You are viewing a plain text version of this content. The canonical link for it is here.
Posted to commits@cassandra.apache.org by "Stu Hood (JIRA)" <ji...@apache.org> on 2010/05/19 01:30:57 UTC
[jira] Updated: (CASSANDRA-932) Use cfids instead of cfnames for
major operations.
[ https://issues.apache.org/jira/browse/CASSANDRA-932?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Stu Hood updated CASSANDRA-932:
-------------------------------
Attachment: 0001-Remove-redundant-cfNamesMap-in-Table.patch
0002-Add-easy-reverse-lookup-from-id-cf.patch
0003-Remove-name-field-from-ColumnFamily-and-clean-up-mem.patch
0001 Removes the cfNamesMap from Table, which was redundant with the global id->cf mapping in CFMetaData.
0002 Exposes id->cf lookups.
0003 Removes name_ from ColumnFamily, and makes the other members more consistent.
As you said, there is probably a lot more that could be done, but it is a start.
> Use cfids instead of cfnames for major operations.
> --------------------------------------------------
>
> Key: CASSANDRA-932
> URL: https://issues.apache.org/jira/browse/CASSANDRA-932
> Project: Cassandra
> Issue Type: Improvement
> Affects Versions: 0.7
> Reporter: Gary Dusbabek
> Assignee: Stu Hood
> Fix For: 0.7
>
> Attachments: 0001-Remove-redundant-cfNamesMap-in-Table.patch, 0002-Add-easy-reverse-lookup-from-id-cf.patch, 0003-Remove-name-field-from-ColumnFamily-and-clean-up-mem.patch
>
>
> CF.id() should be used in place of CF.name() for looking up column families.
> In pragmatic terms this means that:
> * maps that store CF or CFS should use id as the key.
> * CF.name() should be eliminated and replaced by a toString() that includes the cfid and name.
> * other things I can't immediately think of.
> We might even go so far as to change the way we name the sstable files to use cfid instead of name. This would make renaming CFs a simpler (and quicker) operation.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.