From 45d611fe351f6f3847bf329aa053d890d810e2b6 Mon Sep 17 00:00:00 2001
From: Fokko Driesprong <fokko@apache.org>
Date: Thu, 31 Oct 2024 08:08:11 +0100
Subject: [PATCH] Allow for missing operation (#1263)

And then default to `overwrite`, see:
https://lists.apache.org/thread/h9qmrmlgxh91ol0y2v8olt90b9q6p9xr
---
 pyiceberg/table/snapshots.py  | 6 +++++-
 tests/table/test_snapshots.py | 7 +++++++
 2 files changed, 12 insertions(+), 1 deletion(-)

diff --git a/pyiceberg/table/snapshots.py b/pyiceberg/table/snapshots.py
index 829bd60290..c5cb57e691 100644
--- a/pyiceberg/table/snapshots.py
+++ b/pyiceberg/table/snapshots.py
@@ -17,6 +17,7 @@
 from __future__ import annotations
 
 import time
+import warnings
 from collections import defaultdict
 from enum import Enum
 from typing import TYPE_CHECKING, Any, DefaultDict, Dict, Iterable, List, Mapping, Optional
@@ -182,7 +183,10 @@ class Summary(IcebergBaseModel, Mapping[str, str]):
     operation: Operation = Field()
     _additional_properties: Dict[str, str] = PrivateAttr()
 
-    def __init__(self, operation: Operation, **data: Any) -> None:
+    def __init__(self, operation: Optional[Operation] = None, **data: Any) -> None:
+        if operation is None:
+            warnings.warn("Encountered invalid snapshot summary: operation is missing, defaulting to overwrite")
+            operation = Operation.OVERWRITE
         super().__init__(operation=operation, **data)
         self._additional_properties = data
 
diff --git a/tests/table/test_snapshots.py b/tests/table/test_snapshots.py
index ff9d92cea3..b4dde217d4 100644
--- a/tests/table/test_snapshots.py
+++ b/tests/table/test_snapshots.py
@@ -112,6 +112,13 @@ def test_deserialize_snapshot(snapshot: Snapshot) -> None:
     assert actual == snapshot
 
 
+def test_deserialize_snapshot_without_operation(snapshot: Snapshot) -> None:
+    payload = """{"snapshot-id": 25, "parent-snapshot-id": 19, "sequence-number": 200, "timestamp-ms": 1602638573590, "manifest-list": "s3:/a/b/c.avro", "summary": {}, "schema-id": 3}"""
+    with pytest.warns(UserWarning, match="Encountered invalid snapshot summary: operation is missing, defaulting to overwrite"):
+        actual = Snapshot.model_validate_json(payload)
+    assert actual.summary.operation == Operation.OVERWRITE
+
+
 def test_deserialize_snapshot_with_properties(snapshot_with_properties: Snapshot) -> None:
     payload = """{"snapshot-id":25,"parent-snapshot-id":19,"sequence-number":200,"timestamp-ms":1602638573590,"manifest-list":"s3:/a/b/c.avro","summary":{"operation":"append","foo":"bar"},"schema-id":3}"""
     snapshot = Snapshot.model_validate_json(payload)