Horton

Re: unsupportable composite type partition keys

jobinau@gmail.com

View thread View on postgr.es
From: Jobin Augustine <jobinau@gmail.com>
To: Julien Rouhaud <rjuju123@gmail.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Amit Langote <amitlangote09@gmail.com>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2020-09-09T14:17:22Z
Lists: pgsql-hackers 
Is there a way out if someone accidentally executes the same test case
against PG12?

testdb=# create table partitioned (a int, b int)
testdb-#   partition by list ((row(a, b)::partitioned));
CREATE TABLE
testdb=# DROP TABLE partitioned;
ERROR:  cache lookup failed for type 18269


>
> Ah, I wasn't sure that additional tests on a table would be worthwhile
> enough.
> Thanks for tweaking and pushing!
>
>
>

Commits

  1. Fix CheckAttributeType's handling of collations for ranges.

  2. Improve comments in utils/rel.h.

  3. Allow whole-row Vars to be used in partitioning expressions.

  4. Remove equalPartitionDescs().

  5. Load relcache entries' partitioning data on-demand, not immediately.

  6. Disallow partition key expressions that return pseudo-types.

  7. Prevent a rowtype from being included in itself via a range.