You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@spark.apache.org by "Apache Spark (JIRA)" <ji...@apache.org> on 2015/07/27 18:05:04 UTC

[jira] [Commented] (SPARK-9358) Create a UnsafeRowConcat class

    [ https://issues.apache.org/jira/browse/SPARK-9358?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14642926#comment-14642926 ] 

Apache Spark commented on SPARK-9358:
-------------------------------------

User 'yjshen' has created a pull request for this issue:
https://github.com/apache/spark/pull/7698

> Create a UnsafeRowConcat class
> ------------------------------
>
>                 Key: SPARK-9358
>                 URL: https://issues.apache.org/jira/browse/SPARK-9358
>             Project: Spark
>          Issue Type: Sub-task
>          Components: SQL
>            Reporter: Reynold Xin
>            Assignee: Yijie Shen
>
> A class that can be used to concatenate UnsafeRows together.
> {code}
> class UnsafeRowConcat(leftSchema: StructType, rightSchema: StructType) {
>   def concat(left: UnsafeRow, right: UnsafeRow): UnsafeRow = {
>     // ...
>   }
> }
> {code}
> concat should reuse the same buffer space, to avoid constant allocation.
> UnsafeRow has 3 parts: bitset, fixed-length portion, and variable length portion.
> We need to concat the bitsets, the fixed-length portions, append the variable length portions,
> and then update the fields with variable length portion to change the new offsets.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

---------------------------------------------------------------------
To unsubscribe, e-mail: issues-unsubscribe@spark.apache.org
For additional commands, e-mail: issues-help@spark.apache.org