On Fri, Sep 21, 2007 at 02:25:14PM -0700, David Roundy wrote:

I'm still waiting for some sort of agreement that this is a good idea, before converting XMonadContrib (which will be a chore).

I didn't study the new code, and I'm too a bit waiting for that sort of agreement you are talking about to dig into it seriously, but after that I could probably try to be of some help in converting part of the XMonadContrib stuff. Andrea

On Friday 21 September 2007 16:25:14 David Roundy wrote:

Hi all,

This is probably my last patch for the day. It implements layout selection in the class-based approach. There's a bit of hokeyness in the handling of the reading of a SomeLayout, but the code is reasonably clean (in my opinion), and it should be back up to the same feature set as xmonad 0.3, except that layout state is preserved on mod-q.

I'm still waiting for some sort of agreement that this is a good idea, before converting XMonadContrib (which will be a chore).

David

Thu Sep 20 17:01:43 EDT 2007 David Roundy * eliminate a few Eq a constraints in StackSet.

Thu Sep 20 17:05:27 EDT 2007 David Roundy * remove unneeded Ord constraint.

Thu Sep 20 17:28:43 EDT 2007 David Roundy * add (unused) Layout to StackSet.

Thu Sep 20 18:12:48 EDT 2007 David Roundy * move Layout into StackSet. WARNING! This changes the format of StackSet, and will definitely mess up your xmonad state, requiring at minimum a restart!

Fri Sep 21 16:43:16 EDT 2007 David Roundy * make layouts preserved over restart

Fri Sep 21 17:21:59 EDT 2007 David Roundy * add layout selection back into core xmonad using LayoutSelection. This is just a reimplementation of LayoutChoice.

Applied.