HyukjinKwon commented on a change in pull request #28534:
URL: https://github.com/apache/spark/pull/28534#discussion_r428431767
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/analysis/FunctionRegistry.scala
##########
@@ -424,6 +424,9 @@ object FunctionRegistry {
expression[MakeInterval]("make_interval"),
expression[DatePart]("date_part"),
expression[Extract]("extract"),
+ expression[SecondsToTimestamp]("timestamp_seconds"),
+ expression[MilliSecondsToTimestamp]("timestamp_milliseconds"),
+ expression[MicroSecondsToTimestamp]("timestamp_microseconds"),
Review comment:
I think it's okay to have expressions presumably to get away from
`cast(ts as long)` behaviour which is not invasive. But it would be more
interesting to see how other DBMSes solved this problem. From arbitrary
googling, vendors have different approaches.
Seems like BigQuery has these three functions
(https://cloud.google.com/bigquery/docs/reference/standard-sql/timestamp_functions?hl=ko#timestamp_seconds).
Shall we match the naming at least?
----------------------------------------------------------------
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
