[
https://issues.apache.org/jira/browse/SQOOP-1054?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13681582#comment-13681582
]
Venkat Ranganathan commented on SQOOP-1054:
-------------------------------------------
It might not be fully possible to support splitting based on binary columns in
a db platform neutral way. I think doing a query import with an explicit
conversion of the column might help - For example,
select convert(binary_col using utf8), col2,...
from table
where $CONDITIONS
is something that can be used.
> New feature to support importing MySQL table via binary type splitter
> ----------------------------------------------------------------------
>
> Key: SQOOP-1054
> URL: https://issues.apache.org/jira/browse/SQOOP-1054
> Project: Sqoop
> Issue Type: New Feature
> Components: connectors/mysql, tools
> Affects Versions: 1.4.3
> Environment: $uname -a
> Linux bl1253.cn.alcatel-lucent.com 2.6.18-128.7.1.el5PAE #1 SMP Mon Aug 24
> 08:41:30 EDT 2009 i686 athlon i386 GNU/Linux
> Reporter: wenzhi wang
>
> I'm using (Sqoop 1.4.3 + Hadoop 1.0.4) handle data of our product, which
> employs MySQL 5.1.6 and most tables are defined with primary key on binary
> columns. Sqoop does not support binary type column as splitter now, and I
> have to import tables by restricting mapper number to 1 which lowers the
> performance. So, I open this feature to request Sqoop support binary type
> splitter.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
