UNION Set Operator

The UNION set operator combines the result sets of two separate query expressions. The result set of each query must have the same number of columns and compatible data types. UNION automatically removes duplicate records from the result set. UNION ALL returns all duplicate records.


The UNION set operator supports the following syntax:

   { UNION [ ALL ] }



Any SELECT query that Drill supports. See SELECT.

Usage Notes

  • The two SELECT query expressions that represent the direct operands of the UNION must produce the same number of columns. Corresponding columns must contain compatible data types. See Supported Data Types.
  • Multiple UNION operators in the same SELECT statement are evaluated left to right, unless otherwise indicated by parentheses.
  • You can only use * on either side of UNION when the data source has a defined schema, such as data in Hive or views.
  • You must explicitly specify columns.
  • Drill 1.13 and later supports queries on empty directories. An empty directory in a query does not change the results; Drill returns results as if the query does not contain the UNION operator. See Schemaless Tables for more information.


The following example uses the UNION ALL set operator to combine click activity data before and after a marketing campaign. The data in the example exists in the dfs.clicks workspace.

   0: jdbc:drill:> SELECT t.trans_id transaction, t.user_info.cust_id customer
   FROM `clicks/clicks.campaign.json` t
   SELECT u.trans_id, u.user_info.cust_id FROM `clicks/clicks.json` u LIMIT 5;
   | transaction |  customer  |
   | 35232       | 18520      |
   | 31995       | 17182      |
   | 35760       | 18228      |
   | 37090       | 17015      |
   | 37838       | 18737      |

This UNION ALL query returns rows that exist in two files (and includes any duplicate rows from those files): clicks.campaign.json and clicks.json

If a query on either side of the UNION operator queries an empty directory, as shown in the following example where empty_DIR is an empty directory:

   0: jdbc:drill:schema=dfs.tmp> select columns[0] from empty_DIR UNION ALL select cast(columns[0] as int) c1 from `testWindow.csv`;

Drill treats the empty directory as a schemaless table and returns results as if the UNION operator is not included in the query.