[1020] in testers
rt 7.1D: show_console
daemon@ATHENA.MIT.EDU (daemon@ATHENA.MIT.EDU)
Mon Aug 6 09:25:28 1990
From: qjb@ATHENA.MIT.EDU
Date: Mon, 6 Aug 90 09:24:56 -0400
To: probe@MIT.EDU
Cc: testers@MIT.EDU
In-Reply-To: Richard Basch's message of Sun, 5 Aug 90 17:12:12 -0400 <9008052112.AA03705@CRONOS.MIT.EDU>
your controlling tty, so either we have to force something to signal
xlogin/xconsole to re-grab the console
If the point of the program is to cause the display of that
little window that xlogin pops up and calls the "console"
window, then someone should signal Xlogin to remap that window.
I think it's really irrelevant whether the actual console output
goes there or not. It is important because it is the stdout of
the xsession file and the default place where people get writes
and talks. Since the goal is to cause a window to be remapped,
any method that makes assumptions about what kinds of external
events can cause this to happen will be wrong more often than a
program that does it directly.
I seem to remember hearing that it is legal under ICCCM to just
map a window for a client. If so, why not just scan the window
tree until you find the window with class XLogin (or whatever it
is) and name console and then map that window?
