Re: Miscellaneous tab completion issue fixes - Mailing list pgsql-hackers

From Michael Paquier
Subject Re: Miscellaneous tab completion issue fixes
Date
Msg-id Yzur8a4jJJoainiK@paquier.xyz
Whole thread Raw
In response to Re: Miscellaneous tab completion issue fixes  (Dagfinn Ilmari Mannsåker <ilmari@ilmari.org>)
Responses Re: Miscellaneous tab completion issue fixes  (vignesh C <vignesh21@gmail.com>)
List pgsql-hackers
On Mon, Oct 03, 2022 at 06:29:32PM +0100, Dagfinn Ilmari Mannsåker wrote:
> vignesh C <vignesh21@gmail.com> writes:
>> +    else if (TailMatchesCS("\\dRp*"))
>> +        COMPLETE_WITH_QUERY(Query_for_list_of_publications[0].query);
>> +    else if (TailMatchesCS("\\dRs*"))
>> +        COMPLETE_WITH_QUERY(Query_for_list_of_subscriptions[0].query);
>
> These are version-specific queries, so should be passed in their
> entirety to COMPLETE_WITH_VERSIONED_QUERY() so that psql can pick the
> right version, and avoid sending the query at all if the server is too
> old.

+1.

>> +/* add these to Query_for_list_of_roles in OWNER TO contexts */
>> +#define Keywords_for_list_of_owner_to_roles \
>> +"CURRENT_ROLE", "CURRENT_USER", "SESSION_USER"
>
> I think this would read better without the TO, both in the comment and
> the constant name, similar to the below only having GRANT without TO:

Keywords_for_list_of_grant_roles is used in six code paths, so it
seems to me that there is little gain in having a separate #define
here.  Let's just specify the list of allowed roles (RoleSpec) where
OWNER TO is parsed.
--
Michael

Attachment

pgsql-hackers by date:

Previous
From: Andres Freund
Date:
Subject: Re: [RFC] building postgres with meson - v13
Next
From: Amit Langote
Date:
Subject: Re: ExecRTCheckPerms() and many prunable partitions