On Feb 10, 2008, at 14:30, John Koren wrote:
> Jeremy
>
> No it is still a problem for me. Let me give you more detail.
> First of all, the problem occurs only if the wrapper script created by
> Platypus is placed into the Finder's toolbar.
>
> Here are two simple shell scripts starting xterm. (My real script has
> directory path set so the xterm opens in the directory I want).
>
> 1)
> #! /bin/sh
> # start_xterm1.sh
>
> /usr/X11/bin/xterm &
>
> # end start_xterm1.sh
>
> 2)
> #! /bin/sh
> # start_xterm2.sh
>
> /usr/X11/bin/xterm & # call 1
>
> #*** bring X11 to the front ***
> sleep 1
> /usr/bin/osascript -e 'tell application "/Applications/Utilities/
> X11.app" to
> activate' # call 2
>
> # end start_xterm2.sh
>
> The first script (start_xterm1.sh), after being wrapped in Platypus
> and
> placed in the Finder's tool bar, will start xterm but the Finder
> window will
> stay in the foreground. In detail, xterm opens on the top but as the
> script
> exits it jumps back behind the Finder window.
>
> The second script will keep the xterm on the top. 'sleep 1' is needed
> because without it sometimes two xterms would open because of the race
> condition between call 1 and call 2 on the initial launch of X11.
>
> I should also make a correction to my earlier statement that 'open -
> a X11'
> would also bring X11 to the front. In fact it does not but the
> osascript
> does.
>
> What seems to be happening is that when xterm is launched it does
> does not
> let the OS know that it belongs with X11 and therefore after the
> script
> start_xterm1.sh terminates Finder takes over again. One needs to
> activate
> X11 explicitly.
>
> Cheers -John
>
> PS. Jeremy, I can send you the corresponding Platypus wrappers if
> needed.
>
>
>
>
> On Feb 10, 2008 2:07 AM, Jeremy Huddleston
> <
email@hidden> wrote:
>
>>> No you are not the only one. I have open -a X11 in my bundled
>>> scripts mainly
>>> to bring X11 to the front since some X11 windows (e.g. rxvt) stay
>>> behind
>>> Finder windows when launched from Finder.
>>
>> That should not be the case as of 2.1.2 (maybe even 2.1.1, but I'm
>> not
>> sure about that)