Re: [dwm] xmonad: dwm clone in Haskell

From: Stalwart <stlwrt_AT_gmail.com>
Date: Tue, 3 Apr 2007 20:32:39 +0300

If you find good readable, idiotproof XCB doc with examples - please
mail link here

On 4/3/07, Donald Bruce Stewart <dons_AT_cse.unsw.edu.au> wrote:
> arg:
> > On Mon, Apr 02, 2007 at 09:09:30PM -0700, usrucb_AT_gmail.com wrote:
> > > Looks like they're shooting for <400 lines of code:
> > >
> > > http://lennartkolmodin.blogspot.com/2007/04/xmonad.html
> > > http://xmonad.org/
>
> Hey,
>
> (I'm one of the xmonad/dwm-in-haskell devs).
>
> > I think it's very cool that people try to clone dwm in different
> > PLs. However, I doubt it is achievable in lesser than 400 LOC
> > without hiding a lot of stuff into a separate package/lib namely
> > X11-extras - which is part of dwm itself ;)
>
> There's only a little logic in there. Mostly its just bindings from C to
> Haskell, same with the X11 package.
>
> I think the main lines-of-code savings have been two things, I'd say:
>
> * data structures
> - lists are a lot simpler to manipulate in Haskell than in C
>
> * higher order functions
> - whenever we do the same thing twice, we abstract it out as
> a combinator, and use that instead.
>
> > As a side-note: I believe that a higher level library than Xlib
> > might be a great invention - something between widget libs and
> > Xlib itself, which can be used for drawing and which does work
> > with threads instead of a state machine... Such a lib would
> > allow to finish simple X apps in a couple of hours and to hide
> > all braindamage which has evolved during ICCCM and now EWMH
> > times...
>
> We've been considering binding to XCB instead of Xlib -- which we hope
> will simplify things, as well as making threading actually work. We'll
> see if Spencer (other xmonad dev's) summer of code project gets
> accepted.
>
> Cheers (and gratitude for inspiration!)
> Don
>
>
Received on Tue Apr 03 2007 - 19:32:45 UTC

This archive was generated by hypermail 2.2.0 : Sun Jul 13 2008 - 14:40:20 UTC