Horton

Re: Declarative partitioning - another take

Amit Langote <langote_amit_f8@lab.ntt.co.jp>

View thread View on postgr.es
From: Amit Langote <Langote_Amit_f8@lab.ntt.co.jp>
To: Dmitry Ivanov <d.ivanov@postgrespro.ru>
Cc: Robert Haas <robertmhaas@gmail.com>, Michael Paquier <michael.paquier@gmail.com>, Andres Freund <andres@anarazel.de>, Erik Rijkers <er@xs4all.nl>, Amit Langote <amitlangote09@gmail.com>, Rajkumar Raghuwanshi <rajkumar.raghuwanshi@enterprisedb.com>, Ashutosh Bapat <ashutosh.bapat@enterprisedb.com>, Pg Hackers <pgsql-hackers@postgresql.org>, pgsql-hackers-owner@postgresql.org
Date: 2016-12-13T06:58:59Z
Lists: pgsql-hackers

Attachments

Hi,

On 2016/12/13 2:45, Dmitry Ivanov wrote:
> Huh, this code is broken as well. We have to ignore partitions that don't
> have any subpartitions. Patch is attached below (v2).

Good catch and thanks a lot for the patch!  I have revised it a bit and
added some explanatory comments to that function.

Attaching the above patch, along with some other patches posted earlier,
and one more patch fixing another bug I found.  Patch descriptions follow:

0001-Miscallaneous-code-and-comment-improvements.patch

Fixes some obsolete comments while improving others.  Also, implements
some of Tomas Vondra's review comments.

0002-Miscallaneous-documentation-improvements.patch

Fixes inconsistencies and improves some examples in the documentation.
Also, mentions the limitation regarding row movement.

0003-Invalidate-the-parent-s-relcache-after-partition-cre.patch

Fixes a bug reported by Tomas, whereby a parent's relcache was not
invalidated after creation of a new partition using CREATE TABLE PARTITION
OF.  This resulted in tuple-routing code not being to able to find a
partition that was created by the last command in a given transaction.

0004-Fix-a-bug-of-insertion-into-an-internal-partition.patch

Fixes a bug I found this morning, whereby an internal partition's
constraint would not be enforced if it is targeted directly.  See example
below:

create table p (a int, b char) partition by range (a);
create table p1 partition of p for values from (1) to (10) partition by
list (b);
create table p1a partition of p1 for values in ('a');
insert into p1 values (0, 'a');  -- should fail, but doesn't

0005-Fix-a-tuple-routing-bug-in-multi-level-partitioned-t.patch

Fixes a bug discovered by Dmitry Ivanov, whereby wrong indexes were
assigned to the partitions of lower levels (level > 1), causing spurious
"partition not found" error as demonstrated in his email [1].


Sorry about sending some of these patches repeatedly though.

Thanks,
Amit

[1]
https://www.postgresql.org/message-id/e6c56fe9-4b87-4f64-ac6f-bc99675f3f9e%40postgrespro.ru

Commits

  1. Fix typo.

  2. Document trigger-firing behavior for inheritance/partitioning.

  3. Fire per-statement triggers on partitioned tables.

  4. Set ecxt_scantuple correctly for tuple routing.

  5. Fix interaction of partitioned tables with BulkInsertState.

  6. Avoid core dump for empty prepared statement in an aborted transaction.

  7. Fix some problems in check_new_partition_bound().

  8. Remove unnecessary arguments from partitioning functions.

  9. Fix reporting of constraint violations for table partitioning.

  10. Fix tuple routing in cases where tuple descriptors don't match.

  11. Invalid parent's relcache after CREATE TABLE .. PARTITION OF.

  12. Doc: improve documentation about inheritance.