nandorKollar commented on PR #13976: URL: https://github.com/apache/iceberg/pull/13976#issuecomment-3246710475
> > > I think this makes sense to me. I wonder if we should just ignore the root allocator? As long as we close all the Childs we allocate we should be safe > > > > > > Currently, we use root allocator not just in a single place, but also [here](https://github.com/apache/iceberg/blob/main/arrow/src/main/java/org/apache/iceberg/arrow/vectorized/VectorizedArrowReader.java#L856) and [here](https://github.com/apache/iceberg/blob/main/arrow/src/main/java/org/apache/iceberg/arrow/vectorized/VectorizedArrowReader.java#L672) (the two methods are essentially the same, so I'll try to merge them into one). Will see how I can refactor there to use a child allocator. > > Ah yeah we have the places where we need a vector, but it's not part of the actual schema for virtual columns (you are looking at the _last_updated_sequence_number and _row_identifier columns) We should use in both of these places a child allocator no? Or maybe root allocator is also fine, but in that case it is also important to close that too at the right time? -- 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]
