Ethan Quach wrote: > > Tim Knitter wrote: >> >> Ethan Quach wrote: >>> Tim, >>> >>> libbe.c - 781 - needs update. >>> >> Actually that describes what is passed in to beVerifyBEName() just like >> the other functions in libbe.c since that info is hard to determine from >> the generic parameters (PyObject *self, PyObject *args). > > Okay, thats fine then. This is inconsistent with some of the > other block function comments though. >
I made all the "Parameters:" section in the function headers consistent. >>> Another thing that just occurred to me is that there are many >>> places in beadm.py that parse a beName as a commandline argument. >>> Is there any particular reason why you only added the beName verify >>> check to those two places? >>> >> Yes. All the other functions that accept beName as an argument use a >> beName that has already been created or verified. e.g. beadm destroy >> <beName> the beName has already been created so checking it after the >> fact isn't needed. "rename" and "create" are the only beNames that work >> with a freshly devised name from the user. > > Are you saying a user could never type "beadm destroy foo%%" ? > Sure they could type that but they would have never been able to create it initially since the beadm create <beName> argument is checked. However I agree that adding the check for each subcommand is more consistent and will provide a better message for the case you mentioned so I added the check to all the subcommands. WR updated. Thanks Tim > > -ethan > > _______________________________________________ > caiman-discuss mailing list > caiman-discuss at opensolaris.org > http://mail.opensolaris.org/mailman/listinfo/caiman-discuss
