[jira] Commented: (CASSANDRA-1154) basic, single node indexing against one indexexpression

["Jonathan Ellis (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-1154?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12874968#action_12874968 ] 

Jonathan Ellis commented on CASSANDRA-1154:
-------------------------------------------

the one part of this sub-task not dealt with in 1153 or a later sub-task is, here I create index CFs always with BytesType comparator.  This is correct for OPP (technically, BOP) but we need to support mirroring other key orderings, as column orderings.  In other words, just as LocalPartitioner uses the value ordering to store index rows on disk in the correct order (to support operators other than EQ later on), we need to create a PartitonerOrdered[Abstract]Type that orders columns w/in an index row (that is, columns whose name is a row key) by the same order those row keys are ordered on disk.  This allows the optimizations mentioned in CASSANDRA-1156.

> basic, single node indexing against one indexexpression
> -------------------------------------------------------
>
>                 Key: CASSANDRA-1154
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-1154
>             Project: Cassandra
>          Issue Type: Sub-task
>          Components: Core
>            Reporter: Jonathan Ellis
>             Fix For: 0.7
>
>         Attachments: 1154.txt
>
>


-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.