[
https://issues.apache.org/jira/browse/HIVE-1271?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12883051#action_12883051
]
Zheng Shao commented on HIVE-1271:
----------------------------------
I might be too late for the party, but I have a question on removing the field
name comparison for struct type info.
We have 3 choices:
C1: Compare field names case sensitively.
C2: Compare field names case insensitively.
C3: Don't compare field names at all.
The old implementation was following C1, and the new one is following C3.
Is there any reason that we don't do C2? C2 seems to provide some minimal
sanity checks that users will need in practice.
> Case sensitiveness of type information specified when using custom reducer
> causes type mismatch
> -----------------------------------------------------------------------------------------------
>
> Key: HIVE-1271
> URL: https://issues.apache.org/jira/browse/HIVE-1271
> Project: Hadoop Hive
> Issue Type: Bug
> Components: Query Processor
> Affects Versions: 0.5.0
> Reporter: Dilip Joseph
> Assignee: Arvind Prabhakar
> Fix For: 0.6.0
>
> Attachments: HIVE-1271-1.patch, HIVE-1271.patch
>
>
> Type information specified while using a custom reduce script is converted
> to lower case, and causes type mismatch during query semantic analysis . The
> following REDUCE query where field name = "userId" failed.
> hive> CREATE TABLE SS (
> > a INT,
> > b INT,
> > vals ARRAY<STRUCT<userId:INT, y:STRING>>
> > );
> OK
> hive> FROM (select * from srcTable DISTRIBUTE BY id SORT BY id) s
> > INSERT OVERWRITE TABLE SS
> > REDUCE *
> > USING 'myreduce.py'
> > AS
> > (a INT,
> > b INT,
> > vals ARRAY<STRUCT<userId:INT, y:STRING>>
> > )
> > ;
> FAILED: Error in semantic analysis: line 2:27 Cannot insert into
> target table because column number/types are different SS: Cannot
> convert column 2 from array<struct<userId:int,y:string>> to
> array<struct<userid:int,y:string>>.
> The same query worked fine after changing "userId" to "userid".
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
