Mailing-List: contact dev-help@hive.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@hive.apache.org
Content-Type: multipart/alternative;
 boundary="===============3707433728451877702=="
MIME-Version: 1.0
Subject: Re: Review Request 14970: HIVE-4723
 DDLSemanticAnalyzer.addTablePartsOutputs
 eats several exceptions
From: "Szehon Ho" <szehon@cloudera.com>
To: "Szehon Ho" <szehon@cloudera.com>, "hive" <dev@hive.apache.org>,
 "Brock Noland" <brock@cloudera.com>
Date: Mon, 28 Oct 2013 21:57:46 -0000
Message-ID: <20131028215746.7402.1996@reviews.apache.org>
Auto-Submitted: auto-generated
Sender: "Szehon Ho" <noreply@reviews.apache.org>
References: <20131028213333.7400.54291@reviews.apache.org>
In-Reply-To: <20131028213333.7400.54291@reviews.apache.org>
Reply-To: "Szehon Ho" <szehon@cloudera.com>

--===============3707433728451877702==
Content-Type: text/plain; charset="utf-8"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit


> On Oct. 28, 2013, 9:33 p.m., Brock Noland wrote:
> > ql/src/java/org/apache/hadoop/hive/ql/parse/DDLSemanticAnalyzer.java, line 2936
> > <https://reviews.apache.org/r/14970/diff/2/?file=372522#file372522line2936>
> >
> >     Hi,
> >     
> >     Thank you for working on this! It's not actually being logged though.
> >     
> >     See 
> >     
> >     LOG.error("Got HiveException during obtaining list of partitions");
> >     
> >     No exception is being logged. That is the reason I created this JIRA, not to throw a new exception.

Thanks, I got your point :)...  I will add the exception to the log.  The JIRA said "The error message was bad, hive ate an exception, and NPE'ed." thats why I thought that ate exception and NPE was the main problem.

I think there is some value in throwing a semantic exception with the correct error message if we get exception here, instead of allowing it to go forward, as it is a better exception to the end-user that way.  Swallowing it will get either NPE down the line for non-partitioned table, or some exception of DDL Task.  Or should we just keep the existing behavior?  Thanks


- Szehon


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/14970/#review27638
-----------------------------------------------------------


On Oct. 28, 2013, 8:40 p.m., Szehon Ho wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/14970/
> -----------------------------------------------------------
> 
> (Updated Oct. 28, 2013, 8:40 p.m.)
> 
> 
> Review request for hive.
> 
> 
> Bugs: HIVE-4723
>     https://issues.apache.org/jira/browse/HIVE-4723
> 
> 
> Repository: hive-git
> 
> 
> Description
> -------
> 
> Throwing the offending exception, that was swallowed.  Also changing the underlying exception to use the standard error message.
> 
> 
> Diffs
> -----
> 
>   ql/src/java/org/apache/hadoop/hive/ql/metadata/Hive.java b0f124b 
>   ql/src/java/org/apache/hadoop/hive/ql/parse/DDLSemanticAnalyzer.java 037191a 
>   ql/src/test/results/clientnegative/alter_rename_partition_failure3.q.out 384fcbe 
>   ql/src/test/results/clientnegative/touch2.q.out 045121a 
> 
> Diff: https://reviews.apache.org/r/14970/diff/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Szehon Ho
> 
>


--===============3707433728451877702==--