>>>>> "Bart" == Bart Schaefer <schaefer@brasslantern.com> writes:

    Bart> The bug is that the SIGINT should -not- have caused ode to
    Bart> exit. Either ode should have exec'd zsh without forking, so
    Bart> the second zsh inherits PID 11, or it should be catching or
    Bart> ignoring tty-generated signals, or it should have been taken
    Bart> out of the tty process group.

I tried setting SHELL to /bin/csh and run ode, then send INT to it.
It terminated with message 'signal 2'.  This can be avoided if zsh
reset itself as the process group leader.  Should this be the case?

    Bart> I'm a little confused as to why that last didn't happen
    Bart> automatically.  I would have thought that zsh 12 would force
    Bart> jadfal to be the tty group leader, thus preventing the
    Bart> signals from reaching ode.  That certainly seems to happen
    Bart> on my linux machine (using "perl -e 'system zsh'" in place
    Bart> of ode, and "sleep 10" in place of jadfal).  So the bug may
    Bart> be more insidious than one of interrupt handling in ode.

    Bart> You didn't say what operating system you're using.  It's
    Bart> possible that zsh is not doing tty pgrp management
    Bart> correctly.  Do you happen to have the config.h file that was
    Bart> generated when zsh was built?

I am using Digital UNIX 3.2c.  Attached is the config.h generated.