hard to debug. But obviously some people like large quantities of superfluous space... David Mon Aug 27 14:59:23 EDT 2007 David Roundy <droundy(a)darcs.net> * make code more compact in XPrompt.

2 3

'rules' like in dwm
by Karl. 28 Aug '07

28 Aug '07

Hi, I'm currently using dwm, and considering switching to Xmonad. I've been trying to figure out whether Xmonad supports the features that are important to me. The one I haven't been able to figure out for certain is: Does Xmonad support defining rules as to which workspace a new window will appear in? Under dwm I use specific workspaces for specific things - my web browser always goes on the same workspace, regardless of where I am when I launch it. I have some rules defined in my dwm config that specify this. This consistency is one of the essential things that I like about dwm. Another approach, instead of rules defining workspace based on client properties, would be if it is possible for a keybinding to specify where a client should go when it is launched eg. Ctrl-Alt-B = launch browser and make it go to workspace 'web' As far as I can tell so far, under Xmonad, new windows will just appear in the current workspace. thanks, Karl.

3 2

Issue 45 in xmonad: xmonad should be smarter about mouse movement
by codesite-noreply＠google.com 27 Aug '07

27 Aug '07

Issue 45: xmonad should be smarter about mouse movement http://code.google.com/p/xmonad/issues/detail?id=45 Comment #7 by SpencerJanssen: Turns out to be an unclutter issue -- and there is a workaround "unclutter -grab". Issue attribute updates: Status: Invalid -- 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 0