andygrove commented on PR #3536: URL: https://github.com/apache/datafusion-comet/pull/3536#issuecomment-3910028761
not sure how true this is yet, but Claude claims that we there may be a regression related to Iceberg scans: ``` In iceberg_scan.rs, the old code cached (SchemaRef, Arc<dyn SchemaMapper>) and reused it across batches with the same schema. The new adapt_batch_with_expressions() creates a new SparkPhysicalExprAdapterFactory + adapter + expression trees for every batch that needs adaptation. This adds per-batch allocation churn, though the actual data buffers are shared via Arc. ``` -- 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]
