uros-db commented on code in PR #52063:
URL: https://github.com/apache/spark/pull/52063#discussion_r2306869522
##########
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/expressions/datetimeExpressions.scala:
##########
@@ -2808,14 +2814,26 @@ object MakeTimestampLTZExpressionBuilder extends
ExpressionBuilder {
NULL
> SELECT _FUNC_(2024, 13, 22, 15, 30, 0);
NULL
+ > SELECT _FUNC_(DATE'2014-12-28', TIME'6:30:45.887');
+ 2014-12-28 06:30:45.887
+ > SELECT _FUNC_(DATE'2014-12-28', TIME'6:30:45.887', 'CET');
+ 2014-12-27 21:30:45.887
""",
group = "datetime_funcs",
since = "4.0.0")
// scalastyle:on line.size.limit
object TryMakeTimestampLTZExpressionBuilder extends ExpressionBuilder {
override def build(funcName: String, expressions: Seq[Expression]):
Expression = {
val numArgs = expressions.length
- if (numArgs == 6 || numArgs == 7) {
+ if (numArgs == 2 || numArgs == 3) {
+ // Overload for: date, time[, timezone].
+ MakeTimestampFromDateTime(
+ expressions(0),
+ Some(expressions(1)),
+ expressions.drop(2).lastOption
+ )
Review Comment:
I don't think this would provide any additional value for the date/time
constructor. If either date or time is invalid, that means that either the
underlying input data is corrupt or that we get a parse exception for invalid
input literals.
That said, since the results are the same anyway, it shouldn't hurt to
follow the same approach as `TryMakeTimestampNTZ` and indeed use `TryEval`
here. I updated the PR accordingly.
--
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]
