Apache XTable (incubating) is a cross-table converter for lakehouse table formats that facilitates interoperability across data processing systems and query engines.
Depending on how a Hudi table is created, the record key field can be created as a nullable field. This causes issues when we try to sync to Iceberg with this field marked as an identifier field. There are two options we can consider and test:
1) Only set the Iceberg identifier field if the field is non-null and log a warning otherwise
2) Interpret the Hudi record key field as non-null. Note: This may cause issues since the schema for the parquet file will be nullable.
Depending on how a Hudi table is created, the record key field can be created as a nullable field. This causes issues when we try to sync to Iceberg with this field marked as an identifier field. There are two options we can consider and test: 1) Only set the Iceberg identifier field if the field is non-null and log a warning otherwise 2) Interpret the Hudi record key field as non-null. Note: This may cause issues since the schema for the parquet file will be nullable.