Vitalii Diravka created DRILL-6291:
--------------------------------------
Summary: Wrong result for COUNT function and empty file
(schemaless table)
Key: DRILL-6291
URL: https://issues.apache.org/jira/browse/DRILL-6291
Project: Apache Drill
Issue Type: Bug
Components: Functions - Drill
Affects Versions: 1.13.0
Reporter: Vitalii Diravka
Fix For: Future
The count function shouldn't return null result. For the 0 rows case it should
always return 0.
{code}
0: jdbc:drill:zk=local> select count(`last_name`) from
dfs.`/tmp/empty_file.json`;
+---------+
| EXPR$0 |
+---------+
+---------+
No rows selected (0.3 seconds)
{code}
The result should be the similar to:
{code}
0: jdbc:drill:zk=local> select count(`non_existent_column`) from
cp.`employee.json`;
+---------+
| EXPR$0 |
+---------+
| 0 |
+---------+
1 row selected (0.274 seconds)
{code}
Note: empty_file.json is an empty file.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
