Categories:: Query syntax

UNPIVOT¶

Rotates a table by transforming columns into rows. UNPIVOT is a relational operator that accepts two columns (from a table or subquery), along with a list of columns, and generates a row for each column specified in the list. In a query, it is specified in the FROM clause after the table name or subquery.

UNPIVOT is not exactly the reverse of PIVOT because it cannot undo aggregations made by PIVOT.

This operator can be used to transform a wide table (e.g. empid, jan_sales, feb_sales, mar_sales) into a narrower table (e.g. empid, month, sales).

See also:: PIVOT

Syntax¶

SELECT ...
FROM ...
    UNPIVOT [ { INCLUDE | EXCLUDE } NULLS ]
      ( <value_column>
        FOR <name_column> IN (
          <col> [ [ AS ] <col_alias> ],
          ...
        )
      )

[ ... ]

Copy

Parameters¶

{ INCLUDE | EXCLUDE } NULLS

Specifies whether to include or exclude rows with NULLs in the name_column:

INCLUDE NULLS includes rows with NULLs.
EXCLUDE NULLS excludes rows with NULLs.

Default: EXCLUDE NULLS

value_column

The name to assign to the generated column that will be populated with the values from the columns in the column list.

name_column

The name to assign to the generated column that will be populated with the names of the columns in the column list.

column_list

The names of the columns in the source table or subquery that will be rotated into a single pivot column. The column names will populate name_column, and the column values will populate value_column.

The column_list can only contain literal column names, not a subquery.

The columns in column_list must have exactly the same data type, with the following exceptions:

The data types for text strings can be different lengths.
If the columns contain text strings, different columns can use different data types for text. For example, the list can include a VARCHAR column and a CHAR column.

[ AS ] col_alias

Specifies the column alias to use in the result of the UNPIVOT operation instead of the original column names. You can’t use different aliases for the same column name. However, you can’t use the same alias for multiple column names. The AS keyword is optional.

Examples¶

Create a table, monthly_sales, with the following structure and data:

CREATE OR REPLACE TABLE monthly_sales(
  empid INT,
  dept TEXT,
  jan INT,
  feb INT,
  mar INT,
  apr INT);

INSERT INTO monthly_sales VALUES
  (1, 'electronics', 100, 200, 300, 100),
  (2, 'clothes', 100, 300, 150, 200),
  (3, 'cars', 200, 400, 100, 50),
  (4, 'appliances', 100, NULL, 100, 50);

SELECT * FROM monthly_sales;

Copy

+-------+-------------+-----+------+------+-----+
| EMPID | DEPT        | JAN | FEB  | MAR  | APR |
|-------+-------------+-----+------+------+-----|
|     1 | electronics | 100 | 200  | 300  | 100 |
|     2 | clothes     | 100 | 300  | 150  | 200 |
|     3 | cars        | 200 | 400  | 100  |  50 |
|     4 | appliances  | 100 | NULL | 100  |  50 |
+-------+-------------+-----+------+------+-----+

Unpivot the individual month columns to return a single sales value by month for each employee.

SELECT *
  FROM monthly_sales
    UNPIVOT (sales FOR month IN (jan, feb, mar, apr))
  ORDER BY empid;

Copy

+-------+-------------+-------+-------+
| EMPID | DEPT        | MONTH | SALES |
|-------+-------------+-------+-------|
|     1 | electronics | JAN   |   100 |
|     1 | electronics | FEB   |   200 |
|     1 | electronics | MAR   |   300 |
|     1 | electronics | APR   |   100 |
|     2 | clothes     | JAN   |   100 |
|     2 | clothes     | FEB   |   300 |
|     2 | clothes     | MAR   |   150 |
|     2 | clothes     | APR   |   200 |
|     3 | cars        | JAN   |   200 |
|     3 | cars        | FEB   |   400 |
|     3 | cars        | MAR   |   100 |
|     3 | cars        | APR   |    50 |
|     4 | appliances  | JAN   |   100 |
|     4 | appliances  | MAR   |   100 |
|     4 | appliances  | APR   |    50 |
+-------+-------------+-------+-------+

