FreeLists / haiku-appserver / [haiku-appserver] Re: partly paiting

[Stephan Aßmus <superstippi@xxxxxx>]

Hi Adi,

> > after I updated my SVN rep and cleaned up some code here and there... when
> > I first booted into Haiku again, many views didn't paint correctly. For
> > example, menus only paint the first item when the open, the others are
> > painted when they are highlighted "on mouse over". Just to be sure, I
> > reverted my changes, but the problem persists. Is anybody else seeing 
> > this?
> > If so, has anybody some clue which change caused this? It should be a
> > change from after last Sunday.
> 
>     Hi,
> 
>     First, I'm sorry I forgot checkin in Window.cpp.
>     I have experienced this behavior myself, and the problem is, I don't
> remember not having the code I committed in Window.cpp active.
>     If I'm not mistaken, this started happening after a couple checkins
> from Stefano. I'm not sure I remember correctly, Stefano please excuse me
> if I'm wrong.
> 
>     Stephan, if you can't reproduce this anymore, and the checkin I make for
> Window.cpp solved this problem, then Stefano, please accept thousand of
> apologies from me.

No, _your_ checkin (removal of my code for B_VIEW_RESIZED) is the problem. 
:-) 

The problem that my code solved, was that B_VIEW_RESIZED was not sent for all 
views that needed it. The way I solved this was to send a single message 
which contained the tokens from all effected views. The comment you removed 
explained where this was done in Layer.cpp and what the potential problems 
were. 

>     BTW, when I resize Playground window some views do not redraw correctly,
> and I'm quite surprised nobody said anything until now... Am I the only one
> who experiences this behavior?

This behaviour only happens after your checkin for Window.cpp. So I did 
notice, that's why I wrote the initial email. :-)

>     I am now working to solve some issues with the new clipping code. I have
> quite an annoying bug, and it seams it takes a lot more time than expected.
>     Also, I will shortly update the way we mark regions invalid,
> RootLayer::fRedrawReg will be made private and I will introduce 
> ::MarkDirty() +
> ::TriggerRedraw() to simplify the code.

Oh, this will be greatly appreciated!! I think it has been quite obscure for 
everybody involved how to exactly (reliably) trigger a redraw.

>     I started refactoring the input code, and I will continue to do that
> in quite a few places. It's a lot easier now that we have the almost final
> architecture in place. :-)

Yes, thank you very much for your work! After your changes, I think 
app_server will be easier to improve further.

Best regards,
-Stephan