Horton

Re: Building infrastructure for B-Tree deduplication that recognizes when opclass equality is also equivalence

Robert Haas <robertmhaas@gmail.com>

View thread View on postgr.es
From: Robert Haas <robertmhaas@gmail.com>
To: Anastasia Lubennikova <a.lubennikova@postgrespro.ru>
Cc: Antonin Houska <ah@cybertec.at>, Peter Geoghegan <pg@bowt.ie>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Date: 2019-12-28T23:56:41Z
Lists: pgsql-hackers 
On Tue, Dec 24, 2019 at 7:29 AM Anastasia Lubennikova
<a.lubennikova@postgrespro.ru> wrote:
> We can make this 'opcisbitwise' parameter enum (or char) instead of
> boolean to mark
> "always bitwise", "never bitwise" and "maybe bitwise". Though, I doubt
> if it will be helpful in any real use case.

What would be the difference between "never bitwise" and "maybe
bitwise" in that scheme?

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Commits

  1. Add equalimage B-Tree support functions.

  2. Make _bt_keep_natts_fast() use datum_image_eq().

  3. Teach datum_image_eq() about cstring datums.

  4. Fix optimization of foreign-key on update actions

  5. Support all SQL:2011 options for window frame clauses.

  6. Create a "sort support" interface API for faster sorting.