Date: Tue, 3 Apr 2018 10:26:45 -0400
From: Bruce Momjian <bruce@momjian.us>
To: scott.ure@caseware.com, pgsql-docs@lists.postgresql.org
Subject: Re: float and float(p) missing from table 8.1
Message-ID: <20180403142645.GA6472@momjian.us>
References: <152139310065.11961.13080741495660494967@wrigleys.postgresql.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <152139310065.11961.13080741495660494967@wrigleys.postgresql.org>
User-Agent: Mutt/1.5.23 (2014-03-12)
Precedence: bulk

On Sun, Mar 18, 2018 at 05:11:40PM +0000, PG Doc comments form wrote:
> The following documentation comment has been logged on the website:
> 
> Page: https://www.postgresql.org/docs/10/static/datatype.html
> Description:
> 
> Maybe it is intentional, but the float and float(p) data types are missing
> from table 8.1
> (https://www.postgresql.org/docs/10/static/datatype.html#DATATYPE-TABLE).
> float4 and float8 are listed.

Yes, I think it is intentional and is caused by the multiple aliases we
have for numeric values.  As you probably noted from the chart, the
primary type name is 'real', which is ANSI specified, with an alias
listed as float4.  There is also double precision, with an alias of
float8.

float() is more complicated since it can be either real/float4 or double
precision/float8 depending on the length, e.g.:

	CREATE TABLE test(x FLOAT(1), x2 FLOAT(25));
	
	\d test
	                    Table "public.test"
	 Column |       Type       | Collation | Nullable | Default
	--------+------------------+-----------+----------+---------
	 x      | real             |           |          |
	 x2     | double precision |           |          |

The switch from float4/float8 happens at binary digit length 25, and
this is documented:

	https://www.postgresql.org/docs/10/static/datatype-numeric.html#DATATYPE-FLOAT

	PostgreSQL also supports the SQL-standard notations float and float(p)
	for specifying inexact numeric types. Here, p specifies the minimum
	acceptable precision in binary digits. PostgreSQL accepts float(1) to
	float(24) as selecting the real type, while float(25) to float(53)
	select double precision. Values of p outside the allowed range draw an
	error. float with no precision specified is taken to mean double
	precision.


Adding float() to that chart seems like it would add too much complexity.

-- 
  Bruce Momjian  <bruce@momjian.us>        http://momjian.us
  EnterpriseDB                             http://enterprisedb.com

+ As you are, so once was I.  As I am, so you will be. +
+                      Ancient Roman grave inscription +