andygrove opened a new issue, #4136: URL: https://github.com/apache/datafusion-comet/issues/4136
## Description When all struct fields are missing from a Parquet file, Spark's vectorized reader returns `struct<>` (empty struct) as the schema, but Comet's native scan returns the full schema with null values (e.g., `struct<_1:struct<_3:int,_4:bigint>>`). This causes 5 tests to fail in `ParquetIOSuite` when running Spark 4.1.1 SQL tests with Comet enabled: - `vectorized reader: missing all struct fields` - `SPARK-53535: vectorized reader: missing all struct fields, struct with complex fields` - `SPARK-53535: vectorized reader: missing all struct fields, struct with map field only` - `SPARK-53535: vectorized reader: missing all struct fields, struct with cheap map and more expensive array field` - `SPARK-54220: vectorized reader: missing all struct fields, struct with NullType only` These tests are new in Spark 4.1 (SPARK-53535, SPARK-54220). -- 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] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
