[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

[thhsieh: IMdkit problem on OverTheSpot]



----- Forwarded message from thhsieh -----

Date: Tue, 8 Feb 2000 12:04:49 +0800
From: thhsieh
To: xcin@linux.org.tw
Subject: Re: [thhsieh: IMdkit problem on OverTheSpot]
X-Mailer: Mutt 1.0pre3us
In-Reply-To: <389F74EB.93CA070A@coppice.org>

On Tue, Feb 08, 2000 at 09:44:11AM +0800, Steve Underwood wrote:
> debugged and enhanced versions of IMdkit, but no cooperation. Effort is being
> wasted solving the same problems for multiple projects. I think there is good
> reason to extract the IMdkit from xcin, and treat is as a stand-alone library. If
> it is well publicised to the other groups using it, maybe we can all contribute to
> a single more polished library. Maybe it should even become part of the XFree86
> distribution, rather than a part of the IME distributions.

Agreed. I think we can begin the cooperation with the ami-maintainer
and the groups of TurboLinux in Mainland China, but I am not sure if
they have made any important changes or bug fixes to IMdkit.

If we want to contribute IMdkit back to XFree86 to be one of the standard,
who should we contact?

> I doubt the ami library has fixed all the problems, and I know we haven't. I made
> quite large changes when cleaning up sections of the library last summer, and
> Tung-Han has fixed a number of other things since them. The current version is much
> more stable, functional and 64-bit friendly than the original. I suspect that
> merging the good things from the ami version into the xcin version would produce
> the most polished library so far. The other versions of IMdkit I have seen seem far
> less well debugged.

I don't think ami has fixed all the problems of IMdkit, and neither do
we. I have tested it, as I said, it will cause xcin to crash when XIM
client exits. I cann't remember clearly, but it seems much like the bug
we have already fixed long time ago. Although ami has add some codes 
such as monitor_client_win_death callback to try to avoid this problem
(I guess), but this seems not a good solution to me since they call the
gtk function calls from IMdkit directly to handle this problem. 


T.H.Hsieh

----- End forwarded message -----


Reply to: