Message-ID: <27b8149fad9bdf3efd96c7a69d91898e2734f98a.camel@cybertec.at>
Subject: Re: Documentation on Hash and btree Indexes on jsonb
From: Laurenz Albe <laurenz.albe@cybertec.at>
To: holger.vornholt@deutschebahn.com, pgsql-docs@lists.postgresql.org
Date: Mon, 15 Feb 2021 13:53:47 +0100
In-Reply-To: <161320903985.3454.7956608521954674568@wrigleys.postgresql.org>
References: <161320903985.3454.7956608521954674568@wrigleys.postgresql.org>
Content-Type: text/plain; charset="UTF-8"
User-Agent: Evolution 3.36.5 (3.36.5-2.fc32) 
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Precedence: bulk

On Sat, 2021-02-13 at 09:37 +0000, PG Doc comments form wrote:
> The documentation on https://www.postgresql.org/docs/12/datatype-json.html
> claims: 
> "jsonb also supports btree and hash indexes. These are usually useful only
> if it's important to check equality of complete JSON documents."
> 
> According to some older information from 2016 
>    
> https://bitnine.net/blog-postgresql/postgresql-internals-jsonb-type-and-its-indexes/?ckattempt=1
>     https://stackoverflow.com/questions/36075918/postgresql-index-on-json
> it seems to be a valid option to use hash or btree indexes in that case
> however.
> "As a result, in the choice of an index for JSONB type, if the index
> searches a few predetermined attributes, BTREE is the best choice"
> 
> The quoted sentance from the postgres documentation would suggest, that this
> option could be discarded. I would love to hear your feedback on this
> matter.

The PostgreSQL documentation is talking about a B-tree index on the complete
JSON, while the article is talking about a B-tree index on an expression that
extracts one attribute of the JSON.

These are quite different things, so both are correct.

Yours,
Laurenz Albe
-- 
Cybertec | https://www.cybertec-postgresql.com