Eric Hanson created HIVE-6186:
---------------------------------

             Summary: error in vectorized Column-Column comparison filter for 
repeating case
                 Key: HIVE-6186
                 URL: https://issues.apache.org/jira/browse/HIVE-6186
             Project: Hive
          Issue Type: Sub-task
          Components: Query Processor
    Affects Versions: 0.13.0
            Reporter: Eric Hanson


FilterColumnCompareColumn.txt  contains this code:

    } else if (inputColVector1.isRepeating && inputColVector2.isRepeating) {
      if (nullPos1[0] || nullPos2[0]) {
        batch.size = 0; 
      } 

This code examines nullPos1[0] and nullPos2[0] without checking that 
noNulls==false for the appropriate vector. This could lead to incorrect results.

We should review the case analysis for this template more generally to make 
sure that all combinations are covered:

left.noNulls        right.noNulls
T                         T
F                         T
T                         F
F                         F                 




--
This message was sent by Atlassian JIRA
(v6.1.5#6160)

Reply via email to