MaxGekk commented on a change in pull request #25782: [SPARK-29074][SQL]
Optimize `date_format` for foldable `fmt`
URL: https://github.com/apache/spark/pull/25782#discussion_r324195916
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/datetimeExpressions.scala
##########
@@ -589,22 +589,43 @@ case class DateFormatClass(left: Expression, right:
Expression, timeZoneId: Opti
override def inputTypes: Seq[AbstractDataType] = Seq(TimestampType,
StringType)
- override def withTimeZone(timeZoneId: String): TimeZoneAwareExpression =
+ var formatter: Option[TimestampFormatter] = None
+
+ override def withTimeZone(timeZoneId: String): TimeZoneAwareExpression = {
+ if (formatter.isEmpty && right.foldable) {
+ val format = right.eval().toString
+ formatter = Some(TimestampFormatter(
+ format,
+ DateTimeUtils.getZoneId(timeZoneId),
+ Locale.US))
+ }
copy(timeZoneId = Option(timeZoneId))
+ }
override protected def nullSafeEval(timestamp: Any, format: Any): Any = {
- val df = TimestampFormatter(format.toString, zoneId)
- UTF8String.fromString(df.format(timestamp.asInstanceOf[Long]))
+ val tf = if (formatter.isEmpty) {
Review comment:
`.getOrElse` has some overhead of calling the lambda function. I explicitly
avoided its usage in the interpreted mode. For consistency, I could do the same
in the codegen function but I don't think it does matter.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
