[
https://issues.apache.org/jira/browse/HIVE-12311?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gunther Hagleitner reassigned HIVE-12311:
-----------------------------------------
Assignee: Gunther Hagleitner
> explain CTAS fails if the table already exists
> ----------------------------------------------
>
> Key: HIVE-12311
> URL: https://issues.apache.org/jira/browse/HIVE-12311
> Project: Hive
> Issue Type: Bug
> Components: Hive
> Affects Versions: 1.2.1
> Reporter: Carter Shanklin
> Assignee: Gunther Hagleitner
> Priority: Minor
>
> Explain of a CTAS will fail if the table already exists.
> This is an annoyance when you're seeing if a large body of SQL queries will
> function by putting explain in front of every query.
> {code}
> hive> create table temp (x int);
> OK
> Time taken: 0.252 seconds
> hive> create table temp2 (x int);
> OK
> Time taken: 0.407 seconds
> hive> explain create table temp as select * from temp2;
> FAILED: SemanticException org.apache.hadoop.hive.ql.parse.SemanticException:
> Table already exists: mydb.temp
> {code}
> If we compare to Postgres "The Zinc Standard of SQL Compliance":
> {code}
> carter=# create table temp (x int);
> CREATE TABLE
> carter=# create table temp2 (x int);
> CREATE TABLE
> carter=# explain create table temp as select * from temp2;
> QUERY PLAN
> ---------------------------------------------------------
> Seq Scan on temp2 (cost=0.00..34.00 rows=2400 width=4)
> (1 row)
> {code}
> If the CTAS is something complex it would be nice to see the query plan in
> advance.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
