Re: TRUNCATE on foreign tables

Hello,

The attached patch is the revised version of TRUNCATE on foreign tables.

Definition of the callback is revised as follows:

typedef void (*ExecForeignTruncate_function) (List *frels_list,
bool is_cascade,
bool restart_seqs);

The "frels_list" is a list of foreign tables that are connected to a particular
foreign server, thus, the server-id pulled out by foreign tables id should be
identical for all the relations in the list.
Due to the API design, this callback shall be invoked for each foreign server
involved in the TRUNCATE command, not per table basis.

The 2nd and 3rd arguments also informs FDW driver other options of the
command. If FDW has a concept of "cascaded truncate" or "restart sequence",
it can adjust its remote query. In postgres_fdw, it follows the manner of
usual TRUNCATE command.

Best regards,

2020年1月8日(水) 1:08 Kohei KaiGai <kaigai(at)heterodb(dot)com>:
>
> 2020年1月7日(火) 16:03 Michael Paquier <michael(at)paquier(dot)xyz>:
> >
> > On Mon, Jan 06, 2020 at 04:32:39PM -0500, Stephen Frost wrote:
> > > RESTRICT, yes. I don't know about ONLY being sensible as we don't
> > > really deal with inheritance and foreign tables very cleanly today, as I
> > > said up-thread, so I'm not sure if we really want to put in the effort
> > > that it'd require to figure out how to make ONLY make sense.
> >
> > True enough.
> >
> > > The question about how to handle IDENTITY is a good one. I suppose
> > > we could just pass that down and let the FDW sort it out..?
> >
> > Looking at the code, ExecuteTruncateGuts() passes down restart_seqs,
> > so it sounds sensible to me to just pass down that to the FDW
> > callback and the callback decide what to do.
> >
> It looks to me the local sequences owned by a foreign table shall be restarted
> by the core, regardless of relkind of the owner relation. So, even if FDW driver
> is buggy, consistency of the local database is kept, right?
> Indeed, "restart_seqs" flag is needed to propagate the behavior, however,
> it shall be processed on the remote side via the secondary "TRUNCATE" command.
> Is it so sensitive?
>
> Best regards,
> --
> HeteroDB, Inc / The PG-Strom Project
> KaiGai Kohei <kaigai(at)heterodb(dot)com>

--
HeteroDB, Inc / The PG-Strom Project
KaiGai Kohei <kaigai(at)heterodb(dot)com>