Re: row filtering for logical replication

Dilip Kumar <dilipbalaut@gmail.com>

From: Dilip Kumar <dilipbalaut@gmail.com>
To: Greg Nancarrow <gregn4422@gmail.com>
Cc: Peter Smith <smithpb2250@gmail.com>, "tanghy.fnst@fujitsu.com" <tanghy.fnst@fujitsu.com>, Ajin Cherian <itsajin@gmail.com>, "houzj.fnst@fujitsu.com" <houzj.fnst@fujitsu.com>, Euler Taveira <euler@eulerto.com>, Rahila Syed <rahilasyed90@gmail.com>, Peter Eisentraut <peter.eisentraut@enterprisedb.com>, Önder Kalacı <onderkalaci@gmail.com>, japin <japinli@hotmail.com>, Michael Paquier <michael@paquier.xyz>, David Steele <david@pgmasters.net>, Craig Ringer <craig@2ndquadrant.com>, Amit Kapila <amit.kapila16@gmail.com>, Amit Langote <amitlangote09@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2021-11-22T14:27:50Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →
  1. Release cache tuple when no longer needed

  2. Add some additional tests for row filters in logical replication.

  3. Fix one of the tests introduced in commit 52e4f0cd47.

  4. Allow specifying row filters for logical replication of tables.

  5. Move scanint8() to numutils.c

  6. Replace Test::More plans with done_testing

  7. Reduce relcache access in WAL sender streaming logical changes

  8. Small cleanups related to PUBLICATION framework code

  9. Add a view to show the stats of subscription workers.

  10. Allow publishing the tables of schema.

  11. Doc: improve documentation of CREATE/ALTER SUBSCRIPTION.

  12. Add PublicationTable and PublicationRelInfo structs

  13. Remove unused argument "txn" in maybe_send_schema().

  14. Add prepare API support for streaming transactions in logical replication.

  15. Unify PostgresNode's new() and get_new_node() methods

  16. Use l*_node() family of functions where appropriate

  17. Add support for prepared transactions to built-in logical replication.

  18. Restore the portal-level snapshot after procedure COMMIT/ROLLBACK.

  19. Rename a parse node to be more general

  20. Remove unused column atttypmod from initial tablesync query

  21. SEARCH and CYCLE clauses

On Mon, Nov 22, 2021 at 7:14 AM Greg Nancarrow <gregn4422@gmail.com> wrote:
>
> On Thu, Nov 18, 2021 at 12:33 PM Peter Smith <smithpb2250@gmail.com> wrote:
> >
> > PSA new set of v40* patches.
> >

I have a few more comments on 0007,


@@ -783,9 +887,28 @@ pgoutput_row_filter(PGOutputData *data, Relation
relation, HeapTuple oldtuple, H
             ExecDropSingleTupleTableSlot(entry->scantuple);
             entry->scantuple = NULL;
         }
+        if (entry->old_tuple != NULL)
+        {
+            ExecDropSingleTupleTableSlot(entry->old_tuple);
+            entry->old_tuple = NULL;
+        }
+        if (entry->new_tuple != NULL)
+        {
+            ExecDropSingleTupleTableSlot(entry->new_tuple);
+            entry->new_tuple = NULL;
+        }
+        if (entry->tmp_new_tuple != NULL)
+        {
+            ExecDropSingleTupleTableSlot(entry->tmp_new_tuple);
+            entry->tmp_new_tuple = NULL;
+        }

in pgoutput_row_filter, we are dropping the slots if there are some
old slots in the RelationSyncEntry.  But then I noticed that in
rel_sync_cache_relation_cb(), also we are doing that but only for the
scantuple slot.  So IMHO, rel_sync_cache_relation_cb(), is only place
setting entry->rowfilter_valid to false; so why not drop all the slot
that time only and in pgoutput_row_filter(), you can just put an
assert?

2.
+static bool
+pgoutput_row_filter_virtual(Relation relation, TupleTableSlot *slot,
RelationSyncEntry *entry)
+{
+    EState       *estate;
+    ExprContext *ecxt;


pgoutput_row_filter_virtual and pgoutput_row_filter are exactly same
except, ExecStoreHeapTuple(), so why not just put one check based on
whether a slot is passed or not, instead of making complete duplicate
copy of the function.

3.
         oldctx = MemoryContextSwitchTo(CacheMemoryContext);
         tupdesc = CreateTupleDescCopy(tupdesc);
         entry->scantuple = MakeSingleTupleTableSlot(tupdesc, &TTSOpsHeapTuple);

Why do we need to copy the tupledesc? do we think that we need to have
this slot even if we close the relation, if so can you add the
comments explaining why we are making a copy here.



-- 
Regards,
Dilip Kumar
EnterpriseDB: http://www.enterprisedb.com