Serge Rielau created SPARK-59001:
------------------------------------

             Summary: Empty2Null, text, and Hive prune fallback for CHAR/VARCHAR
                 Key: SPARK-59001
                 URL: https://issues.apache.org/jira/browse/SPARK-59001
             Project: Spark
          Issue Type: Sub-task
          Components: Spark Core
    Affects Versions: 5.0.0
            Reporter: Serge Rielau


Under spark.sql.charVarchar.standardSemantics.enabled, CHAR/VARCHAR must
work on V1 write/scan paths that still special-case StringType.

* V1Writes Empty2Null: treat the string family as empty-to-null for
  partition columns (CHAR(0) / empty VARCHAR become NULL), without
  widening Hive's SupportedAttribute list.
* Text datasource: accept CHAR/VARCHAR as text, same as STRING.
* Hive partition prune: when the partition schema includes CHAR/VARCHAR,
  fall back to client-side prunePartitionsFastFallback. Hive still does
  not push CHAR/VARCHAR keys as SupportedAttribute. PAD SPACE compare
  is collation/RTRIM, not a type-level policy.

PR: https://github.com/apache/spark/pull/58255



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

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to