- xmonad - Haskell.org

Issue 17 in project xmonad
by codesite-noreply＠google.com 01 Jun '07

01 Jun '07

Issue 17: User-definable X monad state type http://code.google.com/p/xmonad/issues/detail?id=17 New issue report by dons00: Let config users pass arbitrary state around. Extends X with an addition parameter to be stored in a StateT (that's simplest I think). Then users can happily thread their own state. By default, type XUser = (). Issue attributes: Status: Accepted Owner: dons00 Labels: Type-Enhancement Priority-Medium -- 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

darcs patch: Add TODO about per-config state.
by glasser＠mit.edu 01 Jun '07

01 Jun '07

Thu May 31 23:20:28 EDT 2007 glasser(a)mit.edu * Add TODO about per-config state.

1 0

why no refresh on focus events?
by David Glasser 01 Jun '07

01 Jun '07

I'm working on an "accordion" layout in which the layout depends on which window is focused. (Basically: the focused window takes up most of the screen; the windows before it in the list are strips on the top, and the windows after it are strips on the bottom.) It mostly works, but only when you use focus changes that go through the "windows" function; when a window is focused by mouse-over, nothing happens. This is because of the patch quoted below, which doesn't refresh when you call "focus". I'm wondering what the problem caused by that refresh was, and if there's another way to fix it? (Also, the current doLayout signature is pretty restrictive. It would be nice if it was told the workspace ID of the workspace it is laying out; that way it can get the windowset from the X state and, say, find out what the top window of the workspace being laid out is; W.peek just gives you the focused window of the entire StackSet.) --dave Mon May 28 09:31:27 EDT 2007 Don Stewart <dons(a)cse.unsw.edu.au> * don't refresh on focus events leads to a race. this will affect how gaps are redrawn when moving to a new screen with the mouse. diff -rN -u old-xmonad/Operations.hs new-xmonad/Operations.hs --- old-xmonad/Operations.hs 2007-05-31 18:22:36.000000000 -0400 +++ new-xmonad/Operations.hs 2007-05-31 18:22:36.000000000 -0400 @@ -220,7 +220,7 @@ -- the mouse to a new screen). focus :: Window -> X () focus w = withWorkspace $ \s -> do - if W.member w s then modify (\st -> st { windowset = W.focusWindow w s }) >> refresh + if W.member w s then modify (\st -> st { windowset = W.focusWindow w s }) -- >> refresh else whenX (isRoot w) $ setFocusX w -- we could refresh here, moving gap too. -- XXX a focus change could be caused by switching workspaces in xinerama. -- if so, and the gap is in use, the gap should probably follow the -- David Glasser | glasser(a)mit.edu | http://www.davidglasser.net/

2 1

dynamic configuration? (e.g. hsplugins)
by David Roundy 01 Jun '07

01 Jun '07

I know others here have far more experience in this arena than I do, but are there any plans to support hsplugins or the ghc api, or just plain running under ghci, so that a recompile-reinstall cycle won't be necesary to update one's configuration? I'd assume this'd be in the works for version 0.7 or so, but thought I might as well ask. -- David Roundy Department of Physics Oregon State University

3 5

darcs patch: ignore numlock/capslock on mouse bindings
by Jason Creighton 01 Jun '07

01 Jun '07

Thu May 31 19:51:37 MDT 2007 Jason Creighton <jcreigh(a)gmail.com> * ignore numlock/capslock on mouse bindings

