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

quail kkc and asynchronous processes



I am having some problems with the interaction between quail kkc
methods and asynchronous functions which update the current buffer.
To give a more concrete example:

We have an elisp program here that handles an internal chat system.
Here is an example of what the end of the buffer which handles this
looks:

----------------------------------------------------------------------
billings(chat.wyep)<jonathan billings>[10-Nov-2000, 15:28:12]::
I've been told that xmps can use media player DLLs to play media
player content, but I've never done it myself.
andrzej(chat.wyep)<Andrzej Filinski>[10-Nov-2000, 15:28:18]::
http://www.winehq.com/download.html claims to have RH5 versions.
andrzej(chat.wyep)<Andrzej Filinski>[10-Nov-2000, 15:28:42]::
i've tried that and gave up
andrzej(chat.wyep)<Andrzej Filinski>[10-Nov-2000, 15:28:54]::
(that = xmps)
llaird@ANDREW(cslounge:211.bboards@ANDREW)<l-laird@xxxxxxx>[10-Nov-2000, 15:29:19]::
My housemate asks me whether it's possible to rig something
so that anything posted to a bboard also goes to his inbox.
he says he learned how to do it in 211.
do you know what he's talking about?
md5i<<<
This is a message I am typing which will be sent to user md5i.
----------------------------------------------------------------------

Incoming messages are inserted above the line with the <<< marker in
it asynchronously by a process-filter.  It basically does a 
(save-excursion
   (move-to-correct-location)
   (insert message))

However, if I am am typing in Japanese, and a message comes in while
doing kana to kanji conversion, strange things happen, and part of my
incoming message gets deleted.

Is there any reason the `save-excursion' call would not be sufficient
for protecting the asynchronous insertion from the quail conversion?
Are there any suggested workarounds?

-- 
Michael Duggan
(md5i@xxxxxxxxxx)