Re: [POC] hash partitioning

On Fri, May 12, 2017 at 6:08 PM, amul sul <sulamul(at)gmail(dot)com> wrote:
> Hi,
>
> Please find the following updated patches attached:

I have done some testing with the new patch, most of the cases worked
as per the expectation except below

I expect the planner to select only "Seq Scan on t1" whereas it's
scanning both the partitions?

create table t (a int, b varchar) partition by hash(a);
create table t1 partition of t for values with (modulus 8, remainder 0);
create table t2 partition of t for values with (modulus 8, remainder 1);

postgres=# explain select * from t where a=8;
QUERY PLAN
----------------------------------------------------------
Append (cost=0.00..51.75 rows=12 width=36)
-> Seq Scan on t1 (cost=0.00..25.88 rows=6 width=36)
Filter: (a = 8)
-> Seq Scan on t2 (cost=0.00..25.88 rows=6 width=36)
Filter: (a = 8)
(5 rows)

Some cosmetic comments.
-----------------------------------
+ RangeVar *rv = makeRangeVarFromNameList(castNode(List, nameEl->arg));
+

Useless Hunk.

/*
- * Build a CREATE SEQUENCE command to create the sequence object, and
- * add it to the list of things to be done before this CREATE/ALTER
- * TABLE.
+ * Build a CREATE SEQUENCE command to create the sequence object, and add
+ * it to the list of things to be done before this CREATE/ALTER TABLE.
*/

Seems like, in src/backend/parser/parse_utilcmd.c, you have changed
the existing code with
pgindent. I think it's not a good idea to mix pgindent changes with your patch.

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