Horton

range test for hash index?

Paul A Jungwirth <pj@illuminatedcomputing.com>

View thread View on postgr.es
From: Paul A Jungwirth <pj@illuminatedcomputing.com>
To: PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2019-09-13T19:17:59Z
Lists: pgsql-hackers 
Hello,

I noticed the tests for range types do this:

create table numrange_test2(nr numrange);
create index numrange_test2_hash_idx on numrange_test2 (nr);

Does that need a `using hash`? It seems like that's the intention. We
only use that table for equality comparisions. The script already
creates a table with a btree index further up. If I don't drop the
table I can see it's not a hash index:

regression=# \d numrange_test2
           Table "public.numrange_test2"
 Column |   Type   | Collation | Nullable | Default
--------+----------+-----------+----------+---------
 nr     | numrange |           |          |
Indexes:
    "numrange_test2_hash_idx" btree (nr)

Everything else passes if I change just that one line in the
{sql,expected} files.

Regards,
Paul

Commits

  1. Fix oversight in commit 4429f6a9e3e12bb4af6e3677fbc78cd80f160252.

  2. Support range data types.