AIX 3.1, script and intr

[Vic Abell]

In article <9008142356.AA07620 at mindcrf.mindcraft.com>, karish at mindcrf.UUCP (Chuck Karish) writes:
> In article <5306 at mace.cc.purdue.edu> abe at mace.cc.purdue.edu (Vic Abell) writes:
> >In article <9008100210.AA23730 at mindcrf.mindcraft.com>
> >karish at mindcrf.UUCP writes:
> >>In article <5289 at mace.cc.purdue.edu> abe at mace.cc.purdue.edu (Vic Abell)
> >>writes:
> >>>Does anyone know how to keep the intr character (^C in my case) from
> >>>terminating script?
> 
> The answer is the same.  Type `trap "" 2' to your interactive sh or ksh
> before invoking `script'.  If your shell is ignoring a signal, its
> child will ignore it, too.

This works.
 
> The `onintr' directive to csh doesn't seem to have an option to cause
> the shell to ignore one signal.  `onintr -', though, will cause it (and
> its child, the `script' process) to ignore ALL signals, including
> SIGINT.  This will make it more complicated to recover control of your
> display if your application hangs.


This does not work.  Interactive csh rejects onintr with the message"

	"onintr: Can't from terminal."

A solution that doesn't allow the intr character to be used during a script
session isn't very attractive, however.  The point of wanting to use intr,
especially when scripting a bug's behavior, is to be able to better control
the session.  Since other script programs I have used do handle intr
properly, it appears that IBM has modified the AIX 3.1 script in some way
that has diminished its usefulness for bug reporting.