andygrove commented on issue #38: URL: https://github.com/apache/datafusion-iceberg/issues/38#issuecomment-5934703758
Folding #43 into this issue, since the fix overlaps. With `write.datafusion.fanout.enabled=false`, unpartitioned tables fail too, earlier: `insert_into` calls `sort_by_partition` unconditionally, and unpartitioned inputs have no `_partition` column. ```sql -- t(id INT NOT NULL), unpartitioned, with table property write.datafusion.fanout.enabled=false INSERT INTO t VALUES (1); ``` ``` Error during planning: Partition column '_partition' not found in schema. Ensure the plan has been extended with partition values using project_with_partition. ``` Once `IcebergWriteExec` declares the ordering requirement, the manual `sort_by_partition` call in `insert_into` can go, which removes this failure too. The requirement should only be declared for partitioned tables with fanout disabled. A test that inserts into an unpartitioned table with the property set would cover it. -- 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]
