[jira] [Comment Edited] (SPARK-15276) CREATE TABLE with LOCATION should imply EXTERNAL

2018-03-27 Thread Volodymyr Glushak (JIRA)

[ 
https://issues.apache.org/jira/browse/SPARK-15276?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16415825#comment-16415825
 ] 

Volodymyr Glushak edited comment on SPARK-15276 at 3/27/18 3:55 PM:


[~andrewor14],

HIVE does not "externalise" table  if LOCATION specified. (I reckon Impala 
neither)

Why does Apache Spark introduce different behaviour? 

 

 


was (Author: rumoku):
[~andrewor14],

HIVE does not "externalise" table  if LOCATION specified.

Why does Apache Spark introduce different behaviour? 

 

 

> CREATE TABLE with LOCATION should imply EXTERNAL
> 
>
> Key: SPARK-15276
> URL: https://issues.apache.org/jira/browse/SPARK-15276
> Project: Spark
>  Issue Type: Bug
>  Components: SQL
>Affects Versions: 2.0.0
>Reporter: Andrew Or
>Assignee: Andrew Or
>Priority: Major
> Fix For: 2.0.0
>
>
> If the user runs `CREATE TABLE some_table ... LOCATION /some/path`, then this 
> will still be a managed table even though the table's data is stored at 
> /some/path. The problem is that when we drop the table we'll also delete the 
> data /some/path. This could cause problems if /some/path contains existing 
> data.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org



[jira] [Comment Edited] (SPARK-15276) CREATE TABLE with LOCATION should imply EXTERNAL

2018-03-27 Thread Volodymyr Glushak (JIRA)

[ 
https://issues.apache.org/jira/browse/SPARK-15276?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16415825#comment-16415825
 ] 

Volodymyr Glushak edited comment on SPARK-15276 at 3/27/18 3:51 PM:


[~andrewor14],

HIVE does not "externalise" table  if LOCATION specified.

Why does Apache Spark introduce different behaviour? 

 

 


was (Author: rumoku):
[~andrewor14],

HIVE is not adding EXTERNAL keyword  if LOCATION specified.

Why does Apache Spark introduce different behaviour? 

 

 

> CREATE TABLE with LOCATION should imply EXTERNAL
> 
>
> Key: SPARK-15276
> URL: https://issues.apache.org/jira/browse/SPARK-15276
> Project: Spark
>  Issue Type: Bug
>  Components: SQL
>Affects Versions: 2.0.0
>Reporter: Andrew Or
>Assignee: Andrew Or
>Priority: Major
> Fix For: 2.0.0
>
>
> If the user runs `CREATE TABLE some_table ... LOCATION /some/path`, then this 
> will still be a managed table even though the table's data is stored at 
> /some/path. The problem is that when we drop the table we'll also delete the 
> data /some/path. This could cause problems if /some/path contains existing 
> data.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

-
To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org
For additional commands, e-mail: issues-h...@spark.apache.org