lst-codes commented on pull request #1392: URL: https://github.com/apache/arrow-datafusion/pull/1392#issuecomment-989029092
> Thanks @andrei-ionescu for taking a stab at this. I don't think we can just flatting all the columns like how it's handled in parquet because the rest of the datafusion code base assumes arrow's hierarchical column structure. As a result, column indexes only map to top level columns. I suspect this approach will not cause runtime crashes because we are only expanding the column vector, but it could lead to incorrect read of column stats in places like > > https://github.com/apache/arrow-datafusion/blob/069449f754c92765a2d0dbbf62cd7bac76257892/datafusion/src/physical_optimizer/aggregate_statistics.rs#L181 @houqp Do you think column statistics should include something similar to `Field#fields()`, which returns a list of subfields for nested structures? https://github.com/apache/arrow-rs/blob/master/arrow/src/datatypes/field.rs#L110-L125 -- 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: github-unsubscr...@arrow.apache.org For queries about this service, please contact Infrastructure at: us...@infra.apache.org
