[CARBONDATA-1315] Fix wrong test cases

• Type: Bug
• Status: Open
• Priority: Trivial
• Resolution: Unresolved
• Component/s: test
• Assignee: Unassigned
• Reporter: Liang Chen
• Created: 19/Jul/17 03:02
• Updated: 19/Jul/17 03:02

Description

In spark 1.x, the below query(improper format filter) gets "null" as result; In spark 2.1,system corrects this issue, the below query gets all data, because the where condition doesn't filter any data.
sql("select empid from a12_all_null " + "where empid not in ('china',NULL)")

So, need to fix the wrong test case under Spark 1.x integration module.

The table data as below :
empid,ename,sal,deptno,mgr,gender,dob,comm,desc
1,abc,1233,10,2,,2014-07-01 12:07:28,1234.191,string_null
2,bcd,1322,,3,f,2014-07-01 12:07:28,19.99,int_null
3,cde,4322,,4,m,2014-07-01 12:07:28,16.996,date_null
4, ,43243,,5,m,2014-07-01 12:07:28,999.117,string_space
5,,43242,20,6,m,2017-07-01 12:07:28,99.999,string_null
6,ijk,,20,6,m,2017-07-01 12:07:28,50089,double_null
7,pqr,2422,20,6,m,2017-07-01 12:07:28,32.339,decimal_null