Path: utzoo!mnetor!tmsoft!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!usc!snorkelwacker.mit.edu!bloom-beacon!ANDREAS.WR.USGS.GOV!gjw
From: gjw@ANDREAS.WR.USGS.GOV (Gregory J. Woodhouse)
Newsgroups: comp.windows.x
Subject: Re:  Parent window
Message-ID: <9012210104.AA25485@andreas.wr.usgs.gov>
Date: 21 Dec 90 01:04:51 GMT
Sender: daemon@athena.mit.edu (Mr Background)
Organization: The Internet
Lines: 34

>> From within my client program I would like to be able to pop my
>> 'parent window' forward.  (By 'parent window' I mean the xterm window
>> that the user used to start my program).
>
>First, there may not be any such window.  The program may have been
>started from the user's .xinitrc or .xsession, by a window manager menu
>entry, via rsh, from a dumb terminal, from a non-xterm terminal window,
>etc, etc.
>
>Second, you may not be able to in any case; the window manager can deny
>the request.
>
>Third, even if the window exists and it is successfully uniconized and
>brought forward, it may not be visible even then (think tvtwm with the
>"parent window" off in some other part of the desktop).
>
>> This would allow me to use stderr and stdout for writing error
>> messages.

Why not use pipe or socketpair in conjunction with dup ro force said
output to be written onto a pipe?  Then you could have a program which
reads the "new stderr" and when it sees data it pops a window using
XMapRaised and writes into the window.  Maybe it could even grab the
pointer and wait for a ButtonPress before unmapping the diagnostic
window.

Just an idea.
-----------------------------------------------------------------------------
Gregory Woodhouse          |We know that the center of the earth
gjw@andreas.wr.usgs.gov    |is a fiery molten mass...but it's not
(415) 329-4694 (office)    |good to dwell on it.
(415) 325-7802 (home)      |
U.S. Geological Survey / 345 Middlefield Rd. MS 977 / Menlo Park, CA 94025
-----------------------------------------------------------------------------