cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Vijay (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (CASSANDRA-2710) Get multiple column ranges
Date Thu, 16 Aug 2012 17:21:38 GMT

    [ https://issues.apache.org/jira/browse/CASSANDRA-2710?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13436124#comment-13436124
] 

Vijay commented on CASSANDRA-2710:
----------------------------------

Hi Jonathan, Consider a Use case where we have a hierarchical data, type:latest:1:2, type:v1:1:2,
type:v2:1,2 etc the user might want to query all the data most of the time, sometimes only
the latest and sometimes a specific versions of a given type. 

1) if we model the type to be part of the Row Key then the problem is for 80% or so use case
i will be doing a multi-get (we dont advice OPP so sometimes you might need a index). 
2) if i have all of them in one row then i will be doing multiple calls to get the data out.


I am not arguing the need for it, there are other ways you can get it done (by adding the
type and v1 in the super column name or something like that)... but it will be little more
flexible. I am fine closing the ticket too :) Let me know Thanks! 
                
> Get multiple column ranges
> --------------------------
>
>                 Key: CASSANDRA-2710
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-2710
>             Project: Cassandra
>          Issue Type: Sub-task
>          Components: API, Core
>            Reporter: David Boxenhorn
>            Assignee: Vijay
>              Labels: compositeColumns, cql
>         Attachments: 0001-2710-multiple-column-ranges-cql.patch, 0001-2710-multiple-column-ranges-thrift.patch
>
>
> I have replaced all my super column families with regular column families using composite
columns. I have easily been able to support all previous functionality (I don't need range
delete) except for one thing: getting multiple super columns with a single access. For this,
I would need to get multiple ranges. (I can get multiple columns, or a single range, but not
multiple ranges.) 
> For example, I used to have
> [<superColumnName1>,<subColumnName1..N>],[<superColumnName2>,<subColumnName1..N>]
> and I could get superColumnName1, superColumnName2
> Now I have
> [<len><superColumnName1>0<len><subColumnName1>..<len><superColumnName1>0<len><subColumnNameN>],[<len><superColumnName2>0<len><subColumnName1>..<len><superColumnName2>0<len><subColumnNameN>]
> and I need to get superColumnName1..superColumnName1+, superColumnName2..superColumnName2+
> to get the same functionality
> I would like the clients to support this functionality, e.g. Hector to have .setRages
parallel to .setColumnNames 
> and for CQL to support a syntax like 
> SELECT [FIRST N] [REVERSED] name1..nameN1, name2..nameN2... FROM ...

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message