Vitalii Diravka created DRILL-6774: -------------------------------------- Summary: Wrong data types of empty batches schema for queries with aliases Key: DRILL-6774 URL: https://issues.apache.org/jira/browse/DRILL-6774 Project: Apache Drill Issue Type: Bug Reporter: Vitalii Diravka
{code:java} 0: jdbc:drill:zk=local> select name as full_name from (select CAST(Product AS VARCHAR) AS name from dfs.`/tmp/empty.json`); +------------+ | full_name | +------------+ +------------+ No rows selected (0.177 seconds) {code} The data type for above query is INT:OPTIONAL, but should be VARCHAR:OPTIONAL. It can be verified only via Drill unit test framework or via usage of UNION ALL operator with other query: The usual result: {code:java} 0: jdbc:drill:zk=local> SELECT full_name FROM cp.`employee.json` LIMIT 2; +------------------+ | full_name | +------------------+ | Sheri Nowmer | | Derrick Whelply | +------------------+ 2 rows selected (0.207 seconds) {code} But after UNION ALL with the above empty output: {code:java} 0: jdbc:drill:zk=local> select name as full_name from (select CAST(Product AS VARCHAR) AS name from dfs.`/tmp/empty.json`) UNION ALL SELECT full_name FROM cp.`employee.json` LIMIT 2; +------------+ | full_name | +------------+ | null | | null | +------------+ 2 rows selected (0.198 seconds) {code} Perhaps it is a regression of DRILL-5546 and the solution could be similar to DRILL-6773 -- This message was sent by Atlassian JIRA (v7.6.3#76005)