On Thu, Aug 19, 1999 at 10:29:56AM -0400, I wrote:
> Looking at this sequence, Some tools could be improved:
>
> pdbtool:
> - add a user _name_ to a _named_ group. The best solution would
> probably be to `generalize' the processing of id-numbers, by allowing
> names to be entered in places we currently expect an id. Whenever the
> id is not a numeric value, assume it is a name, and perform a lookup.
That was easy :), I've implemented, tested and checked a new version of
the pdbtool.c file into the CVS which accepts both ids and names for the
following functions:
'i id/name' - read info associated with id/name from the database.
'n id/name' - (same as above, left it for compatibility with existing
scripts)
'd id/name' - delete a user or group from the database.
'ag groupid/name id/name' - Add user/group to a group.
'rg groupid/name id/name' - Remove user/group from a group.
> au:
> - Accept on the commandline:
> au nu <adminname> <username> <userpassword> [<userinfo>]
>
> And only query for the administrator's password. This way the
> password doesn't need to be stored in the script.
>
> createvol_rep:
> - Accept a dumpday on the commandline. If dayname is None, no backups,
> and when not specified query like we do now.
Anyone interested in fixing these up?
Jan
