- xmonad - Haskell.org

Issue 133 in xmonad: xmonad-contrib source code "cleanup" - in search of a user code location
by codesite-noreply＠google.com 02 Feb '08

02 Feb '08

Issue 133: xmonad-contrib source code "cleanup" - in search of a user code location http://code.google.com/p/xmonad/issues/detail?id=133 Comment #2 by dons00: How about interfaces get some docs describing their purpose, and point to where to find specific instances? Can you make these tickets more focused? I'm finding them vague and rambling, so they're difficult to address or close. -- 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 125 in xmonad: Tabbed layout not working anymore
by codesite-noreply＠google.com 02 Feb '08

02 Feb '08

Issue 125: Tabbed layout not working anymore http://code.google.com/p/xmonad/issues/detail?id=125 Comment #18 by andrea.rossato: since I read, from the main developer and xmonad author: "I think this solution is *really* ugly. Also, I don't think we can necessarily rely on the property that decorations windows are the only override-redirect windows in the windowset." I just replay that I have no replay but to ask to revert to the previous system, since my patch is the logic consequence of the design chosen, which probably requires such *really* ugly stuff. I'm sorry for the inconvenience. Also because I have no other ideas on how to solve this issue, if not by reverting to the back version. ps: the link error where just my stupidity. Since windowremover came after windownavigator, in the cabal file I accidentally renamed the first. Thanks, Spencer, for pointing that out. Issue attribute updates: Owner: SpencerJanssen -- 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

please revert back
by Andrea Rossato 02 Feb '08

02 Feb '08

Hi, I'm feeling in a quite difficult situation. I have pushed a huge set of patches with the assumption they were being approved by the main developers and now, instead of a supportive environment I feel like my code is being rejected . it seems to work, but its just a big ugly hack. I'm really far from arguing that this is false. As I said I'm a legal scholar doing legal, economic and philosophical research. I do not really have the competence for having a contest with the main developer, Spencer. Moreover in the silence of some the other developers and a mixed approach but Brent, each one far more expert then me on the subject. Which leads me to think that it is me who's being wrong, and, since I have no (scientific) interest in avoiding admitting my faults - my reputation here is far from being at stake - I have no problem to acknowledged I did a mistake by abusing the trust that has been given to me, by granting me write access to the contributory repository. While I can assure you that I did all that with plenty of good faith, truly (even if only subjectively) thinking to improve xmonad, you may also understand that I'm now not in a position to joyfully maintain my code, if I have to argue. Because I'm not in the position of feeling I'm i the right position. I hope you understand that there is no harsh in my words, but just the acknowledgment that a computer scientist has more knowledge than me, and that for me it would just be wasted pretending that's not true. So please revert. We are at the very beginning and people are distracted by the release, so the change would go almost unnoticed. At a latter time, after deeper code analysis, you may consider to slowly introduce some of the changes I proposed, if you find them not too hackish. I really really hope you'll appreciate the constructive intent of this message, and the fact the its intention is supportive of this community. Cheers, Andrea

2 2

Issue 125 in xmonad: Tabbed layout not working anymore
by codesite-noreply＠google.com 01 Feb '08

01 Feb '08

Issue 125: Tabbed layout not working anymore http://code.google.com/p/xmonad/issues/detail?id=125 Comment #16 by andrea.rossato: > The only thing that remains, from a user point of view, is the > ability to click on the tabs (decorations) to switch to the window. this requires a change in the way we handle messages. Basically, a layout modifier is not allowed to call "focus w" (the function to place a window in focus), because that modifies the stack the layout modifier is operating with. As I wrote here: http://www.haskell.org/pipermail/xmonad/2008-January/004398.html I do not really grasp the hows and whys, but I'm sure it's true. I think this issue is the same of issue #111. (but I remember there should be a similar issue, but I cannot find it presently). > One question: shouldn't 'windowRemove' be used inside 'tabbed' in > 'XMonad.Layout.Tabbed'. So that users don't have to remember to apply it to 'tabbed'? yes, you are right. But I didn't because of that linker issue. I thought that first we must test the patch, and possibly understand if it breaks ghc-6.6.1 compatibility, that would be quite an issue. Cheers, Andrea I'm accepting this bug. See: http://www.haskell.org/pipermail/xmonad/2008-February/004533.html Issue attribute updates: Status: Accepted Owner: andrea.rossato -- 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

3 2

Issue 133 in xmonad: xmonad-contrib source code "cleanup" - in search of a user code location
by codesite-noreply＠google.com 01 Feb '08

01 Feb '08

Issue 133: xmonad-contrib source code "cleanup" - in search of a user code location http://code.google.com/p/xmonad/issues/detail?id=133 Comment #1 by andrea.rossato: this stupid defaults..;) I did not accept this patch. Also because it needs review. changed to new. Issue attribute updates: Status: New Owner: --- -- 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

Re: [xmonad] Moving windows to a workspace based on their title / console applications in terminal emulators
by lobzang＠free.fr 01 Feb '08

