[immodule-qt] Re: receiving order inversion between QKeyEvent and QIMEvent (Re: scim-qtimm 0.7 released)

[LiuCougar]

I tested but I could not reproduce any crashes, it just works fine in my box.

Regards,

On Thu, 09 Sep 2004 02:02:36 +0200, Mike FABIAN <mfabian at suse.de> wrote:
> YamaKen <yamaken at bp.iij4u.or.jp> さんは書きました:
> 
> > At Wed, 08 Sep 2004 14:30:37 +0200,
> > mfabian at suse.de wrote:
> >>
> >> YamaKen <yamaken at bp.iij4u.or.jp> さんは書きました:
> >>
> >> > At Tue, 7 Sep 2004 15:08:39 +0000,
> >> > liucougar at gmail.com wrote:
> >> >>
> >> >> Thanks, looks fine to me
> >> >
> >> > Oops, I had a trivial mistake. Try new patch attached in this
> >> > mail.
> >>
> >> I tried
> >>
> >>     qt-x11-immodule-unified-qt3.3.3-20040819.diff
> >>
> >> together with the event-inversion patch
> >>
> >>     qt-x11-immodule-unified-qt3.3.3-20040819-event-inversion-20040908.diff
> >>
> >> today and encountered problems when trying to check whether
> >> M17N-vi-viqr works.
> >>
> >> I tried LiuCougar's simple test to type 'ds' with M17N-vi-viqr and
> >> check whether it comes out in the correct order, but then the client
> >> just hangs consuming 100% CPU time. I tried with edittest and
> >> scim-qtimm:
> >
> > Thank you for reporting. Would you provide me the backtrace and
> > source list around interrupted point?
> 
> backtrace attached. I used M17N-t-latin-post input method via SCIM
> and typed
> 
>     a"b
> 
> then edittest did hang and I interrupted it from gdb to get the
> backtrace.
> 
> I have also attached the source code around the interrupt point,
> it is in qevent.h in the class QEvent.
> 
> I marked the line where I interrupted it like this:
> 
>       // <--- interrupted point
> 
> 
> 
> 


-- 
"People's characters are strengthened through struggle against
difficulties; they are weakened by comfort."
- Old Chinese adage