Categories:

Query syntax

FROM

Specifies the tables, views, or table functions to use in a SELECT statement.

See also:

AT | BEFORE , CHANGES , CONNECT BY , JOIN , ASOF JOIN, MATCH_RECOGNIZE, PIVOT , SAMPLE / TABLESAMPLE , UNPIVOT, Working with Joins, Analyzing time-series data

Syntax

SELECT ...
FROM objectReference [ JOIN objectReference [ ... ] ]
[ ... ]
Copy

Where:

objectReference ::=
   {
      [<namespace>.]<object_name>
           [ AT | BEFORE ( <object_state> ) ]
           [ CHANGES ( <change_tracking_type> ) ]
           [ MATCH_RECOGNIZE ]
           [ PIVOT | UNPIVOT ]
           [ [ AS ] <alias_name> ]
           [ SAMPLE ]
     | <table_function>
           [ PIVOT | UNPIVOT ]
           [ [ AS ] <alias_name> ]
           [ SAMPLE ]
     | ( VALUES (...) )
           [ SAMPLE ]
     | [ LATERAL ] ( <subquery> )
           [ [ AS ] <alias_name> ]
     | @[<namespace>.]<stage_name>[/<path>]
           [ ( FILE_FORMAT => <format_name>, PATTERN => '<regex_pattern>' ) ]
           [ [AS] <alias_name> ]
     | DIRECTORY( @<stage_name> )
   }
Copy

Parameters

JOIN

Subclause that specifies to perform a join between two or more tables (or views or table functions). The join can be an inner join, outer join, or other type of join. The join can use the keyword JOIN or an alternative supported join syntax. For more details about joins, see JOIN and ASOF JOIN.

[ AS ] alias_name

Specifies a name given to the object reference it is attached to. Can be used with any of the other subclauses in the FROM clause.

Alias names must follow the rules for Object identifiers.

VALUES

The VALUES clause can specify literal values or expressions to be used in the FROM clause. This clause can contain table and column aliases (not shown in the diagram above). For more details about the VALUES clause, see VALUES.

Object or table function clause

[namespace.]object_name

Specifies the name of the object (table or view) being queried.

The object name can be qualified using namespace (in the form of db_name.schema_name.object_name or schema_name.object_name). A namespace is not required if the context can be derived from the current database and schema for the session.

When specifying a table/view name to query, you can also specify the following optional subclause:

AT | BEFORE ( object_state )

Optional subclause that specifies the time-based or event-based historical state of the table or view for Time Travel. For more details, see AT | BEFORE.

MATCH_RECOGNIZE

Optional subclause for finding sequences of rows that match a pattern. For more details, see MATCH_RECOGNIZE.

table_function

Specifies a system-defined table function, a UDF table function, or a class method to call within the FROM clause. For details, see the following topics:

PIVOT | UNPIVOT

Optional subclause that specifies to pivot or unpivot the results of the FROM clause. For more details, see PIVOT and UNPIVOT.

SAMPLE

Optional subclause that specifies to sample rows from the table/view. For more details, see SAMPLE / TABLESAMPLE.

Inline view clause

[ LATERAL ] ( subquery )

Specifies an inline view within the FROM clause. If the optional LATERAL keyword is used, then the subquery can refer to columns from other tables (or views or table functions) that are in the current FROM clause and to the left of the inline view.

For more information about subqueries in general, see Working with Subqueries.

Staged file clause

@[namespace.]stage_name[/path]

Specifies a named stage to be queried (or ~ for referring to the stage for the current user or % followed by a table name for referring to the stage for the specified table).

When querying a stage, you can also optionally specify a named file format and pattern:

( FILE_FORMAT => format_name [ , PATTERN => 'regex_pattern' ] )

Specifies a named file format object to use for the stage and a pattern to filter the set of files in the stage.

For more details about querying stages, see Querying Data in Staged Files.

Directory table clause

DIRECTORY( @<stage_name> )

Specifies the name of a stage that includes a directory table.

Hierarchical query result

hierarchical_query_result

A hierarchical query result is the result set from using a clause such as CONNECT BY to query a table of hiearchical data. For more details, see CONNECT BY.

Usage notes

  • Object names are SQL identifiers. They are case-insensitive by default. To preserve case, enclose them between double quotes (" ").

Examples

Create a table and load data into it:

CREATE TABLE ftable1 (retail_price FLOAT, wholesale_cost FLOAT, description VARCHAR);
INSERT INTO ftable1 (retail_price, wholesale_cost, description) 
  VALUES (14.00, 6.00, 'bling');
Copy

Here is a basic example of using the FROM clause:

SELECT description, retail_price, wholesale_cost 
    FROM ftable1;
+-------------+--------------+----------------+
| DESCRIPTION | RETAIL_PRICE | WHOLESALE_COST |
|-------------+--------------+----------------|
| bling       |           14 |              6 |
+-------------+--------------+----------------+
Copy

This example is identical to the previous example, but specifies the table name qualified by the schema for the table:

SELECT description, retail_price, wholesale_cost 
    FROM temporary_doc_test.ftable1;
+-------------+--------------+----------------+
| DESCRIPTION | RETAIL_PRICE | WHOLESALE_COST |
|-------------+--------------+----------------|
| bling       |           14 |              6 |
+-------------+--------------+----------------+
Copy

This example creates an inline view and then uses it in the query:

SELECT v.profit 
    FROM (SELECT retail_price - wholesale_cost AS profit FROM ftable1) AS v;
+--------+
| PROFIT |
|--------|
|      8 |
+--------+
Copy

This example queries a sample of 10% of the data in the table:

SELECT *
    FROM sales SAMPLE(10);
Copy

This example executes a UDTF (user-defined table function):

SELECT *
    FROM TABLE(Fibonacci_Sequence_UDTF(6.0::FLOAT));
Copy

These examples use an AT clause to return historical data from the following specified points in the past:

  • One day earlier than the current time (-86400 = -3600 * 24).

  • Specific time and day.

SELECT *
    FROM sales AT(OFFSET => -86400);
SELECT *
    FROM sales AT(TIMESTAMP => '2018-07-27 12:00:00'::TIMESTAMP);
Copy

For more details about AT, see AT | BEFORE.

This example queries files located in a named stage:

SELECT
    v.$1, v.$2, ...
  FROM
    @my_stage( FILE_FORMAT => 'csv_format', PATTERN => '.*my_pattern.*') v;
Copy

This example retrieves all metadata columns in a directory table for a stage named mystage:

SELECT * FROM DIRECTORY(@mystage);
Copy

This example retrieves the FILE_URL column values from a directory table for files greater than 100 K bytes in size:

SELECT FILE_URL FROM DIRECTORY(@mystage) WHERE SIZE > 100000;
Copy

This example retrieves the FILE_URL column values from a directory table for comma-separated value files:

SELECT FILE_URL FROM DIRECTORY(@mystage) WHERE RELATIVE_PATH LIKE '%.csv';
Copy
Language: English