[
https://issues.apache.org/jira/browse/SPARK-61119?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Hyukjin Kwon updated SPARK-61119:
---------------------------------
Fix Version/s: connect-rust-4.3.0
> Set the element type of the array literals of the Rust core, as PySpark does
> ----------------------------------------------------------------------------
>
> Key: SPARK-61119
> URL: https://issues.apache.org/jira/browse/SPARK-61119
> Project: Spark
> Issue Type: Bug
> Components: Connect
> Affects Versions: connect-rust-4.3.0
> Reporter: Hyukjin Kwon
> Priority: Major
> Fix For: connect-rust-4.3.0
>
>
> The {{spark.connect.Expression.Literal}} of {{LiteralExpression::Array}} of
> the Rust core has no element type: {{to_proto}} sets the elements of
> {{Literal.Array}}, but neither its {{element_type}} nor the {{data_type}} of
> the literal. A Spark 4.3.0 server rejects such a literal, also of elements,
> with {{CONNECT_INVALID_PLAN.ARRAY_LITERAL_MISSING_DATA_TYPE}}. So a column of
> {{Expression::literal(LiteralExpression::Array { .. })}} fails on the server.
> PySpark's {{LiteralExpression.to_plan}} sets the {{element_type}} of an array
> literal, also of nested arrays, and the server accepts it, also for an empty
> array.
> The fix is to set the {{element_type}} of the array literals of the core, as
> PySpark does.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
