Use ChainMap

Author: NoahStapp

pymongo/asynchronous/client_bulk.py

@@ -21,6 +21,7 @@
 import copy
 import datetime
 import logging
+from collections import ChainMap
 from collections.abc import MutableMapping
 from itertools import islice
 from typing import (
@@ -133,10 +134,9 @@ def add_insert(self, namespace: str, document: _DocumentOut) -> None:
         validate_is_document_type("document", document)
         # Generate ObjectId client side.
         if not (isinstance(document, RawBSONDocument) or "_id" in document):
-            new_document = {"_id": ObjectId()}
-            new_document.update(document)
-            document.clear()
-            document.update(new_document)
+            document = ChainMap(document, {"_id": ObjectId()})
+        elif not isinstance(document, RawBSONDocument) and "_id" in document:
+            document = ChainMap(document, {"_id": document["_id"]})
         cmd = {"insert": -1, "document": document}
         self.ops.append(("insert", cmd))
         self.namespaces.append(namespace)

pymongo/synchronous/client_bulk.py

@@ -21,6 +21,7 @@
 import copy
 import datetime
 import logging
+from collections import ChainMap
 from collections.abc import MutableMapping
 from itertools import islice
 from typing import (
@@ -133,10 +134,9 @@ def add_insert(self, namespace: str, document: _DocumentOut) -> None:
         validate_is_document_type("document", document)
         # Generate ObjectId client side.
         if not (isinstance(document, RawBSONDocument) or "_id" in document):
-            new_document = {"_id": ObjectId()}
-            new_document.update(document)
-            document.clear()
-            document.update(new_document)
+            document = ChainMap(document, {"_id": ObjectId()})
+        elif not isinstance(document, RawBSONDocument) and "_id" in document:
+            document = ChainMap(document, {"_id": document["_id"]})
         cmd = {"insert": -1, "document": document}
         self.ops.append(("insert", cmd))
         self.namespaces.append(namespace)

test/mockupdb/test_id_ordering.py

@@ -50,3 +50,21 @@ def test_id_ordering(self):
             request = server.receives()
             self.assertEqual("_id", next(iter(request["ops"][0]["document"])))
             request.reply({"ok": 1})
+
+        # Re-ordering user-supplied _id fields is not required by the spec, but PyMongo does it for performance reasons
+        with going(collection.insert_one, {"x": 1, "_id": 111}):
+            request = server.receives()
+            self.assertEqual("_id", next(iter(request["documents"][0])))
+            request.reply({"ok": 1})
+
+        with going(collection.bulk_write, [InsertOne({"x1": 1, "_id": 1111})]):
+            request = server.receives()
+            self.assertEqual("_id", next(iter(request["documents"][0])))
+            request.reply({"ok": 1})
+
+        with going(
+            client.bulk_write, [InsertOne(namespace="db.coll", document={"x2": 1, "_id": 11111})]
+        ):
+            request = server.receives()
+            self.assertEqual("_id", next(iter(request["ops"][0]["document"])))
+            request.reply({"ok": 1})