weimingdiit opened a new issue, #2425:
URL: https://github.com/apache/auron/issues/2425

   **Is your feature request related to a problem? Please describe.**
   
   Auron Iceberg native scan converts supported Iceberg residual filters into 
native scan pruning predicates.
   
   However, Iceberg `STARTS_WITH` filters are currently not converted by 
`IcebergScanSupport`. For prefix string predicates such as `LIKE 'a%'`, the 
query can still be executed with a native post-scan filter, but the equivalent 
starts-with predicate is not passed into native scan pruning.
   
   **Describe the solution you'd like**
   
   Convert Iceberg `STARTS_WITH` predicates on string columns to Spark 
`StartsWith` expressions before passing them to 
`NativeConverters.convertScanPruningExpr`.
   
   This keeps the change narrow and reuses the existing native pruning-mode 
`starts_with` scalar function support.
   
   **Describe alternatives you've considered**
   
   One alternative is to enable more string pruning predicates, such as string 
equality or string IN, at the same time.
   
   This PR intentionally does not do that, to avoid broadening string pruning 
semantics beyond the existing `STARTS_WITH` native support.
   
   **Additional context**
   
   Auron already supports Spark `StartsWith` in scan-pruning expression 
conversion. The missing piece is the Iceberg `STARTS_WITH` to Spark 
`StartsWith` mapping.


-- 
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]

Reply via email to