On Fri, Nov 02, 2012 at 04:21:56PM -0500, Adam Heath wrote: > On 11/02/2012 04:19 PM, Ben Pfaff wrote: > > I still don't understand. Why would an existing script call ovs-vsctl > > with --may-exist as a global option? The call would not do anything > > useful, because it would exit with a fatal error every time. Taking > > this argument to its logical conclusion, one must never extend any > > program by adding a new option, because it could break some existing > > script that tries to use that (nonexistent) option. > > Exactly. Your patch breaks existing scripts, by suddenly allowing for > --may-exist to work where it used to not.
You still haven't explained why an existing script would call ovs-vsctl with --may-exist as a global option. Why would it do that? > I'm saying you should just change the error message, but do not allow > for it to be used as a global option. It's an option, and it's easier to do than the alternative patch that I sent out, but I don't understand why you think it will break any real script. Who is in the habit of writing useless scripts that do nothing but provoke errors? _______________________________________________ dev mailing list dev@openvswitch.org http://openvswitch.org/mailman/listinfo/dev
