We have Iceberg tables stored in OCI Object Storage with the standard structure:
table/
├── data/
└── metadata/
I found the AIDP sample notebook for working with Iceberg through Spark using a Hadoop Catalog, which is an example to support reading and writing Iceberg tables from notebooks. [github.com]
However, I have not found a way to ingest or register those Iceberg tables as native objects in the AIDP Master Catalog, and Iceberg is not listed in the supported table formats documentation. [docs.oracle.com]
Is Iceberg catalog ingestion currently supported?
If not, are there any recommended workarounds besides accessing the tables directly from Spark via an Iceberg catalog? Note: I am not looking to ingest the raw parquet only, as that would lose the critical manifest/metadata.
Thanks.
