From kwrite-devel  Mon May 30 02:22:12 2005
From: Thomas =?utf-8?q?L=C3=BCbking?= <thomas.luebking () web ! de>
Date: Mon, 30 May 2005 02:22:12 +0000
To: kwrite-devel
Subject: Re: KDE/kdelibs/kate/part
Message-Id: <200505300422.13442.thomas.luebking () web ! de>
X-MARC-Message: https://marc.info/?l=kwrite-devel&m=111741976328773

On Sunday, 29. May 2005 22:42, you wrote:
> SVN commit 419560 by cullmann:
> ...
> if this solves it, we have some problem with editStart/End and indenters
afaics now it's no way about the indenting anyway.
i can remove the whole indenting and just keep a editStart(); editEnd(); right 
after of before the editStart() - editEnd() block around the insertion and 
heve the problem still existing.
on the other hand i can either remove the editStart() - editEnd() block around 
the indention or extend the editStart() - editEnd() block around the 
insertion to end after the indention and the problem is fixed (at this level 
- there may be more problems than just on pasting text)

the whole thing smells as if kate simply dislikes fast repetitions of (real, 
editSessionNumber == 0) editStart() - editEnd() blocks - especially as THE 
SAME LAG OCCURS IF I JUST REMOVE THE FINAL "editEnd()" (though i get some 
additional assert output, but kate behaves exactly the same)
to test this, i simply appendend "editStart(); editEnd();" to 
KateDocument::newLine(.)
-> simply typing enter knocks kate (i tested kwrite) out!

i just don't know why, according to Qt doc, signals/slots are strict linear so 
no chance for a lifelock here...

Thomas
ps: PLEASE KEEP ME IN CC
-- 
Think, think different. But essentially: Think!
_______________________________________________
KWrite-Devel mailing list
KWrite-Devel@kde.org
https://mail.kde.org/mailman/listinfo/kwrite-devel