01 Feb '08

Hey ! Indeed there is a bug. I did a simple test : 1)with urxvt urxvt -name icq --> xprop says WM_CLASS(STRING) = "icq", "URxvt" --> xmonad puts the window matching URxvt pattern , not icq 2)with xterm xterm -class icq --> xprop says WM_CLASS(STRING) = "xterm", "icq" --> the window is placed in my icq workspace. Again xmonad seems to take only the second class string to match So for now you can probably use xterm... Lobzang. Quoting Bastian Schnitzler <baschni(a)gmail.com>: > Hi! > I did: > > [ className =? "Icq" --> doF (W.shift "2") ], > [ className =? "irssiterm" --> doF (W.shift "2") ], > [ className =? "muttterm" --> doF (W.shift "2") ], > > and still urxvt -name Icq -title Icq stays on the same workspace, same for > other classnames. > Does it work for you? > > Bastian > > On 09:06 Thu 31 Jan , lobzang(a)free.fr wrote: > > Hi, > > > > urxvt -name icq -e centerim > > urxvt -name mail -e mutt > > > > should do the trick ( assuming you update your xmonad window rules : > > , className =? "icq" --> doF (W.shift "icq" ) > > , className =? "mail" --> doF (W.shift "mail" ) > > > > Lobzang > > > > Quoting Bastian Schnitzler <baschni(a)gmail.com>: > > > > > Hi all! > > > For many purposes I am using console applications, like centerim, > > > irssi or mutt. Now when I start xmonad, I'd like to open a bunch of > > > applications (Can I do this via .xinitrc), like firefox, mutt, etc. > > > These should then be moved to a workspace, based on the application > > > running. So when I start an rxvt-Terminal with mutt, I want this to be > > > moved to workspace 3, how can I do this? On top of that I can't figure > > > out how to move gimp to a special workspace. I have setten it to > > > float, but I tried the class names, gimp, Gimp, gimp-2.2 etc. but it > > > won't go to the workspace I want, when I start it. Anyway what is this > > > class name to identify a window? Is it the process name? > > > Thanks in advance, > > > Bastian Schnitzler > > > > > > > > >

4 4

Issue 133 in xmonad: xmonad-contrib source code "cleanup" - in search of a user code location
by codesite-noreply＠google.com 01 Feb '08

01 Feb '08

Issue 133: xmonad-contrib source code "cleanup" - in search of a user code location http://code.google.com/p/xmonad/issues/detail?id=133 New issue report by andrea.rossato: I think we should find a better way to separate modules for user interfaces, and modules for developers APIs. For instance, in the IRC channel there were complains about the fact the XMonad.Layout.Decoration did not have user documentation and a user could not figure how to turn decorations on. The fact is that Decoration is just an API, and you can find user code in Tabbed. SimpleDecoration and DwmStyle are combinators, and I do not really judge the user code, but APIs. Since it seems we can get to a quite expressive level of layers composition, I suggest to elaborate a strategy for placing user code in a single place. On a longer period of time, I hope that the class system and the related instances (even Decoration - with (this is necessary :( the needed XUtils and Font libraries) will be proven safe enough to be included in the core, that would just export an abstract layer that can provide the full functionality of a highly programmable WM *without* any sensitive Xlib call. This hope guided my last hacking efforts. Issue attributes: Status: Accepted Owner: andrea.rossato Labels: Type-Defect 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

Issue 131 in xmonad: We need a mouse interface to the decoration layer
by codesite-noreply＠google.com 01 Feb '08

01 Feb '08

Issue 131: We need a mouse interface to the decoration layer http://code.google.com/p/xmonad/issues/detail?id=131 Comment #1 by andrea.rossato: I'm starting to think, and I would like some feed back on that, that the separation between arranger/decoration is a bad design choice, since a messaging system is probably impossible. I must confess I had/have the (probably unjustified) hope that by solving #111 we would have a safe messaging system where a modifier can send a message to another modifier (or the underlying layout) and have this one return Just (with a Operations.windows call), without entering a recursive non terminating loop: like a safe "focus w" call. A second dream was this: have the first modifier send a message, have the second respond with a send message to the first while returning Nothing. - Decoration to arranger: send me the arranger state. - Arranger to decoration: here's my state >> return Nothing and then decoration set rectangles and attributes. But I think this is not working too. The idea of the separation came from the thrill of an arranger implemented with pure code and the arrow interface ;). But if the arranger cannot modify itself without returning just, mine are just dreams, and the design is broken. The fact that the design is broken can be proved this way: if you put the arranger over the decoration and you start moving, you'll see that the decoration is one step behind. Joining them would probably make the code ugly: Decoration *cannot* be pure. I hope someone can come up with a better strategy and give me some direction. -- 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 132 in xmonad: We should have a convention for layout names
by codesite-noreply＠google.com 01 Feb '08

01 Feb '08

Issue 132: We should have a convention for layout names http://code.google.com/p/xmonad/issues/detail?id=132 New issue report by andrea.rossato: I would suggest to set some conventional rules on out to export layouts to our users. I think we should export a simple one, without any configuration option, and one, with a "'", for a more complex one where some configuration parameters are available, and so on of higher levels of configurability. For instance: tabbed tabbed' shrinker theme tabbedResizable tabbedResizable' shrinker theme and so on. I'm just opening this issue as a way to gather ideas and then choose what seems the best. Issue attributes: Status: New Owner: ---- 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

Issue 125 in xmonad: Tabbed layout not working anymore
by codesite-noreply＠google.com 01 Feb '08

01 Feb '08

Issue 125: Tabbed layout not working anymore http://code.google.com/p/xmonad/issues/detail?id=125 Comment #17 by andrea.rossato: I came to the conclusion that WindowRemover should be mover to XMonad.Layout.Tabbed, and not exported. It should just be used for that kind of special type of layou tabbed it. This way we have an implementation of both that specific idea of a tabbed layout where tabs represent not visible windows, and the one where a tabbed layout is a stack of windows with visible lables. The user will chose one, and will know that, while the second is composable, the first may be not (which means it also can: for instance, I believe that verticalReflect should work with both) cheers ps did you ever try applying Reflect to tabbed? reflectHorizontal is just messy, but vertical will place tabs at the bottom. If you use the WindowArranger, keep in mind that vertical actions must be reversed. -- 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