[GitHub] [spark] MaxGekk commented on a change in pull request #33278: [WIP][SPARK-36044][SQL] Suport TimestampNTZ in functions unix_timestamp/to_unix_timestamp

[GitBox <gi...@apache.org>]

MaxGekk commented on a change in pull request #33278:
URL: https://github.com/apache/spark/pull/33278#discussion_r666821477



##########
File path: sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/datetimeExpressions.scala
##########
@@ -1091,8 +1091,13 @@ abstract class ToTimestamp
   override protected def formatString: Expression = right
   override protected def isParsing = true
 
+  override def forTimestampNTZ: Boolean = left.dataType match {
+    case TimestampNTZType => true
+    case _ => false
+  }
+

Review comment:
       Would be simpler to invoke isInstanceOf? Like
   ```scala
   override def forTimestampNTZ: Boolean = left.dataType.isInstanceOf[TimestampNTZType]
   ```

##########
File path: sql/core/src/test/scala/org/apache/spark/sql/DateFunctionsSuite.scala
##########
@@ -587,16 +587,22 @@ class DateFunctionsSuite extends QueryTest with SharedSparkSession {
         val date2 = Date.valueOf("2015-07-25")
         val ts1 = Timestamp.valueOf("2015-07-24 10:00:00.3")
         val ts2 = Timestamp.valueOf("2015-07-25 02:02:02.2")
+        val ntzTs1 = LocalDateTime.parse("2015-07-24T10:00:00.3")
+        val ntzTs2 = LocalDateTime.parse("2015-07-25T02:02:02.2")
         val s1 = "2015/07/24 10:00:00.5"
         val s2 = "2015/07/25 02:02:02.6"
         val ss1 = "2015-07-24 10:00:00"
         val ss2 = "2015-07-25 02:02:02"
         val fmt = "yyyy/MM/dd HH:mm:ss.S"
-        val df = Seq((date1, ts1, s1, ss1), (date2, ts2, s2, ss2)).toDF("d", "ts", "s", "ss")
+        val df = Seq((date1, ts1, ntzTs1, s1, ss1), (date2, ts2, ntzTs2, s2, ss2)).toDF(
+          "d", "ts", "ntzTs", "s", "ss")
         checkAnswer(df.select(unix_timestamp(col("ts"))), Seq(
           Row(secs(ts1.getTime)), Row(secs(ts2.getTime))))
         checkAnswer(df.select(unix_timestamp(col("ss"))), Seq(
           Row(secs(ts1.getTime)), Row(secs(ts2.getTime))))
+        checkAnswer(df.select(unix_timestamp(col("ntzTs"))), Seq(
+          Row(secs(DateTimeUtils.microsToMillis(DateTimeUtils.localDateTimeToMicros(ntzTs1)))),
+          Row(secs(DateTimeUtils.microsToMillis(DateTimeUtils.localDateTimeToMicros(ntzTs2))))))

Review comment:
       It would be nice to check that the type of the result is `timestamp_ntz` but not `timestamp_ltz`.




-- 
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: reviews-unsubscribe@spark.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org