[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kmail-devel
Subject:    Re: Bug#27707: kmail crash: delete in a special constellation
From:       Don Sanders <sanders () kde ! org>
Date:       2001-06-28 12:59:56
[Download RAW message or body]

On Wednesday 27 June 2001 14:37, Marc Mutz wrote:
> On Wednesday 27 June 2001 05:05, Don Sanders wrote:
> > On Tuesday 26 June 2001 12:02, Marc Mutz wrote:
> > > On Monday 25 June 2001 23:47, Achim Bohnet wrote:
> > > <snip>
> > >
> > > > Delete first message (has no followups) in folder
> > > > listing, then delete second message (2 followups,
> > > > not expanded) --> crash.
> > >
> > > <snip>
> > >
> > > Reproducable. Reason:
> > I'm willing to look at it but I can't reproduce it.
> > Which threading mode is being used?
>
> "Open threads containing unread/new"
> sort by Date A (descending date)
> The first message in the folder is then by one "Jules
> Bean".

Could someone please send me a copy of the folder to help 
me reproduce the bug.

>
> Still reproducable with KMail CVS from 1/2 hour ago.
>
> %&#$! This seems to be a race condition!?
> If I apply this patch:
>
> I'm no longer able to reproduce the problem, if I revert
> it, the crash is back.

Ok.

> BTW: why are
> KMHeaderItem::{setOpen,text,setup,pixmap,paintCell}() not
> declared virtual?

I believe the base class declares them virtual so it is not 
necessary to redeclare them virtual. 

I'm indifferent to whether they are redeclared virtual or 
not.

> BTW2: What's this supposed to do?
>      const char req_col = (char)column;
>      if(mKey.isEmpty() || mKey.left(1) != &req_col) {

Good question. Well spotted.

> Aren't we relying on *(&req_col+1) == \0 here?

No comment.

> Shouldn't this rather be:
>
> mKey[0] != req_col

Yes, that looks right to me. Thanks. I committed the fix to 
cvs.

Don.

_______________________________________________
Kmail Developers mailing list
Kmail@master.kde.org
http://master.kde.org/mailman/listinfo/kmail

[prev in list] [next in list] [prev in thread] [next in thread]