map_keys

Description

Returns an array of all the keys in the specified map.

Syntax

map_keys(any_map)

Parameters

any_map: the MAP value from which you want to retrieve keys.

Return value

The return value is in the format of array<keyType>. The element type in the array matches the key type in the map.

If the input is NULL, NULL is returned. If a key or value in the MAP value is NULL, NULL is processed as a normal value and contained in the result.

Examples

Query MAP data from a CelerData native table

CelerData supports defining MAP columns when you create a table. This example uses table test_map, which contains the following data:

CREATE TABLE test_map(
    col_int INT,
    col_map MAP<VARCHAR(50),INT>
  )
DUPLICATE KEY(col_int);

INSERT INTO test_map VALUES
(1,map{"a":1,"b":2}),
(2,map{"c":3}),
(3,map{"d":4,"e":5});

SELECT * FROM test_map ORDER BY col_int;
+---------+---------------+
| col_int | col_map       |
+---------+---------------+
|       1 | {"a":1,"b":2} |
|       2 | {"c":3}       |
|       3 | {"d":4,"e":5} |
+---------+---------------+
3 rows in set (0.05 sec)

Obtain all the keys from each row of the col_map column.

select map_keys(col_map) from test_map order by col_int;
+-------------------+
| map_keys(col_map) |
+-------------------+
| ["a","b"]         |
| ["c"]             |
| ["d","e"]         |
+-------------------+
3 rows in set (0.05 sec)

Query MAP data from data lake

This example uses Hive table hive_map, which contains the following data:

SELECT * FROM hive_map ORDER BY col_int;
+---------+---------------+
| col_int | col_map       |
+---------+---------------+
|       1 | {"a":1,"b":2} |
|       2 | {"c":3}       |
|       3 | {"d":4,"e":5} |
+---------+---------------+

After a Hive catalog is created in your CelerData cloud account, you can use this catalog and the map_keys() function to obtain all the keys from each row of the col_map column.

select map_keys(col_map) from hive_map order by col_int;
+-------------------+
| map_keys(col_map) |
+-------------------+
| ["a","b"]         |
| ["c"]             |
| ["d","e"]         |
+-------------------+
3 rows in set (0.05 sec)
In this article
Copyright © 2024 CelerData