On Fri, Feb 10, 2017 at 1:52 PM, Michael Paquier
<michael.paquier@gmail.com> wrote:
> On Tue, Feb 7, 2017 at 6:35 AM, Peter Eisentraut
> <peter.eisentraut@2ndquadrant.com> wrote:
>> On 2/6/17 10:54 AM, Fujii Masao wrote:
>>> Yes, that's an option. And, if we add dbid to pg_stat_subscription,
>>> I'm tempted to add all the pg_subscription's columns except subconninfo
>>> into pg_stat_subscription. Since subconninfo may contain security-sensitive
>>> information, it should be excluded. But it seems useful to expose other
>>> columns. Then, if we expose all the columns except subconninfo, maybe
>>> it's better to just revoke subconninfo column on pg_subscription instead of
>>> all columns. Thought?
>>
>> I think previous practice is to provide a view such as pg_subscriptions
>> that contains all the nonprivileged information.
>
> OK, I think that I see the point you are coming at:
> pg_stat_get_subscription (or stat tables) should not be used for
> psql's tab completion. So gathering all things discussed, we have:
> - No tab completion for publications.
No, the tab-completions for ALTER/DROP PUBLICATION should show the local
publications because those commands drop and alter the local ones. OTOH,
CREATE/ALTER SUBSCRIPTOIN ... PUBLICATION should show nothing because
the remote publications in the publisher side should be specified there.
> - Fix the meta-commands.
Yes.
> - Addition of a view pg_subscriptions with all the non-sensitive data.
> (- Or really extend pg_stat_subscriptions with the database ID and use
> it for tab completion?)
Probably I failed to get Peter's point... Anyway IMO that we can expose all the
columns except the sensitive information (i.e., subconninfo field)
in pg_subscription to even non-superusers. Then we can use pg_subscription
for the tab-completion for ALTER/DROP SUBSCRIPTION.
Regards,
--
Fujii Masao