[
https://issues.apache.org/jira/browse/HADOOP-2264?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12545117
]
stack commented on HADOOP-2264:
-------------------------------
What is 'MyCast'? Or UDC? Is it code that a user would have to add to the
CLASSPATH (like PIG) for it to work? If its only input is the column name,
then it looks like you'd have to provide a UDC/MyCast per available formatter?
I suppose we could write one that works with Writables -- i.e. pass cell bytes
to readField and use the toString as formatter -- and that'd be good enough for
most needs.
Would 'formatter' be a better name than 'MyCast/UDC'? Default it could take
column name only and format using Writables. Optionally you could pass in the
name of a class to do custom formatting?
> Support an OutputFormat for HQL row data
> ----------------------------------------
>
> Key: HADOOP-2264
> URL: https://issues.apache.org/jira/browse/HADOOP-2264
> Project: Hadoop
> Issue Type: New Feature
> Components: contrib/hbase
> Reporter: Paul Saab
>
> Currently when selecting a row, if the data does not convert to a String the
> hbase shell will print garbage. It would be nice if HQL supported a
> mechanism to format individual columns. Something along the lines of:
> select col1: format(SomeFormatClass), col2: format(AnotherFormatClass) from
> table ....
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
