[
https://issues.apache.org/jira/browse/SPARK-61850?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Dimitrije Zdravkovic updated SPARK-61850:
-----------------------------------------
Labels: pull-requests-available (was: pull-request-available)
> Add GAP FILL syntax, parser node and feature gate
> -------------------------------------------------
>
> Key: SPARK-61850
> URL: https://issues.apache.org/jira/browse/SPARK-61850
> Project: Spark
> Issue Type: Sub-task
> Components: SQL
> Affects Versions: 4.4.0
> Reporter: Dimitrije Zdravkovic
> Priority: Major
> Labels: pull-requests-available
>
> Add the syntax for the \{{GAP FILL}} relation operator, as the first step of
> SPARK-61276.
> * Parse \{{relation GAP FILL (...)}} into a new \{{UnresolvedGapFill}} plan
> node. The pipe forms \{{|> GAP FILL (...)}} and, when the single-character
> pipe is enabled, \{{| GAP FILL (...)}} also work.
> * Add 9 non-reserved keywords: \{{BACKWARD}}, \{{BFILL}}, \{{CONSTANT}},
> \{{FFILL}}, \{{FILL}}, \{{FORWARD}}, \{{GAP}}, \{{IS_GENERATED}}, \{{LINEAR}}.
> * Add the internal conf \{{spark.sql.gapFillRelationOperator.enabled}},
> default \{{false}}.
> Every \{{GAP FILL}} query is still rejected with
> \{{UNSUPPORTED_FEATURE.GAP_FILL}}. Resolution and execution come in later
> sub-tasks.
--
This message was sent by Atlassian Jira
(v10.3.26#10030026)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]