[jira] [Created] (DRILL-1958) select * fails on a text file with 3500 columns or more

["Rahul Challapalli (JIRA)" <ji...@apache.org>]

Rahul Challapalli created DRILL-1958:
----------------------------------------

             Summary: select * fails on a text file with 3500 columns or more
                 Key: DRILL-1958
                 URL: https://issues.apache.org/jira/browse/DRILL-1958
             Project: Apache Drill
          Issue Type: Bug
          Components: Storage - Text & CSV
            Reporter: Rahul Challapalli
            Assignee: Hanifi Gunes


git.commit.id.abbrev=b491cdb

The dataset contains a single record with 3500 columns. 'select *' returns an empty set. However with 3000 columns 'select *' has no issues.

Selecting an individual columns worked with both the data sets. Also json does not seem to have this problem

{code}
0: jdbc:drill:schema=dfs.wide-columns> select * from `temp.tsv`;
+--+
|  |
+--+
+--+
No rows selected (0.174 seconds)
0: jdbc:drill:schema=dfs.wide-columns> select columns[0] from `temp.tsv`;
+------------+
|   EXPR$0   |
+------------+
| 4788998.32516 |
+------------+
1 row selected (0.244 seconds)
{code}

There was no activity in the log files when the select * did not return anything. I attached the data file



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)