Correctly convert partition_spec and sort_order (apache#544)

HonahX · Mar 28, 2024 · d174760 · d174760
1 parent a371077
commit d174760
Show file tree

Hide file tree

Showing 2 changed files with 15 additions and 9 deletions.
diff --git a/pyiceberg/table/metadata.py b/pyiceberg/table/metadata.py
@@ -306,6 +306,11 @@ def construct_partition_specs(cls, data: Dict[str, Any]) -> Dict[str, Any]:
                 fields = data[PARTITION_SPEC]
                 data[PARTITION_SPECS] = [{SPEC_ID: INITIAL_SPEC_ID, FIELDS: fields}]
                 data[DEFAULT_SPEC_ID] = INITIAL_SPEC_ID
+            elif data.get("partition_spec") is not None:
+                # Promote the spec from partition_spec to partition-specs
+                fields = data["partition_spec"]
+                data[PARTITION_SPECS] = [{SPEC_ID: INITIAL_SPEC_ID, FIELDS: fields}]
+                data[DEFAULT_SPEC_ID] = INITIAL_SPEC_ID
             else:
                 data[PARTITION_SPECS] = [{"field-id": 0, "fields": ()}]
 
@@ -328,7 +333,7 @@ def set_sort_orders(cls, data: Dict[str, Any]) -> Dict[str, Any]:
         Returns:
             The TableMetadata with the sort_orders set, if not provided.
         """
-        if not data.get(SORT_ORDERS):
+        if not data.get(SORT_ORDERS) and not data.get("sort_orders"):
             data[SORT_ORDERS] = [UNSORTED_SORT_ORDER]
         return data
 

diff --git a/tests/table/test_metadata.py b/tests/table/test_metadata.py
@@ -235,6 +235,11 @@ def test_new_table_metadata_with_explicit_v1_format() -> None:
 
     expected_spec = PartitionSpec(PartitionField(source_id=2, field_id=1000, transform=IdentityTransform(), name="bar"))
 
+    expected_sort_order = SortOrder(
+        SortField(source_id=1, transform=IdentityTransform(), direction=SortDirection.ASC, null_order=NullOrder.NULLS_LAST),
+        order_id=1,
+    )
+
     expected = TableMetadataV1(
         location="s3://some_v1_location/",
         table_uuid=actual.table_uuid,
@@ -252,20 +257,16 @@ def test_new_table_metadata_with_explicit_v1_format() -> None:
         snapshots=[],
         snapshot_log=[],
         metadata_log=[],
-        sort_orders=[
-            SortOrder(
-                SortField(
-                    source_id=1, transform=IdentityTransform(), direction=SortDirection.ASC, null_order=NullOrder.NULLS_LAST
-                ),
-                order_id=1,
-            )
-        ],
+        sort_orders=[expected_sort_order],
         default_sort_order_id=1,
         refs={},
         format_version=1,
     )
 
     assert actual.model_dump() == expected.model_dump()
+    assert actual.schemas == [expected_schema]
+    assert actual.partition_specs == [expected_spec]
+    assert actual.sort_orders == [expected_sort_order]
 
 
 def test_invalid_format_version(example_table_metadata_v1: Dict[str, Any]) -> None: