RDBC-1037 RDBC-1038 Fix load_into_stream and implement stream_into/to_stream by redknightlois · Pull Request #278 · ravendb/ravendb-python-client

redknightlois · 2026-03-02T15:12:25Z

Issue link

https://issues.hibernatingrhinos.com/issue/RDBC-1037
https://issues.hibernatingrhinos.com/issue/RDBC-1038

Additional description

RDBC-1037 – session.advanced.load_into_stream() and load_starting_with_into_stream() were broken: the implementation was reading from the output stream instead of writing to it, so callers always received an empty result. Fixed to serialize documents and write them to the provided stream.

RDBC-1038 – Adds session.advanced.stream_into(query, output), which streams query results incrementally to an output stream without buffering the full result set in memory. Also adds DocumentQuery.to_stream() and RawDocumentQuery.to_stream() as convenience wrappers. The output format matches the C# StreamInto wire format ({"Results":[...]}).

Type of change

Bug fix
Regression bug fix
Optimization
New feature

How risky is the change?

Low
Moderate
High
Not relevant

Backward compatibility

Non breaking change
Ensured. Please explain how has it been implemented?
Breaking change
Not relevant

Is it platform specific issue?

Yes. Please list the affected platforms.
No

Documentation update

This change requires a documentation update. Please mark the issue on YouTrack using Documentation Required tag.
No documentation update is needed

Testing by Contributor

Tests have been added that prove the fix is effective or that the feature works
Internal classes added to the test class (e.g. entity or index definition classes) have the lowest possible access modifier (preferable private)
It has been verified by manual testing
Existing tests verify the correct behavior

Testing by RavenDB QA team

This change requires a special QA testing due to possible performance or resources usage implications (CPU, memory, IO). Please mark the issue on YouTrack using QA Required tag.
No special testing by RavenDB QA team is needed

Is there any existing behavior change of other features due to this change?

Yes. Please list the affected features/subsystems and provide appropriate explanation
No

UI work

It requires further work in the Studio. Please mark the issue on YouTrack using Studio Required tag.
No UI work is needed

poissoncorp · 2026-04-10T11:22:56Z

please bump PR to run checks

poissoncorp · 2026-04-10T11:58:56Z

ravendb/documents/session/document_session.py

        return load_operation.get_documents(object_type)

-    def _load_internal_stream(self, keys: List[str], operation: LoadOperation, stream: Optional[bytes] = None) -> None:
+    def _load_internal_stream(self, keys: List[str], operation: LoadOperation, stream=None) -> None:


please provide typehint for stream parameter
should be fixed type, Union, or Protocol

write a docstring on the use-case of this method - when to use, should provide file stream/BytesIO?

What if we use BinaryIO which is the parallel to Stream in C#?

The other usage is attachments stream which is a readable stream (or raw bytes). StoreAttachmentParameters.__init__ is untyped and PutAttachmentOperation.__init__ typed bytes. We would need to think in the public API what impact it has if we want to standarize.

ravendb/documents/session/document_session.py

ravendb/documents/session/query.py

poissoncorp · 2026-04-10T15:18:25Z

ravendb/documents/session/document_session.py


-        def stream_into(self):  # query: Union[DocumentQuery, RawDocumentQuery], output: iter):
-            pass
+        def stream_into(self, query: AbstractDocumentQuery, output: BinaryIO) -> None:


let's add docstring here

poissoncorp · 2026-04-10T15:18:33Z

ravendb/documents/session/document_session.py

@@ -404,16 +407,19 @@ def load_starting_with(
    def load_starting_with_into_stream(


let's add docstring here

poissoncorp · 2026-04-10T15:18:46Z

ravendb/documents/session/query.py

        self._suggest_using(suggestion_or_builder)
        return SuggestionDocumentQuery(self)

+    def to_stream(self, output: BinaryIO) -> None:


let's add docstring here

poissoncorp · 2026-04-10T15:18:58Z

ravendb/documents/session/query.py

        self._projection(projection_behavior)
        return self

+    def to_stream(self, output: BinaryIO) -> None:


let's add docstring here

…ssion streaming APIs - _load_internal_stream: write serialized result to output stream instead of reading from it; broaden exception handler from IOError to Exception and use proper raise-from chaining; filter None fields from to_json() output so the stream matches the server wire format (only fields the server actually sent) - load_starting_with_into_stream: add required output parameter, validate it, fix grammar error in id_prefix guard message, delegate to internal helper; filter None fields at write site - stream_into: implemented; writes {"Results":[...]} incrementally to output as JSONL items arrive — no full in-memory buffering; matches C# StreamInto format expected by JObject.Load / json.GetValue("Results") tests - DocumentQuery.to_stream / RawDocumentQuery.to_stream: convenience wrappers over session.advanced.stream_into Tests: - dotnet migrated test (RDBC-1037): load_into_stream and load_starting_with_into_stream write valid JSON to BytesIO; every integration test asserts exact count and field values - dotnet migrated test (RDBC-1038): stream_into and to_stream write Results JSON; each integration test asserts count and field values for every result; unit test verifies the method rejects non-query arguments at runtime - JVM migrated test: adds test_can_load_by_ids_into_stream (port of C# CanLoadByIdsIntoStream) alongside the existing load_starting_with test

redknightlois force-pushed the RDBC-1037 branch 2 times, most recently from 4dee4e0 to a6503a6 Compare March 6, 2026 17:07

redknightlois changed the base branch from v7.1 to v7.2 March 6, 2026 21:11