Skip to content
Merged
5 changes: 5 additions & 0 deletions pymongo/_client_bulk_shared.py
Original file line number Diff line number Diff line change
Expand Up @@ -16,6 +16,7 @@
"""Constants, types, and classes shared across Client Bulk Write API implementations."""
from __future__ import annotations

from collections import ChainMap
from typing import TYPE_CHECKING, Any, Mapping, MutableMapping, NoReturn

from pymongo.errors import ClientBulkWriteException, OperationFailure
Expand Down Expand Up @@ -63,6 +64,10 @@ def _throw_client_bulk_write_exception(
"""Raise a ClientBulkWriteException from the full result."""
# retryWrites on MMAPv1 should raise an actionable error.
if full_result["writeErrors"]:
# Unpack ChainMaps into the original document only
for doc in full_result["writeErrors"]:
if "document" in doc["op"] and isinstance(doc["op"]["document"], ChainMap):
doc["op"]["document"] = doc["op"]["document"].maps[0]
full_result["writeErrors"].sort(key=lambda error: error["idx"])
err = full_result["writeErrors"][0]
code = err["code"]
Expand Down
7 changes: 7 additions & 0 deletions pymongo/message.py
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@
import datetime
import random
import struct
from collections import ChainMap
from io import BytesIO as _BytesIO
from typing import (
TYPE_CHECKING,
Expand Down Expand Up @@ -1111,6 +1112,12 @@ def _check_doc_size_limits(
# key and the index of its namespace within ns_info as its value.
op_doc[op_type] = ns_info[namespace] # type: ignore[index]

# Since the data document itself is nested within the insert document
# it won't be automatically re-ordered by the BSON conversion.
# We use ChainMap here to make the _id field the first field instead.
if real_op_type == "insert":
op_doc["document"] = ChainMap(op_doc["document"], {"_id": op_doc["document"]["_id"]}) # type: ignore[index]
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The only reason we need to unwrap ChainMap later is because we're mutating op_doc. Instead we can do this:

        doc_to_encode = op_doc
        if real_op_type == "insert":
            doc = op_doc["document"]
            if not isinstance(doc, RawBSONDocument):
                doc_to_encode = op_doc.copy()  # Shallow copy
                doc_to_encode["document"] = ChainMap(doc, {"_id": doc["_id"]})  # type: ignore[index]
        op_doc_encoded = _dict_to_bson(doc_to_encode, False, opts)

We also still need the RawBSONDocument here. We should add a test to ensure RawBSONDocument is not inflated after a call to bulk_write.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ah, I see! That's a slick solution.


# Encode current operation doc and, if newly added, namespace doc.
op_doc_encoded = _dict_to_bson(op_doc, False, opts)
op_length = len(op_doc_encoded)
Expand Down
7 changes: 6 additions & 1 deletion pymongo/monitoring.py
Original file line number Diff line number Diff line change
Expand Up @@ -189,7 +189,7 @@ def connection_checked_in(self, event):
from __future__ import annotations

import datetime
from collections import abc, namedtuple
from collections import ChainMap, abc, namedtuple
from typing import TYPE_CHECKING, Any, Mapping, Optional, Sequence

from bson.objectid import ObjectId
Expand Down Expand Up @@ -625,6 +625,11 @@ def __init__(
raise ValueError(f"{command!r} is not a valid command")
# Command name must be first key.
command_name = next(iter(command))
# Unpack ChainMaps into the original document only
if command_name == "bulkWrite" and "ops" in command:
for doc in command["ops"]:
if "document" in doc and isinstance(doc["document"], ChainMap):
doc["document"] = doc["document"].maps[0]
super().__init__(
command_name,
request_id,
Expand Down
93 changes: 93 additions & 0 deletions test/mockupdb/test_id_ordering.py
Original file line number Diff line number Diff line change
@@ -0,0 +1,93 @@
# Copyright 2024-present MongoDB, Inc.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
# http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

from __future__ import annotations
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let's add the boilerplate License comment.


from test import PyMongoTestCase

import pytest

from pymongo import InsertOne

try:
from mockupdb import MockupDB, OpMsg, go, going

_HAVE_MOCKUPDB = True
except ImportError:
_HAVE_MOCKUPDB = False


from bson.objectid import ObjectId

pytestmark = pytest.mark.mockupdb


class TestIdOrdering(PyMongoTestCase):
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Can you add a link to the crud spec that describes this test?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Once the spec is merged, yes.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Was the spec merged?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes. added!

def test_id_ordering(self):
server = MockupDB()
server.autoresponds(
"hello",
isWritablePrimary=True,
msg="isdbgrid",
minWireVersion=0,
maxWireVersion=25,
helloOk=True,
serviceId=ObjectId(),
)
server.run()
self.addCleanup(server.stop)

# We also verify that the original document contains an _id field after each insert
document = {"x": 1}

client = self.simple_client(server.uri, loadBalanced=True)
collection = client.db.coll
with going(collection.insert_one, document):
request = server.receives()
self.assertEqual("_id", next(iter(request["documents"][0])))
request.reply({"ok": 1})
self.assertIn("_id", document)

document = {"x1": 1}

with going(collection.bulk_write, [InsertOne(document)]):
request = server.receives()
self.assertEqual("_id", next(iter(request["documents"][0])))
request.reply({"ok": 1})
self.assertIn("_id", document)

document = {"x2": 1}
with going(client.bulk_write, [InsertOne(namespace="db.coll", document=document)]):
request = server.receives()
self.assertEqual("_id", next(iter(request["ops"][0]["document"])))
request.reply({"ok": 1})
self.assertIn("_id", document)

# 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})
Loading