JingsongLi commented on a change in pull request #10268:
[Flink-14599][table-planner-blink] Support precision of TimestampType in blink
planner
URL: https://github.com/apache/flink/pull/10268#discussion_r349445639
##########
File path:
flink-table/flink-table-planner-blink/src/main/java/org/apache/flink/table/planner/expressions/converter/ExpressionConverter.java
##########
@@ -135,8 +138,24 @@ public RexNode visit(ValueLiteralExpression valueLiteral)
{
return
relBuilder.getRexBuilder().makeTimeLiteral(TimeString.fromCalendarFields(
valueAsCalendar(extractValue(valueLiteral, java.sql.Time.class))), 0);
case TIMESTAMP_WITHOUT_TIME_ZONE:
- return
relBuilder.getRexBuilder().makeTimestampLiteral(TimestampString.fromCalendarFields(
-
valueAsCalendar(extractValue(valueLiteral, java.sql.Timestamp.class))), 3);
+ TimestampType timestampType = (TimestampType)
type;
+ Class<?> clazz =
valueLiteral.getOutputDataType().getConversionClass();
+ LocalDateTime datetime = null;
+ if (clazz == LocalDateTime.class) {
+ datetime = extractValue(valueLiteral,
LocalDateTime.class);
+ } else if (clazz == Timestamp.class) {
+ datetime = extractValue(valueLiteral,
Timestamp.class).toLocalDateTime();
+ } else {
+ throw new
TableException(String.format("Invalid literal of %s.",
clazz.getCanonicalName()));
+ }
+ return
relBuilder.getRexBuilder().makeTimestampLiteral(
+ new TimestampString(
Review comment:
I would say it's worth adding methods.
First of all, this transformation is not simple. It requires 7 + method
calls.
It's not a single constructor, it's easy to forget `withNanos`.
Besides, I don't think we will have nowhere to put down a method of reusing
code.
----------------------------------------------------------------
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
