Randy J. Ray wrote:
>
> >
> > I'm somewhat late into this thread, but I think that the two primitives
> > should be:
> >
> > Shell makes a subshell
> >
> > Exec forks a subshell with an implicit 'exec', or even just fork
> > a process with no shell at all. The fvwm command should
> > mirror the behavior of the shell command by that name,
> > because that's what people will expect it to do.
> >
> > 3 is overkill, but one is not enough.
> >
>
> This would be disastrous, as it would break existing rc files that use Exec
> expecting it to spawn a shell as it has done thus far. *If* a new keyword is
> added, it has to reference the new functionality, not break the existing
> functionality.
>
> Randy
SO WHAT! This is the reason that 2.0.x is labelled BETA. Breaking people's
configurations is not sufficient reason to prevent a design change. It
should be announced prominently, but people using beta software can be
reasonably expected to make small (or even large) changes to their
configuration files in the interest of having better quality in a RELEASE.
Stig
--
Visit the official FVWM web page at <URL:http://www.hpc.uh.edu/fvwm/>.
To unsubscribe from the list, send "unsubscribe fvwm" in the body of a
message to majordomo_at_hpc.uh.edu.
To report problems, send mail to fvwm-owner_at_hpc.uh.edu.
Received on Thu Apr 17 1997 - 12:50:26 BST