lirui-apache commented on pull request #14336: URL: https://github.com/apache/flink/pull/14336#issuecomment-767419287
@wuchong Thanks for your review. I've updated to address your comment. And after some more thoughts, I added logic in `HiveScalarFunction` to reuse previous argument types during type inference. `HiveScalarFunction` should only use `STRING` type for real literal arguments. However, calcite converts `CAST(literal)` to `literal` for RexNode, which means hive function cannot differentiate the two cases on RelNode level. So in that case, `HiveScalarFunction` should reuse the previous types and not force `STRING` type for literals. ---------------------------------------------------------------- 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]
