[DRILL-2297] INTERVAL YEAR columns are missing the converted type when stored in parquet files

• Type: Bug
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 0.8.0
• Fix Version/s: Future
• Component/s: Storage - Parquet
• Labels: interval
• Assignee: Unassigned
• Reporter: Abdel Hakim Deneche
• Created: 24/Feb/15 23:11
• Updated: 12/Nov/15 16:57
• Resolved: 12/Nov/15 16:56

Description

When creating a parquet file with the following query:

CREATE TABLE dfs.tmp.`test_intervalyear` AS SELECT interval '1-2' YEAR TO MONTH AS INTERVALYEAR_col FROM dfs.data.`test.json`;

parquet tools display the following metadata for the generated parquet file:

creator:          parquet-mr 

file schema:      root 
----------------------------------------------------------------------
INTERVALYEAR_col: REQUIRED INT32 R:0 D:0

row group 1:      RC:5 TS:52 
----------------------------------------------------------------------

querying the file back displays the following results:

0: jdbc:drill:zk=local> select * from dfs.tmp.`test_intervalyear`;
+------------------+
| intervalyear_col |
+------------------+
| 14               |
| 14               |
| 14               |
| 14               |
| 14               |
+------------------+