[SPARK-25452] Query with where clause is giving unexpected result in case of float column

• Type: Bug
• Status: Resolved
• Priority: Major
• Resolution: Incomplete
• Affects Version/s: 2.3.1
• Component/s: SQL
• Labels: bulk-closed

Environment

Spark 2.3.1

Hadoop 2.7.2

• Assignee: Unassigned
• Reporter: Ayush Anubhava
• Created: 18/Sep/18 11:20
• Updated: 12/Dec/22 18:10
• Resolved: 25/May/21 01:43

Description

Description : Query with clause is giving unexpected result in case of float column 

 

Query with filter less than equal to is giving inappropriate result

0: jdbc:hive2://10.18.18.214:23040/default> create table k2 ( a int, b float);
+---------+--+
| Result |
+---------+--+
+---------+--+
0: jdbc:hive2://10.18.18.214:23040/default> insert into table k2 values (0,0.0);
+---------+--+
| Result |
+---------+--+
+---------+--+
0: jdbc:hive2://10.18.18.214:23040/default> insert into table k2 values (1,1.1);
+---------+--+
| Result |
+---------+--+
+---------+--+
0: jdbc:hive2://10.18.18.214:23040/default> select * from k2 where b >=0.0;
+----+--------------------+--+
| a | b |
+----+--------------------+--+
| 0 | 0.0 |
| 1 | 1.100000023841858 |
+----+--------------------+--+

Query with filter less than equal to is giving in appropriate result

0: jdbc:hive2://10.18.18.214:23040/default> select * from k2 where b <=1.1;
+----+------+--+
| a | b |
+----+------+--+
| 0 | 0.0 |
+----+------+--+
1 row selected (0.299 seconds)