|From:||Stephen Frost <sfrost(at)snowman(dot)net>|
|To:||Noah Misch <noah(at)leadboat(dot)com>|
|Cc:||Jose Luis Tallon <jltallon(at)adv-solutions(dot)net>, pgsql-hackers(at)postgresql(dot)org|
|Subject:||Re: pg_dump dump catalog ACLs|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
* Noah Misch (noah(at)leadboat(dot)com) wrote:
> On Tue, Apr 05, 2016 at 05:50:18PM -0400, Stephen Frost wrote:
> > @@ -14187,18 +14869,65 @@ dumpTable(Archive *fout, TableInfo *tbinfo)
> > + "FROM pg_catalog.pg_attribute at "
> > + "JOIN pg_catalog.pg_class c ON (at.attrelid = c.oid) "
> > + "LEFT JOIN pg_init_privs pip ON "
> Since pg_attribute and pg_class require schema qualification here, so does
> pg_init_privs. Likewise elsewhere in the patch's pg_dump changes.
Attached is a patch to qualify the tables used in that query. I
reviewed all of the other pg_init_privs uses and they all match the
qualification level of the other tables in those queries. This isn't
too surprising as this is the only query which happens outside of the
This patch doesn't do anything but qualify the table usage and that
query is checked by the pg_dump regression suite and ran fine for me,
so, if there are no objections, I'll push this later on this afternoon.
|Next Message||Vladimir Gordiychuk||2016-05-10 17:15:33||Re: Stopping logical replication protocol|
|Previous Message||Robert Haas||2016-05-10 16:56:17||Re: asynchronous and vectorized execution|