Skip to content

Statements

The following statement forms are supported.

SELECT

Retrieve rows from zero or more relations.

[ statement UNION [ ALL ] ... ]
SELECT [ DISTINCT [ ON (<columns>) ] ] <expression> [, ..]
  FROM <relation> [AS <alias>]
   FOR <period>
       [ INNER ] JOIN <relation> | <function> | (<subquery>)
       CROSS JOIN <relation> | <function> | (<subquery>)
       LEFT [ OUTER | ANTI | SEMI ] JOIN <relation> | <function> | (<subquery>)
       RIGHT [ OUTER | ANTI | SEMI ] JOIN <relation> | <function> | (<subquery>)
       FULL [ OUTER ] JOIN <relation> | <function> | (<subquery>)
                      ON <expression>
                      USING (<columns>)
 WHERE <expression> [ AND | OR | XOR .. ]
 GROUP BY 
       HAVING <expression> [ AND | OR | XOR .. ]
 ORDER BY <expression> [, ..]
OFFSET <offset>
 LIMIT <limit>

UNION class

statement UNION [ ALL ] statement

The UNION class appends the results of two queries together one after the other. The names and types of the resulting columns are taken from the first statement, names in the second statement are ignored and types are coerced where possible.

The default behaviour of the UNION class is to deduplicate rows, to return all rows, including duplicates the ALL modifier must be used.

SELECT clause

SELECT [ DISTINCT [ ON (columns )]] expression [, ...]

The SELECT clause specifies the list of columns that will be returned by the query. While it appears first in the clause, logically the expressions here are executed after most other clauses. The SELECT clause can contain arbitrary expressions that transform the output, as well as aggregate functions.

The DISTINCT modifier is specified, only unique rows are included in the result set. In this case, each output column must be of a type that allows comparison. DISTINCT ON () will perform a distinct on the specified columns and select one value for other columns.

FROM / JOIN clauses

FROM relation [AS alias] [FOR period] [WITH (NO_CACHE, NO_PARTITION, NO_PUSH_PROJECTION, NO_PUSH_SELECTION)] [, ...] 

FROM relation [AS alias] [FOR period] [ INNER ] JOIN relation [FOR period] < USING (columns) | ON condition >

FROM relation [AS alias] [FOR period] LEFT [ OUTER | ANTI | SEMI ] JOIN relation [FOR period] < USING (columns) | ON condition >

FROM relation [AS alias] [FOR period] RIGHT [ OUTER | ANTI | SEMI ] JOIN relation [FOR period] < USING (columns) | ON condition >

FROM relation [AS alias] [FOR period] FULL [OUTER ] JOIN relation [FOR period]

FROM relation [AS alias] [FOR period] CROSS JOIN < relation [FOR period] | UNNEST(column) >

The FROM clause specifies the source of the data on which the remainder of the query should operate. Logically, the FROM clause is where the query starts execution. The FROM clause can contain a single relation, a combination of multiple relations that are joined together, or another SELECT query inside a subquery node.

JOIN clauses allow you to combine data from multiple relations. If no JOIN qualifier is provided, INNER will be used. JOIN qualifiers are mutually exclusive. ON and USING clauses are also mutually exclusive and can only be used with INNER and LEFT joins.

See Joins for more information on JOIN syntax and functionality.

FOR clause

FOR date

FOR DATES BETWEEN start AND end

FOR DATES IN range

FOR DATES SINCE start

The FOR clause is a non-standard clause which filters data by the date it was recorded for. When provided FOR clauses must directly follow the relation in a FROM or JOIN clause. If not provided FOR TODAY is assumed.

See Time Travel for more information on FOR syntax and functionality.

WHERE clause

WHERE condition

The WHERE clause specifies any filters to apply to the data. This allows you to select only a subset of the data in which you are interested. Logically the WHERE clause is applied immediately after the FROM clause.

GROUP BY / HAVING clauses

GROUP BY expression [, ...]

HAVING group_filter

The GROUP BY clause specifies which grouping columns should be used to perform any aggregations in the SELECT clause. If the GROUP BY clause is specified, the query is always an aggregate query, even if no aggregations are present in the SELECT clause. The HAVING clause specifies filters to apply to aggregated data, HAVING clauses require a GROUP BY clause.

GROUP BY expressions may use column numbers, however, this is not recommended for statements intended for reuse.

ORDER BY / LIMIT / OFFSET clauses

ORDER BY expression [ ASC | DESC ] [, ...]

OFFSET count

LIMIT count

ORDER BY, LIMIT and OFFSET are output modifiers. Logically they are applied at the very end of the query. The OFFSET clause discards initial rows from the returned set, the LIMIT clause restricts the amount of rows fetched, and the ORDER BY clause sorts the rows on the sorting criteria in either ascending or descending order.

ORDER BY expressions may use column numbers, however, this is not recommended for statements intended for reuse.

EXPLAIN

Show the logical execution plan of a statement.

EXPLAIN
statement

The EXPLAIN clause outputs a summary of the execution plan for the query in the SELECT statement.

Warning

The data returned by the EXPLAIN statement is intended for interactive usage only and the output format may change between releases. Applications should not depend on the output of the EXPLAIN statement.

EXECUTE

Execute a preprated statement.

EXECUTE statement_name[(<parameter=value[, ...]>)]

The EXECUTE clause executes a prepared statement, the parameters supplied in the invocation clause are used to populate placeholders in the prepared statement. The supplied parameters must be named, for example EXECUTE PLANETS_BY_ID (id=1).

SET

Specifies the value of a variable, the variable is available to the scope of the executing query batch.

SET variable = value

User defined variable names must be prefixed with an 'at' symbol (@) and the value must be a literal value. The variable can be used within SELECT clauses within the same query batch. A SET statement without a SELECT statement is invalid.

System parameters can also be temporarily for a query batch and are prefixed with a dollar sign ($).

Related: SHOW VARIABLES and SHOW PARAMETER

SHOW COLUMNS

List the columns in a relation along with their data type. Without any modifiers, SHOW COLUMNS only reads a single morsel of data before returning.

SHOW [EXTENDED] [FULL] COLUMNS
FROM relation
 FOR period

EXTENDED modifier

Inclusion of the EXTENDED modifier includes summary statistics about the columns which take longer and more memory to create than the standard summary information without the modifier. The summary information varies between column types and values.

FULL modifier

Inclusion of the FULL modifier uses the entire dataset in order to return complete column information, rather than just the first morsel from the dataset.

FOR clause

FOR date

FOR DATES BETWEEN start AND end

FOR DATES IN range

FOR DATES SINCE start

The FOR clause is a non-standard clause which filters data by the date it was recorded for. When provided FOR clauses must directly follow the relation name the FROM clause. If not provided FOR TODAY is assumed.

See Time Travel for more information on FOR syntax and functionality.

SHOW CREATE VIEW

Show an approximation of the SQL to create a specified relation.

SHOW CREATE VIEW view

Returns the SQL (including comments) which is executed when the view is accessed.