kevinjqliu commented on code in PR #2478:
URL: https://github.com/apache/iceberg-python/pull/2478#discussion_r2380748364
##########
tests/integration/test_catalog.py:
##########
@@ -343,3 +345,66 @@ def test_update_namespace_properties(test_catalog:
Catalog, database_name: str)
else:
assert k in update_report.removed
assert "updated test description" ==
test_catalog.load_namespace_properties(database_name)["comment"]
+
+
[email protected]
[email protected]("test_catalog", CATALOGS)
+def test_update_table_schema(test_catalog: Catalog, table_schema_nested:
Schema, table_name: str, database_name: str) -> None:
+ identifier = (database_name, table_name)
+ test_catalog.create_namespace(database_name)
+ table = test_catalog.create_table(identifier, table_schema_nested)
+
+ update = table.update_schema().add_column("new_col", LongType())
+ update.commit()
+
+ loaded = test_catalog.load_table(identifier)
+
+ assert loaded.schema().find_field("new_col", case_sensitive=False)
+
+
[email protected]
[email protected]("test_catalog", CATALOGS)
+def test_update_table_schema_conflict(
+ test_catalog: Catalog, table_schema_nested: Schema, table_name: str,
database_name: str
+) -> None:
+ identifier = (database_name, table_name)
+ test_catalog.create_namespace(database_name)
+ table = test_catalog.create_table(identifier, table_schema_nested)
+
+ update = table.update_schema().add_column("new_col", LongType())
+
+ # update the schema concurrently so that the original update fails
+ concurrent_table = test_catalog.load_table(identifier)
+ # The test schema is assumed to have a `bar` column that can be deleted.
+ with concurrent_table.update_schema(allow_incompatible_changes=True) as
concurrent_update:
+ concurrent_update.set_identifier_fields("foo")
+ concurrent_update.update_column("foo", required=True)
+ concurrent_update.delete_column("bar")
Review Comment:
nit: i think this test will fail if there are _any_ conflicting updates,
right?
if so, we can simplify this part
##########
tests/integration/test_catalog.py:
##########
@@ -343,3 +345,66 @@ def test_update_namespace_properties(test_catalog:
Catalog, database_name: str)
else:
assert k in update_report.removed
assert "updated test description" ==
test_catalog.load_namespace_properties(database_name)["comment"]
+
+
[email protected]
[email protected]("test_catalog", CATALOGS)
+def test_update_table_schema(test_catalog: Catalog, table_schema_nested:
Schema, table_name: str, database_name: str) -> None:
+ identifier = (database_name, table_name)
+ test_catalog.create_namespace(database_name)
+ table = test_catalog.create_table(identifier, table_schema_nested)
+
+ update = table.update_schema().add_column("new_col", LongType())
+ update.commit()
+
+ loaded = test_catalog.load_table(identifier)
+
+ assert loaded.schema().find_field("new_col", case_sensitive=False)
+
+
[email protected]
[email protected]("test_catalog", CATALOGS)
+def test_update_table_schema_conflict(
+ test_catalog: Catalog, table_schema_nested: Schema, table_name: str,
database_name: str
+) -> None:
+ identifier = (database_name, table_name)
+ test_catalog.create_namespace(database_name)
+ table = test_catalog.create_table(identifier, table_schema_nested)
+
+ update = table.update_schema().add_column("new_col", LongType())
+
+ # update the schema concurrently so that the original update fails
+ concurrent_table = test_catalog.load_table(identifier)
+ # The test schema is assumed to have a `bar` column that can be deleted.
+ with concurrent_table.update_schema(allow_incompatible_changes=True) as
concurrent_update:
+ concurrent_update.set_identifier_fields("foo")
+ concurrent_update.update_column("foo", required=True)
+ concurrent_update.delete_column("bar")
+
+ # attempt to commit the original update
+ with pytest.raises(CommitFailedException, match="Requirement failed:
current schema"):
+ update.commit()
+
+ loaded = test_catalog.load_table(identifier)
+ assert loaded.schema() == concurrent_table.schema()
+
+
[email protected]
[email protected]("test_catalog", CATALOGS)
+def test_update_table_schema_then_revert(
Review Comment:
nit: "revert" here is misleading.
after the columns are after added then removed, those field-ids will not be
reused. but we do end up with the same table schema
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
