hive-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Keegan Mosley (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HIVE-3236) allow column names to be prefixed by table alias in select all queries
Date Sat, 07 Jul 2012 00:47:35 GMT

     [ https://issues.apache.org/jira/browse/HIVE-3236?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Keegan Mosley updated HIVE-3236:
--------------------------------

    Description: 
When using "CREATE TABLE x AS SELECT ..." where the select joins tables with hundreds of columns
it is not a simple task to resolve duplicate column name exceptions (particularly with self-joins).
The user must either manually specify aliases for all duplicate columns (potentially hundreds)
or write a script to generate the data set in a separate select query, then create the table
and load the data in.


There should be some conf flag that would allow queries like

"create table joined as select one.\*, two.\* from mytable one join mytable two on (one.duplicate_field
= two.duplicate_field1);"

to create a table with columns one_duplicate_field and two_duplicate_field.

  was:
When using "CREATE TABLE x AS SELECT ..." where the select joins tables with hundreds of columns
it is not a simple task to resolve duplicate column name exceptions (particularly with self-joins).
The user must either manually specify aliases for all duplicate columns (potentially hundreds)
or write a script to generate the data set in a separate select query, then create the table
and load the data in.


There should be some conf flag that would allow queries like

"create table joined as select one.&#42, two.&#42 from mytable one join mytable two
on (one.duplicate_field = two.duplicate_field1);"

to create a table with columns one_duplicate_field and two_duplicate_field.

    
> allow column names to be prefixed by table alias in select all queries
> ----------------------------------------------------------------------
>
>                 Key: HIVE-3236
>                 URL: https://issues.apache.org/jira/browse/HIVE-3236
>             Project: Hive
>          Issue Type: Improvement
>          Components: Query Processor
>    Affects Versions: 0.10.0, 0.9.1
>            Reporter: Keegan Mosley
>            Priority: Minor
>             Fix For: 0.10.0
>
>         Attachments: HIVE-3236.1.patch.txt
>
>
> When using "CREATE TABLE x AS SELECT ..." where the select joins tables with hundreds
of columns it is not a simple task to resolve duplicate column name exceptions (particularly
with self-joins). The user must either manually specify aliases for all duplicate columns
(potentially hundreds) or write a script to generate the data set in a separate select query,
then create the table and load the data in.
> There should be some conf flag that would allow queries like
> "create table joined as select one.\*, two.\* from mytable one join mytable two on (one.duplicate_field
= two.duplicate_field1);"
> to create a table with columns one_duplicate_field and two_duplicate_field.

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message