uros-b commented on code in PR #56616:
URL: https://github.com/apache/spark/pull/56616#discussion_r3444968795
##########
sql/core/src/test/scala/org/apache/spark/sql/TimestampNanosFunctionsSuiteBase.scala:
##########
@@ -481,6 +481,41 @@ abstract class TimestampNanosFunctionsSuiteBase extends
SharedSparkSession {
checkAnswer(ltz.select(unix_nanos(col("c"))), Row(null))
}
}
+
+ test("SPARK-57526: timestamp_nanos builds nanosecond-precision TIMESTAMP_LTZ
values") {
+ // 1230219000123456789 ns since the epoch -> 2008-12-25 15:30:00.123456789
UTC. The result is a
+ // TIMESTAMP_LTZ(9); collecting it yields the absolute Instant regardless
of the session zone.
+ val nanos = 1230219000123456789L
+ val instant = Instant.parse("2008-12-25T15:30:00.123456789Z")
+ val sqlRes = spark.sql(s"SELECT timestamp_nanos($nanos)")
+ val colRes = spark.range(1).select(timestamp_nanos(lit(nanos)))
+ // The SQL and Scala Column API agree, return the expected instant, and
keep the LTZ(9) type.
+ checkAnswer(sqlRes, colRes)
+ checkAnswer(sqlRes, Row(instant))
+ assert(sqlRes.schema.head.dataType === TimestampLTZNanosType(9))
+
+ // A BIGINT argument is implicitly cast to DECIMAL, so the integral
literal works directly.
Review Comment:
Nit: This comment seems inaccurate; the expression uses ExpectsInputTypes
(not ImplicitCastInputTypes), so a BIGINT is not cast to DECIMAL — it goes
through the dedicated IntegralType path (BigInteger.valueOf(... longValue())).
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
