rdblue commented on a change in pull request #3677:
URL: https://github.com/apache/iceberg/pull/3677#discussion_r763317445



##########
File path: python/tests/table/test_metadata.py
##########
@@ -0,0 +1,155 @@
+import io
+import json
+from unittest.mock import MagicMock, patch
+
+import pytest
+from jsonschema.exceptions import ValidationError
+
+from iceberg.table.metadata import TableMetadata
+
+EXAMPLE_TABLE_METADATA_V1 = {
+    "format-version": "1",
+    "table-uuid": "foo-table-uuid",
+    "location": "s3://foo/bar/baz.metadata.json",
+    "last-sequence-number": 3,
+    "last-updated-ms": 1600000000000,
+    "last-column-id": 4,
+    "schemas": [],
+    "current-schema-id": 5,
+    "partition-spec": [],
+    "partition-specs": [],
+    "default-spec-id": 6,
+    "last-partition-id": 7,
+    "properties": {"read.split.target-size": 134217728},
+    "current-snapshot-id": 1.1,
+    "snapshots": [],
+    "snapshot-log": [],
+    "metadata-log": [],
+    "sort-orders": [],
+    "default-sort-order-id": 8,
+}
+EXAMPLE_TABLE_METADATA_V2 = {
+    "format-version": "2",
+    "table-uuid": "foo-table-uuid",
+    "location": "s3://foo/bar/baz.metadata.json",
+    "last-sequence-number": 3,
+    "last-updated-ms": 1600000000000,
+    "last-column-id": 4,
+    "schemas": [],
+    "current-schema-id": 5,
+    "partition-spec": [],
+    "partition-specs": [],
+    "default-spec-id": 6,
+    "last-partition-id": 7,
+    "properties": {"read.split.target-size": 134217728},
+    "current-snapshot-id": 1.1,
+    "snapshots": [],
+    "snapshot-log": [],
+    "metadata-log": [],
+    "sort-orders": [],
+    "default-sort-order-id": 8,
+}
+
+
[email protected](
+    "metadata, version",
+    [
+        (
+            EXAMPLE_TABLE_METADATA_V1,
+            1,
+        ),
+        (
+            EXAMPLE_TABLE_METADATA_V2,
+            2,
+        ),
+    ],
+)
+def test_from_dict(metadata, version):
+    """Test standard initialization of a TableMetadata instance from a 
dictionary"""
+    table_metadata = TableMetadata(metadata=metadata, version=version)
+    table_metadata.validate()
+
+
+def test_invalid_v1_metadata():
+    """Test raising a ValidationError on invalid v1 metadata"""
+    with pytest.raises(ValidationError):
+        TableMetadata.validate_v1(metadata={"foo": "bar"})
+
+
+def test_invalid_v2_metadata():
+    """Test raising a ValidationError on invalid v2 metadata"""
+    with pytest.raises(ValidationError):
+        TableMetadata.validate_v2(metadata={"foo": "bar"})
+
+
+def test_unknown_table_metadata_version():
+    """Test raising a ValueError when an invalid table metadata version is 
provided"""
+    with pytest.raises(ValueError) as exc_info:
+        TableMetadata(metadata=EXAMPLE_TABLE_METADATA_V2, version=3).validate()
+    assert "Unknown table metadata version 3" in str(exc_info)
+
+
+def test_from_byte_stream():
+    """Test generating a TableMetadata instance from a file-like byte stream"""
+    data = bytes(json.dumps(EXAMPLE_TABLE_METADATA_V2), encoding="utf-8")
+    byte_stream = io.BytesIO(data)
+    TableMetadata.from_byte_stream(byte_stream, version=1).validate()
+
+
+@patch("iceberg.table.metadata.boto3")
+def test_from_s3(mock_boto3):
+    """Test generating a TableMetadata instance from a metadata file on S3"""
+    data = bytes(json.dumps(EXAMPLE_TABLE_METADATA_V2), encoding="utf-8")
+    byte_stream = io.BytesIO(data)
+
+    mock_s3_client = MagicMock()
+    mock_s3_client.get_object.return_value = {"Body": byte_stream}
+
+    mock_boto3_session = MagicMock()
+    mock_boto3_session.client.return_value = mock_s3_client
+
+    mock_boto3.Session.return_value = mock_boto3_session
+
+    TableMetadata.from_s3("s3://foo/bar/metadata.json", version=2).validate()
+
+
+def test_metadata_parsing():
+    """Test retrieveing values from TableMetadata instance"""
+    table_metadata = TableMetadata(metadata=EXAMPLE_TABLE_METADATA_V2, 
version=2)
+    table_metadata.validate()
+
+    assert table_metadata.format_version == "2"
+    assert table_metadata.table_uuid == "foo-table-uuid"
+    assert table_metadata.location == "s3://foo/bar/baz.metadata.json"
+    assert table_metadata.last_sequence_number == 3
+    assert table_metadata.last_updated_ms == 1600000000000
+    assert table_metadata.last_column_id == 4
+    assert table_metadata.schemas == []
+    assert table_metadata.current_schema_id == 5
+    assert table_metadata.partition_spec == []
+    assert table_metadata.partition_specs == []
+    assert table_metadata.default_spec_id == 6
+    assert table_metadata.last_partition_id == 7
+    assert table_metadata.properties.read_split_target_size == 134217728
+    assert table_metadata.current_snapshot_id == 1.1
+    assert table_metadata.snapshots == []
+    assert table_metadata.snapshot_log == []
+    assert table_metadata.metadata_log == []
+    assert table_metadata.sort_orders == []
+    assert table_metadata.default_sort_order_id == 8
+
+    assert table_metadata.properties._metadata == {"read.split.target-size": 
134217728}

Review comment:
       This test looks good to me.




-- 
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]

Reply via email to