The following example is the same as the previous example, but it uses aliases for the column names:

SELECT *
  FROM monthly_sales
    UNPIVOT (sales FOR month IN (
      jan AS january,
      feb AS february,
      mar AS march,
      apr AS april)
    )
  ORDER BY empid;

Copy

+-------+-------------+----------+-------+
| EMPID | DEPT        | MONTH    | SALES |
|-------+-------------+----------+-------|
|     1 | electronics | JANUARY  |   100 |
|     1 | electronics | FEBRUARY |   200 |
|     1 | electronics | MARCH    |   300 |
|     1 | electronics | APRIL    |   100 |
|     2 | clothes     | JANUARY  |   100 |
|     2 | clothes     | FEBRUARY |   300 |
|     2 | clothes     | MARCH    |   150 |
|     2 | clothes     | APRIL    |   200 |
|     3 | cars        | JANUARY  |   200 |
|     3 | cars        | FEBRUARY |   400 |
|     3 | cars        | MARCH    |   100 |
|     3 | cars        | APRIL    |    50 |
|     4 | appliances  | JANUARY  |   100 |
|     4 | appliances  | MARCH    |   100 |
|     4 | appliances  | APRIL    |    50 |
+-------+-------------+----------+-------+

The previous SELECT statements exclude NULLs by default. So, they don’t include a row for appliances in February in the results. To include NULLs in the results, run the following SQL statement:

SELECT *
  FROM monthly_sales
    UNPIVOT INCLUDE NULLS (sales FOR month IN (jan, feb, mar, apr))
  ORDER BY empid;

Copy

+-------+-------------+-------+-------+
| EMPID | DEPT        | MONTH | SALES |
|-------+-------------+-------+-------|
|     1 | electronics | JAN   |   100 |
|     1 | electronics | FEB   |   200 |
|     1 | electronics | MAR   |   300 |
|     1 | electronics | APR   |   100 |
|     2 | clothes     | JAN   |   100 |
|     2 | clothes     | FEB   |   300 |
|     2 | clothes     | MAR   |   150 |
|     2 | clothes     | APR   |   200 |
|     3 | cars        | JAN   |   200 |
|     3 | cars        | FEB   |   400 |
|     3 | cars        | MAR   |   100 |
|     3 | cars        | APR   |    50 |
|     4 | appliances  | JAN   |   100 |
|     4 | appliances  | FEB   |  NULL |
|     4 | appliances  | MAR   |   100 |
|     4 | appliances  | APR   |    50 |
+-------+-------------+-------+-------+

This output includes a row for appliances in February.

Instead of selecting all columns with *, you can include specific columns in the SELECT list and reference the UNPIVOT value_column and name_column. The following example is similar to the previous example, but it specifies the value_column sales and the name_column month in the SELECT list. The query excludes the empid column:

SELECT dept, month, sales
  FROM monthly_sales
    UNPIVOT INCLUDE NULLS (sales FOR month IN (jan, feb, mar, apr))
  ORDER BY dept;

Copy

+-------------+-------+-------+
| DEPT        | MONTH | SALES |
|-------------+-------+-------|
| appliances  | JAN   |   100 |
| appliances  | FEB   |  NULL |
| appliances  | MAR   |   100 |
| appliances  | APR   |    50 |
| cars        | JAN   |   200 |
| cars        | FEB   |   400 |
| cars        | MAR   |   100 |
| cars        | APR   |    50 |
| clothes     | JAN   |   100 |
| clothes     | FEB   |   300 |
| clothes     | MAR   |   150 |
| clothes     | APR   |   200 |
| electronics | JAN   |   100 |
| electronics | FEB   |   200 |
| electronics | MAR   |   300 |
| electronics | APR   |   100 |
+-------------+-------+-------+