[
https://issues.apache.org/jira/browse/SPARK-57852?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Max Gekk updated SPARK-57852:
-----------------------------
Shepherd: Max Gekk
> Support TIME endpoints in the sequence function
> -----------------------------------------------
>
> Key: SPARK-57852
> URL: https://issues.apache.org/jira/browse/SPARK-57852
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Affects Versions: 4.3.0
> Reporter: Max Gekk
> Priority: Major
>
> This sub-task is part of the umbrella SPARK-57550 (extend support for the
> TIME data type).
> h2. Problem
> {{Sequence}} (collectionOperations.scala {{checkInputDataTypes}} ~L3316-3333)
> allows only integral / {{DateType}} / {{TimestampType}} /
> {{TimestampNTZType}} endpoints; {{TimeType}} falls to {{case _ => false}}, so
> {{sequence}} over TIME raises an analysis error.
> h2. Goal
> Generate an array of TIME values from TIME start/stop with a day-time
> interval step, preserving the declared precision and honoring the within-day
> domain.
> h2. Scope
> Extend {{Sequence.checkInputDataTypes}} and the implementation to accept
> {{AnyTimeType}} endpoints with a day-time interval step, reusing the
> {{TimeAddInterval}} semantics.
> h2. Acceptance criteria
> * {{sequence(TIME '08:00:00', TIME '10:00:00', INTERVAL '30' MINUTE)}} yields
> the expected TIME array of the input precision.
> h2. Testing
> {{CollectionExpressionsSuite}}; TIME golden files.
> h2. Dependencies
> None hard (uses resolved {{TimeAddInterval}}). Coordinate with the TIME +/-
> INTERVAL out-of-range semantics sub-task for the within-day/overflow behavior.
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
