4.3 rc1 tasks widget crashiness, multi-head update

Since we keep getting flooded with the bug reports about this issue, I figured I should try and stem the flow with a blog entry: in rc1 there's a crash in the tasks widget that was fixed a few hours after 4.3 rc1 was officially tagged for the last time. The backtraces usually end up in methods called manualSortingRequest or removeGroup, but somewhere in libtaskmanager. It only impacts people who have panel layouts where the tasks widget may resize and it is triggered by windows coming and going in just the "right" numbers. In any case, it's fixed (probably before most of your got your hands on it) and I hope we get an rc2 out the door pretty shortly here.

I have to say it's pretty impressive how many of you are testing the betas and rc's for 4.3 and reporting bugs. That's very cool.

Speaking of testing, some people stepped up to test the multi-head changes I made the other day. The distinction of first-back-with-results goes to "DarkMetatron" on irc. Yesterday he came with:

[00:16] Hi, I read your blogpost about multihead. I use this setup and the bug is a total showstopper. How can I help?

I told him what he could do (install kdelibs and kdebase from svn, try it out and let me know how it goes) and today he came back with:

[16:19] Hi.. I am running kdelibs and kdebase from svn now and so far it works just as it should. Your changes work

We discussed it a bit more than that, of course, but there's the important bit. If I can get more testing of this feature over the next month or two, I will backport it to 4.3.1 or 4.3.2 depending on when they come out and how much testing this new code gets.

I also have a blog drafted that I'll probably publish tomorrow after I go through it once more about what exactly I'm planning on doing now that I've transitioned out of the KDE e.V. board. Until then: congrats to the new board members and it sounds like Akademy and GCDS is doing quite well. :) For those of you there, please keep us all informed!