[ 
https://issues.apache.org/jira/browse/HIVE-3576?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13484004#comment-13484004
 ] 

Harsh J commented on HIVE-3576:
-------------------------------

This also only happens on tables partitioned by a string type key, not int 
types, for example:

{code}
hive> create table str_part_table (a string) partitioned by (dt string);
OK
Time taken: 0.054 seconds
hive> create table int_part_table (a string) partitioned by (dt int);   
OK
Time taken: 0.031 seconds
hive> alter table str_part_table drop if exists partition (dt=2007);
FAILED: SemanticException [Error 10006]: Partition not found dt = 2007
hive> alter table int_part_table drop if exists partition (dt=2007);
OK
Time taken: 0.091 seconds
hive> alter table str_part_table drop if exists partition (dt='2007');
OK
Time taken: 0.06 seconds
hive> alter table int_part_table drop if exists partition (dt='2007');
OK
Time taken: 0.065 seconds
{code}
                
> Regression: ALTER TABLE DROP IF EXISTS PARTITION throws a SemanticException 
> if Partition is not found
> -----------------------------------------------------------------------------------------------------
>
>                 Key: HIVE-3576
>                 URL: https://issues.apache.org/jira/browse/HIVE-3576
>             Project: Hive
>          Issue Type: Bug
>          Components: Metastore, Query Processor
>    Affects Versions: 0.9.0
>            Reporter: Harsh J
>
> Doing a simple "{{ALTER TABLE testtable DROP IF EXISTS 
> PARTITION(dt=NONEXISTENTPARTITION)}}" fails with a SemanticException of the 
> 10006 kind (INVALID_PARTITION).
> This does not respect the {{hive.exec.drop.ignorenonexistent}} condition 
> either, since there are no if-check-wraps around this area, when fetching 
> partitions from the store.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

Reply via email to