[
https://issues.apache.org/jira/browse/DRILL-3781?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14933525#comment-14933525
]
Victoria Markman commented on DRILL-3781:
-----------------------------------------
This fix is verified with 1.2.0
#Thu Sep 24 00:27:47 UTC 2015
git.commit.id.abbrev=d2caa62
Verification was done on a drill view:
with 1.1.0
{code}
0: jdbc:drill:schema=dfs> create or replace view v1(a1, b1, c1) as select
cast(a1 as integer), cast(b1 as varchar(10)), cast(c1 as date) from t1;
+-------+-------------------------------------------------------------+
| ok | summary |
+-------+-------------------------------------------------------------+
| true | View 'v1' replaced successfully in 'dfs.subqueries' schema |
+-------+-------------------------------------------------------------+
1 row selected (0.446 seconds)
0: jdbc:drill:schema=dfs> describe v1;
+--------------+--------------------+--------------+
| COLUMN_NAME | DATA_TYPE | IS_NULLABLE |
+--------------+--------------------+--------------+
| a1 | INTEGER | YES |
| b1 | CHARACTER VARYING | YES |
| c1 | DATE | YES |
+--------------+--------------------+--------------+
3 rows selected (0.911 seconds)
0: jdbc:drill:schema=dfs> select CURRENT_DATE, count(*) from v1 group by
CURRENT_DATE;
Error: PARSE ERROR: From line 1, column 48 to line 1, column 59: Column
'CURRENT_DATE' not found in any table
[Error Id: 97e61cdc-f0c6-4ae4-9194-88583f6e2603 on atsqa4-133.qa.lab:31010]
(state=,code=0)
{code}
with 1.2.0
{code}
0: jdbc:drill:schema=dfs> select CURRENT_DATE, count(*) from v1 group by
CURRENT_DATE;
+---------------+---------+
| CURRENT_DATE | EXPR$1 |
+---------------+---------+
| 2015-09-28 | 10 |
+---------------+---------+
1 row selected (2.642 seconds)
0: jdbc:drill:schema=dfs> select count(*) from v1 group by CURRENT_DATE;
+---------+
| EXPR$0 |
+---------+
| 10 |
+---------+
1 row selected (0.516 seconds)
{code}
Added test to: Functional/Passing/aggregation/bugs
> Using CURRENT_DATE in a group by throws a column not found error for hive
> tables and csv files
> ----------------------------------------------------------------------------------------------
>
> Key: DRILL-3781
> URL: https://issues.apache.org/jira/browse/DRILL-3781
> Project: Apache Drill
> Issue Type: Bug
> Components: Functions - Drill, Query Planning & Optimization
> Reporter: Rahul Challapalli
> Assignee: Jinfeng Ni
> Fix For: 1.2.0
>
> Attachments:
> 0002-DRILL-3781-Group-by-system-function-in-schema-based-.patch,
> csv-error.log, hive_error.log
>
>
> Commit # : e43155d8eabb6fc2d0fa4c68c25d6e7c59bf4521
> Using CURRENT_DATE in a group by seems to failing against hive and csv files.
> With parquet and json, there seems to be no issues.
> Query against a hive table :
> {code}
> select CURRENT_DATE from student_hive group by CURRENT_DATE;
> Error: PARSE ERROR: From line 1, column 48 to line 1, column 59: Column
> 'CURRENT_DATE' not found in any table
> [Error Id: e7d7df50-c5e8-4eda-990a-050b9a2b188e on qa-node190.qa.lab:31010]
> (state=,code=0)
> {code}
> Query against csv files :
> {code}
> select CURRENT_DATE from `temp.tbl` group by CURRENT_DATE;
> Error: DATA_READ ERROR: Selected column 'CURRENT_DATE' must have name
> 'columns' or must be plain '*'
> File Path maprfs:///drill/testdata/temp.tbl
> Fragment 0:0
> [Error Id: 1856f171-966e-4078-bbea-7ff3e9e22e15 on qa-node190.qa.lab:31010]
> (state=,code=0)
> {code}
> A similar query against json and parquet seems to be working fine
> {code}
> select current_date from `a.json` group by current_date;
> +---------------+
> | current_date |
> +---------------+
> | 2015-09-14 |
> +---------------+
> select CURRENT_DATE from cp.`tpch/lineitem.parquet` group by CURRENT_DATE;
> +---------------+
> | CURRENT_DATE |
> +---------------+
> | 2015-09-14 |
> +---------------+
> {code}
> I attached the log files for the failing conditions. Let me know if you need
> anything
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
