[jira] [Created] (SPARK-57854) Support the parenthesized interval-qualified datetime difference syntax

["Max Gekk (Jira)" <ji...@apache.org>]

Max Gekk created SPARK-57854:
--------------------------------

             Summary: Support the parenthesized interval-qualified datetime difference syntax
                 Key: SPARK-57854
                 URL: https://issues.apache.org/jira/browse/SPARK-57854
             Project: Spark
          Issue Type: Sub-task
          Components: SQL
    Affects Versions: 4.3.0
            Reporter: Max Gekk


This sub-task is part of the umbrella SPARK-57550 (extend support for the TIME data type).

h2. Problem
ANSI defines a parenthesized, interval-qualified datetime difference, e.g. {{(t1 - t2) DAY TO SECOND}} / {{(t1 - t2) HOUR}} (std ~L16762-16763). Spark only supports the bare {{t1 - t2}} form, which for TIME yields a fixed {{DayTimeIntervalType(HOUR, SECOND)}} ({{SubtractTimes}}). There is no parser/resolver support for the qualified form.

h2. Goal
Add grammar and analyzer support for the parenthesized interval-qualified datetime difference (for TIME and, consistently, DATE/TIMESTAMP), producing the requested interval qualifier with documented rounding/truncation.

h2. Scope
Extend the SQL grammar and {{AstBuilder}} for the {{( <datetime> - <datetime> ) <interval qualifier>}} form; resolve to the appropriate subtraction expression cast to the requested qualifier.

h2. Acceptance criteria
* {{(TIME '10:00:00' - TIME '08:00:00') HOUR}} returns an {{INTERVAL HOUR}} value; other qualifiers behave per ANSI.

h2. Testing
Parser suite (e.g. {{ExpressionParserSuite}}) and {{TimeExpressionsSuite}}; golden files.

h2. Dependencies
None hard; largest/most speculative item in this set. Relates to the ANSI OVERLAPS predicate sub-task (SPARK-57586).




--
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