Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 21 additions & 0 deletions pyiceberg/table/update/__init__.py
Original file line number Diff line number Diff line change
Expand Up @@ -193,6 +193,11 @@ class RemoveStatisticsUpdate(IcebergBaseModel):
snapshot_id: int = Field(alias="snapshot-id")


class RemovePartitionSpecsUpdate(IcebergBaseModel):
action: Literal["remove-partition-specs"] = Field(default="remove-partition-specs")
spec_ids: List[int] = Field(alias="spec-ids")


class RemoveSchemasUpdate(IcebergBaseModel):
action: Literal["remove-schemas"] = Field(default="remove-schemas")
schema_ids: List[int] = Field(alias="schema-ids")
Expand Down Expand Up @@ -227,6 +232,7 @@ class RemovePartitionStatisticsUpdate(IcebergBaseModel):
RemovePropertiesUpdate,
SetStatisticsUpdate,
RemoveStatisticsUpdate,
RemovePartitionSpecsUpdate,
RemoveSchemasUpdate,
SetPartitionStatisticsUpdate,
RemovePartitionStatisticsUpdate,
Expand Down Expand Up @@ -595,6 +601,21 @@ def _(update: RemoveStatisticsUpdate, base_metadata: TableMetadata, context: _Ta
return base_metadata.model_copy(update={"statistics": statistics})


@_apply_table_update.register(RemovePartitionSpecsUpdate)
def _(update: RemovePartitionSpecsUpdate, base_metadata: TableMetadata, context: _TableMetadataUpdateContext) -> TableMetadata:
for remove_spec_id in update.spec_ids:
if not any(spec.spec_id == remove_spec_id for spec in base_metadata.partition_specs):
raise ValueError(f"Partition spec with id {remove_spec_id} does not exist")

if base_metadata.default_spec_id in update.spec_ids:
raise ValueError(f"Cannot remove default partition spec: {base_metadata.default_spec_id}")

partition_specs = [spec for spec in base_metadata.partition_specs if spec.spec_id not in update.spec_ids]

context.add_update(update)
return base_metadata.model_copy(update={"partition_specs": partition_specs})


@_apply_table_update.register(RemoveSchemasUpdate)
def _(update: RemoveSchemasUpdate, base_metadata: TableMetadata, context: _TableMetadataUpdateContext) -> TableMetadata:
# This method should error if any schemas do not exist.
Expand Down
34 changes: 34 additions & 0 deletions tests/table/test_init.py
Original file line number Diff line number Diff line change
Expand Up @@ -66,6 +66,7 @@
)
from pyiceberg.table.statistics import BlobMetadata, PartitionStatisticsFile, StatisticsFile
from pyiceberg.table.update import (
AddPartitionSpecUpdate,
AddSnapshotUpdate,
AddSortOrderUpdate,
AssertCreate,
Expand All @@ -76,6 +77,7 @@
AssertLastAssignedPartitionId,
AssertRefSnapshotId,
AssertTableUUID,
RemovePartitionSpecsUpdate,
RemovePartitionStatisticsUpdate,
RemovePropertiesUpdate,
RemoveSchemasUpdate,
Expand Down Expand Up @@ -1294,6 +1296,38 @@ def test_update_metadata_log_overflow(table_v2: Table) -> None:
assert len(new_metadata.metadata_log) == 1


def test_remove_partition_spec_update(table_v2: Table) -> None:
base_metadata = table_v2.metadata
new_spec = PartitionSpec(PartitionField(source_id=2, field_id=1001, transform=IdentityTransform(), name="y"), spec_id=1)
metadata_with_new_spec = update_table_metadata(base_metadata, (AddPartitionSpecUpdate(spec=new_spec),))

assert len(metadata_with_new_spec.partition_specs) == 2

update = RemovePartitionSpecsUpdate(spec_ids=[1])
updated_metadata = update_table_metadata(
metadata_with_new_spec,
(update,),
)

assert len(updated_metadata.partition_specs) == 1


def test_remove_partition_spec_update_spec_does_not_exist(table_v2: Table) -> None:
update = RemovePartitionSpecsUpdate(
spec_ids=[123],
)
with pytest.raises(ValueError, match="Partition spec with id 123 does not exist"):
update_table_metadata(table_v2.metadata, (update,))


def test_remove_partition_spec_update_default_spec(table_v2: Table) -> None:
update = RemovePartitionSpecsUpdate(
spec_ids=[0],
)
with pytest.raises(ValueError, match="Cannot remove default partition spec: 0"):
update_table_metadata(table_v2.metadata, (update,))


def test_remove_schemas_update(table_v2: Table) -> None:
base_metadata = table_v2.metadata
assert len(base_metadata.schemas) == 2
Expand Down