1 0

layout combinators
by Joe Thornber 01 Jun '07

01 Jun '07

I'm not going to get time to look at this again until the weekend, so I thought I'd post what I had for people to muse over. Combining splitters rather than rectangles is a neat idea (I hope). However I think we do need a combinator that merges rectangles back together again. Consider a layout consisting of three columns, where the middle column is twice the width of the out ones. To do this I'd like to create four columns and then merge together the two middle columns: eg, merge [1..2] . hcat . replicate 4 $ unit I really don't like passing indexes to merge, so if anyone can come up with a nicer way I'd be very interested. For now, this approach ignores the whole issue of scaling factors given by the user to get the window sizes exactly as they want. But I do wonder if that can be done as a seperate distortion pass over the resultant rects anyway. - Joe -- -- Layout combinators -- type Splitter = Rect -> [Rect] unit :: Splitter unit = (: []) catDir :: Direction -> [Splitter] -> Splitter catDir dir splitters = concat . zipWith ($) splitters . divideMany (length splitters) dir hcat, vcat :: [Splitter] -> Splitter hcat splitters = catDir East splitters vcat splitters = catDir South splitters hjoin, vjoin :: Splitter -> Splitter -> Splitter vjoin left right = vcat [left, right] hjoin left right = hcat [left, right] mkLayout :: (Int -> Splitter) -> Layout mkLayout s = Layout { doLayout = l, modifyLayout = const Nothing} where l sc ws = return . zip ws . map fromRect . s (length ws) . toRect $ sc -- -- Layouts -- ejtTall :: Layout ejtTall = mkLayout tall' where tall' 1 = unit tall' n = unit `hjoin` (vcat . replicate (n - 1) $ unit) ejtWide :: Layout ejtWide = mkLayout wide' where wide' 1 = unit wide' n = unit `vjoin` (hcat . replicate (n - 1) $ unit) ejtColumn :: Layout ejtColumn = mkLayout $ \_ -> hcat . replicate 4 $ (unit 'vjoin' unit)

2 1

darcs patch: Rename withWorkspace to withWindowSet.
by glasser＠mit.edu 01 Jun '07

01 Jun '07

Thu May 31 20:13:25 EDT 2007 glasser(a)mit.edu * Rename withWorkspace to withWindowSet.

2 1

a patch for 6.4 users
by Daniel Wagner 31 May '07

31 May '07

Hello, I copied GHC 6.6's instance of Read, and thought I'd save other 6.4 users from having to dig around for it. ~d New patches: [add a Read instance for Map daniel(a)wagner-home.com**20070531210704 This adds an instance of Read for Maps. If you have GHC 6.4, see the README for how to enable it; otherwise, this shouldn't affect you. ] { hunk ./README 56 + +Note: If you are using GHC 6.4, you will need to add a line to StackSet.hs. At + the end of the list of imports, add + + import ReadMap () + + to provide GHC with an instance of Read for Data.Map. addfile ./ReadMap.hs hunk ./ReadMap.hs 1 +module ReadMap where + +-- stolen from GHC 6.6 sources + +import Data.Map +import Text.Read +import GHC.Read + +instance (Ord k, Read k, Read e) => Read (Map k e) where + readPrec = parens $ prec 10 $ do + Ident "fromList" <- lexP + xs <- readPrec + return (fromList xs) + + readListPrec = readListPrecDefault } Context: [Revert accidental change to border color Spencer Janssen <sjanssen(a)cse.unl.edu>**20070531145509] [comments on why fullscreen tiling doesn't work with `implicit' floating Don Stewart <dons(a)cse.unsw.edu.au>**20070531090537] [clean up mouse code a bit Don Stewart <dons(a)cse.unsw.edu.au>**20070531085308] [first shot at a floating layer Jason Creighton <jcreigh(a)gmail.com>**20070531044733 This is a first attempting at a floating layer: mod-button1: move window mod-button2: swapMaster mod-button3: resize window mod-t: make floating window tiled again Moving or resizing a window automatically makes it floating. Known issues: Hard to manage stacking order. You can promote a window to move it to the top, (which you can do with mod-button2) but it should be easier than that. Moving a window by dragging it to a different Xinerama screen does not move it to that workspace. Code is ugly. ] [remove LOC cap (but still print count after tests) Jason Creighton <jcreigh(a)gmail.com>**20070531043417] [TAG 0.2 Spencer Janssen <sjanssen(a)cse.unl.edu>**20070531010004] Patch bundle hash: a81ee5c5006c3eaf508e8b9ea2610874ece3ee19

1 0

Issue 16 in project xmonad
by codesite-noreply＠google.com 31 May '07

31 May '07

Issue 16: xmonad does not mark non-visible windows as iconafied http://code.google.com/p/xmonad/issues/detail?id=16 Comment #1 by dons00: The issue here is that an external client is calling X behind our back. If we properly denote hidden windows as hidden to X (via the iconified property?) then clients will have a better chance to know what is visible on the screen. There'd still be issues with fullscreen mode, but setting iconfified (if that's the right thing to do) on 'Operations.hide' seems easy enough. Issue attribute updates: Status: Accepted -- 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

Issue 15 in project xmonad
by codesite-noreply＠google.com 31 May '07

31 May '07

Issue 15: Enter one-line summary http://code.google.com/p/xmonad/issues/detail?id=15 Comment #1 by dons00: (No comment was entered for this change.) 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