snuyanzin commented on code in PR #17677:
URL: https://github.com/apache/flink/pull/17677#discussion_r864566895
##########
flink-table/flink-table-planner/src/main/scala/org/apache/flink/table/planner/codegen/calls/FloorCeilCallGen.scala:
##########
@@ -105,13 +105,27 @@ class FloorCeilCallGen(
case _ =>
operand.resultType.getTypeRoot match {
case LogicalTypeRoot.TIMESTAMP_WITHOUT_TIME_ZONE =>
- val longTerm = s"${terms.head}.getMillisecond()"
- s"""
- |$TIMESTAMP_DATA.fromEpochMillis(
- | ${qualifyMethod(arithmeticIntegralMethod.get)}(
- | $longTerm,
- | (long) ${unit.startUnit.multiplier.intValue()}))
+ val millis = s"${terms.head}.getMillisecond()"
+
+ unit match {
+ case MILLISECOND =>
+ val nanos =
+
s"${qualifyMethod(arithmeticIntegralMethod.get)}(${terms.head}.getNanoOfMillisecond(),
" +
Review Comment:
The problem is that `org.apache.flink.table.data.TimestampData` contains
these `millis`, `nanoOfMillis` in 2 separate variables
https://github.com/apache/flink/blob/90e98ba7c858b740365fc9ffdf0afc7d791373f2/flink-table/flink-table-common/src/main/java/org/apache/flink/table/data/TimestampData.java#L46-L50
.
It means that to get `nanos of second` it is required to get both `nano of
millisecond` and `millis` and then convert to `nanos of second` since there is
no corresponding methods.
Or another way is extending `org.apache.flink.table.data.TimestampData` to
provide getter of `nanos of second` and probably the ability of being
initialized by `nanos of second`
--
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]
