You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@spark.apache.org by "Uroš Bojanić (Jira)" <ji...@apache.org> on 2026/08/23 14:02:00 UTC

[jira] [Commented] (SPARK-58948) Deduplicate nulls-buffer creation in columnar decompressors

    [ https://issues.apache.org/jira/browse/SPARK-58948?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=18107119#comment-18107119 ] 

Uroš Bojanić commented on SPARK-58948:
--------------------------------------

In progress: https://github.com/apache/spark/pull/58228.

> Deduplicate nulls-buffer creation in columnar decompressors
> -----------------------------------------------------------
>
>                 Key: SPARK-58948
>                 URL: https://issues.apache.org/jira/browse/SPARK-58948
>             Project: Spark
>          Issue Type: Improvement
>          Components: SQL
>    Affects Versions: 4.3.0
>            Reporter: Uroš Bojanić
>            Priority: Major
>              Labels: pull-request-available
>
> Each columnar decompressor (PassThrough, RunLengthEncoding, DictionaryEncoding, BooleanBitSet, IntDelta, LongDelta) repeats the same two-line idiom to build the nulls buffer: buffer.duplicate().order(ByteOrder.nativeOrder()) followed by rewind(). This extracts a protected createNullsBuffer(buffer) helper on the shared Decoder trait and routes the six call sites through it. No behavior change; it also normalizes one site that called rewind without parentheses.



--
This message was sent by Atlassian Jira
(v8.20.10#820010)

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