[
https://issues.apache.org/jira/browse/PHOENIX-3262?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15481034#comment-15481034
]
James Taylor commented on PHOENIX-3262:
---------------------------------------
For simplicity and to match the design for global indexes, we map column
families from the data table to their prefixed "shadow" column families in the
index. That way, you can rewrite the query to be against the index table
without any additional info. You'd only get multiple column families if you
have covered columns from the corresponding data table column families.
Typically local indexes aren't covered, so you'd have the one shadow column
family that had the empty KeyValue from the data table.
Sure, the design could be changed, but I'd prioritize PHOENIX-1590 instead as
it's a good win when dropping view indexes and views. Better bang for the buck
IMHO.
> When dropping a local index, delete the column family instead of individual
> rows.
> ---------------------------------------------------------------------------------
>
> Key: PHOENIX-3262
> URL: https://issues.apache.org/jira/browse/PHOENIX-3262
> Project: Phoenix
> Issue Type: Bug
> Reporter: Lars Hofhansl
> Fix For: 4.9.0
>
>
> Currently when Phoenix drops a local index it places a delete marker for
> every single row, instead we can just drop the L#<x> column family.
> {{HBaseAdmin.deleteColumn}} in 0.98. And {{HBaseAdmin.deleteColumnFamily}} in
> 1.0+.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
