pola-rs/polars · error · TypeError

cannot sink to static Iceberg table

Error message

cannot sink to static Iceberg table: {type(target) = }, {getattr(target, 'catalog', None) = }

What it means

The resolved catalog is pyiceberg's NoopCatalog, a catalog that performs no persistence. Sinking to such a table would silently discard data, so Polars raises TypeError identifying the target type and its catalog attribute.

Solutions

Load the table from a real catalog (SQLCatalog, HiveCatalog, RESTCatalog, etc.) and sink to that
Pass a table identifier string plus a configured catalog= argument instead of a NoopCatalog-backed Table

Example fix

// before
table = NoopCatalog('noop').load_table('db.t')
lf.sink_iceberg(table)
// after
catalog = load_catalog('prod', **props)
lf.sink_iceberg('db.t', catalog=catalog)

Defensive patterns

Strategy: type-guard

Validate before calling

from pyiceberg.catalog.noop import NoopCatalog
if isinstance(target, Table) and isinstance(target.catalog, NoopCatalog):
    raise TypeError('target table uses NoopCatalog; cannot sink')

Type guard

def is_sinkable_table(target) -> bool:
    from pyiceberg.catalog.noop import NoopCatalog
    return not (hasattr(target, 'catalog') and isinstance(target.catalog, NoopCatalog))

Try / catch

try:
    lf.sink_iceberg(target, catalog=catalog)
except TypeError as e:
    if 'static Iceberg table' in str(e):
        target = catalog.load_table(table_identifier)
        lf.sink_iceberg(target, catalog=catalog)
    else:
        raise

Prevention

Never pass tables loaded from NoopCatalog to write operations
Verify the catalog backend (REST/Hive/SQL) before sinking
Use a string identifier plus catalog= argument in tests of real writes

When it happens

Trigger: Passing a pyiceberg Table loaded from NoopCatalog as target to sink_iceberg / IcebergSinkState.new, or constructing a table whose .catalog is a NoopCatalog (e.g. from pyiceberg static/test helpers).

Common situations: Testing code paths with NoopCatalog and then trying to write real data; loading a table via a catalog stub instead of Hive/REST/SQL catalog; forgetting to configure a real catalog.

Understand the failure class

Background: "is not a compatible type" / "cannot merge" errors: when a value's type doesn't match what the library requires — this error's family across 65 libraries.

Related errors

AI-assisted analysis of pola-rs/polars@fe841f959e (2026-09-18). Data as JSON: /api/errors/3ceb4f3581220cd3. Report an issue: GitHub.

Appendix: source

Thrown at py-polars/src/polars/io/iceberg/_sink.py:548

            )
            if isinstance(target, str)
            else (
                IcebergCatalogConfig(
                    class_=type(target.catalog),
                    name=target.catalog.name,
                    properties=target.catalog.properties,
                )
            )
        )

        from pyiceberg.catalog.noop import NoopCatalog

        if catalog_config.class_ is NoopCatalog:
            msg = (
                "cannot sink to static Iceberg table: "
                f"{type(target) = }, {getattr(target, 'catalog', None) = }"
            )
            raise TypeError(msg)

        return IcebergSinkState(
            py_catalog_class_module=catalog_config.class_.__module__,
            py_catalog_class_qualname=catalog_config.class_.__qualname__,
            catalog_name=catalog_config.name,
            catalog_properties=catalog_config.properties,
            table_name=target if isinstance(target, str) else ".".join(target.name()),
            mode=mode,
            schema_mode=schema_mode,
            snapshot_properties=snapshot_properties or {},
            iceberg_storage_properties=storage_options or {},
            compression=compression,
            compression_level=compression_level,
            row_group_size=row_group_size,
            maintain_order=maintain_order,
            sink_uuid_str=gen_uuid_v7().hex(),
            table_=NoPickleOption(target if not isinstance(target, str) else None),
            source_schema=None,

View on GitHub (pinned to fe841f959e)