https://github.com/apache/arrow-datafusion/pull/3976 Adds tests for applying filters during the parquet scan However, it currently uses a single file with a specific layout This ticket tracks improvements to test coverage we could/should add: - [ ] Limit row page size to ensure multiple pages are being written - [ ] Choose row page size / sort to ensure that page pruning happens / doesn't happen for multiple pages