On Sat, Sep 15, 2007 at 12:09:53PM +0900, Osamu Aoki wrote:
>
> As I see for SCIM:
>
> XIM_PROGRAM=/usr/bin/scim
> XIM_ARGS="-d"
>
> With these, the following is actually executed
> # execute XIM_PROGRAM
> [ -n "$XIM_PROGRAM" -a -x "$XIM_PROGRAM" ] && eval "$XIM_PROGRAM $XIM_ARGS &"
> ||
> true
>
> Changing SCIM setting to something like
> XIM_PROGRAM="if [ -z \"$(pidof scim-launcher|xargs -n1 ps --no-headers -o
> user,pid --pid 2>/dev/null|sed -n \"s/^${USER} *\([1-9][0-9]*\)$/\1/p\")\" ]
> ; then /usr/bin/scim -d >& /dev/null ; fi"
Eww, that's one ugly line. But I suppose it's worthwhile if bugs are
fixed.
> XIM=ARGS=""
(I suppose that's XIM_ARGS with a typo.) Why "-d" is not needed here?
I am not sure "scim -d" and "scim -d &" are functionally equivalent.
> Maybe, I need to reassign this to SCIM. I will test it :-)
Osamu, I'm rather ignorant in the im-switch business, is this related to
the scim bug #431294?
Ming
2007.09.14
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
