Text Size: Normal / Large

Home → Documentation → Manuals → PostgreSQL 7.3

This page in other versions: 9.2 / 9.1 / 9.0 / 8.4 | Unsupported versions: 9.3 / 8.3 / 8.2 / 8.1 / 8.0 / 7.4 / 7.3 / 7.2 / 7.1 / devel

PostgreSQL 7.3.21 Documentation
Prev	Chapter 6. Functions and Operators	Next

6.14. Aggregate Functions

Aggregate functions compute a single result value from a set of input values. Table 6-33 show the built-in aggregate functions. The special syntax considerations for aggregate functions are explained in Section 1.2.5. Consult the PostgreSQL 7.3.21 Tutorial for additional introductory information.

Table 6-33. Aggregate Functions

Function	Argument Type	Return Type	Description
`avg(expression)`	`smallint`, `integer`, `bigint`, `real`, `double precision`, `numeric`, or `interval`.	`numeric` for any integer type argument, `double precision` for a floating-point argument, otherwise the same as the argument data type	the average (arithmetic mean) of all input values
`count(*)`		`bigint`	number of input values
`count(expression)`	any	`bigint`	number of input values for which the value of `expression` is not null
`max(expression)`	any numeric, string, or date/time type	same as argument type	maximum value of `expression` across all input values
`min(expression)`	any numeric, string, or date/time type	same as argument type	minimum value of `expression` across all input values
`stddev(expression)`	`smallint`, `integer`, `bigint`, `real`, `double precision`, or `numeric`.	`double precision` for floating-point arguments, otherwise `numeric`.	sample standard deviation of the input values
`sum(expression)`	`smallint`, `integer`, `bigint`, `real`, `double precision`, `numeric`, or `interval`	`bigint` for `smallint` or `integer` arguments, `numeric` for `bigint` arguments, `double precision` for floating-point arguments, otherwise the same as the argument data type	sum of `expression` across all input values
`variance`(`expression`)	`smallint`, `integer`, `bigint`, `real`, `double precision`, or `numeric`.	`double precision` for floating-point arguments, otherwise `numeric`.	sample variance of the input values (square of the sample standard deviation)

It should be noted that except for count, these functions return a null value when no rows are selected. In particular, sum of no rows returns null, not zero as one might expect. The function coalesce may be used to substitute zero for null when necessary.

Prev	Home	Next
Miscellaneous Functions	Up	Subquery Expressions

Comments

Jan. 6, 2003, 6:27 p.m.

The aggregates do not make use of indexes. If I understand correctly, they do a seq scan even if an index is available. Thus, on tables larger that a few thousand rows they take forever to complete.

For production use all queries with MIN() and MAX() should be rewritten with SELECT ... ORDER BY ... LIMIT 1.

Sept. 9, 2003, 5:10 p.m.

The lack of support for indexing with aggregates has been discussed ad nauseum on the mailing lists and is in the FAQ. Suffice it to say that our inability to use indexes for MIN() and MAX() is due to the ectensibility of PostgreSQL; it is the price we pay for supporting custom aggregates and data types.

Fixes for MIN() and MAX() have been proposed for PostgreSQL 7.5, but the implementation is far more difficult than it would appear to the layman.

Nov. 24, 2004, 4:45 p.m.

Here\'s how I\'ve been able to do a \"conditional count\":

Define a table:
create table cities (id serial not null unique, title varchar, state varchar);

Assume that you want a count of all the cities, as well as a count of those just in California:

select count(*) as total, sum(case when state=\'CA\' then 1 else 0 end) as ca_total from cities;

No need to group by - you\'ll get a count of all the cities, as well as just those in California!

I think that case bypasses indexes, so this might not be a good performer.