+1 - didn't think of STRICT but yeah, that makes sense to me :-)

regards,

Karl
On Tue, Oct 2, 2018 at 10:06 AM Bertrand Delacretaz
<[email protected]> wrote:
>
> Hi,
>
> On Tue, Oct 2, 2018 at 7:46 AM Karl Pauls <[email protected]> wrote:
> > Can’t we stay BC and just introduce a new command that has the new behavior
> > and keep the old one as is?...
>
> Great idea, this is much simpler to implement and manage indeed, at
> the cost of making the language slightly more complicated.
>
> I think having to differentiate between service and non-service users
> when deleting is somewhat of an edge case, which a STRICT modifier can
> address - so how about:
>
>   DELETE STRICT USER -> delete user only if it's NOT a service user
>   DELETE STRICT SERVICE USER -> delete user only if it's a service user
>   DELETE USER -> delete both service and non-service user
>   DELETE SERVICE USER -> same but deprecated, indicated by an INFO log
>
> In all cases there's no failure if the target user doesn't exist, as
> in the current version.
>
> -Bertrand (who's so happy to see Shared Neurons in action ;-)



-- 
Karl Pauls
[email protected]

Reply via email to