[jira] Commented: (CASSANDRA-71) Range query support

["Jun Rao (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/CASSANDRA-71?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12704187#action_12704187 ] 

Jun Rao commented on CASSANDRA-71:
----------------------------------

I reviewed this patch and here are my comments:
1. I'd prefer changing get_range to get_key_range or get_row_range.
2. Remove unreferenced packages in RangeVerbHandler.
3. In Table.getKeyRange(), why is there a constant in the following line? Do you intend to add a count parameter in the get_range interface?
        while (keys.size() < 10000)
4. It seems to me that directly adding an iterator interface to FileStruct makes the class confusing, since there are now two sets of iterator interfaces and it's not clear which is which. I'd prefer leaving the current FileStruct interface as it is and  making FileStruct an input to FileStructIterator or creating a FileStruct inside FileStructIterator.
5. In FileStructIterator, the keys iterated could still be smaller than startKey. So you need to check them.
6. I am not sure if it's a good idea to hide IOExceptions in FileStructIterator. It's better to let upper layers handle it.
7. Please add some new test cases.


> Range query support
> -------------------
>
>                 Key: CASSANDRA-71
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-71
>             Project: Cassandra
>          Issue Type: New Feature
>            Reporter: Jonathan Ellis
>            Assignee: Jonathan Ellis
>             Fix For: 0.3
>
>         Attachments: 71.patch
>
>
> Scan for keys by range (between X and Y) or prefix (starting with P).

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