[lake/iceberg] Support multi-bucket-key tables #4069

Search before asking

• I searched in the issues and found nothing similar.

Motivation

Fluss supports composite bucket keys (e.g., DISTRIBUTED BY (tenant_id, user_id)). Paimon handles these natively. Iceberg's integration currently throws UnsupportedOperationException for multi-key tables (restriction added in #2499).

Since Fluss already computes the composite hash and stores the result in the __bucket system column, the Iceberg partition spec can use identity(__bucket) for multi-key cases without requiring Iceberg to understand the composite hash.

Solution

Remove the bucketKeys.size() > 1 check in IcebergLakeCatalog.createPartitionSpec(). For multi-key tables, use identity(BUCKET_COLUMN_NAME) partitioning.

Willingness to contribute

• I'm willing to submit a PR!