[DRILL-1642] flatten function is dropping the last record when the json file contains even no of records

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Duplicate
• Fix Version/s: 0.7.0
• Component/s: Storage - JSON
• Assignee: Jason Altekruse
• Reporter: Rahul Kumar Challapalli
• Created: 06/Nov/14 01:16
• Updated: 07/May/15 01:06
• Resolved: 09/Nov/14 20:30

Description

git.commit.id.abbrev=5adadfa

The below query does not produce the right output. It is missing the last record from the data file. The record count is even for the data set. However if it is odd, the flatten operator has not issues .

0: jdbc:drill:schema=dfs> select rownum, flatten(complex) from `flatten.json`;
+------------+------------+
|   rownum   |   EXPR$1   |
+------------+------------+
| 1          | {"col1":3} |
| 1          | {"col2":2,"col3":1} |
| 1          | {"col1":7} |
| 2          | {"col2":2,"col3":1} |
| 2          | {"col1":7} |
| 3          | {"col1":2,"col3":1} |
+------------+------------+

Attached the data file. Let me know if you need anything more.

Issue Links

• duplicates: DRILL-1643 Group count incorrect in repeated map vector