langchain-ai/langgraph
1
0
Fork 0
mirror of https://github.com/langchain-ai/langgraph.git synced 2026-07-15 12:35:43 -04:00
Code Issues 314 Packages Projects Releases 17 Wiki Activity

[PR #5789] fix(langgraph): support Pydantic default values with reducer functions #4647

New Issue
Closed
opened 2026-02-20 17:50:31 -05:00 by yindo · 0 comments
yindo commented 2026-02-20 17:50:31 -05:00
Owner
Copy Link
Copy Source

Original Pull Request: https://github.com/langchain-ai/langgraph/pull/5789

State: closed
Merged: No

Description: This PR fixes a bug where Pydantic Field(default=...) and Field(default_factory=...) values were not being applied when used with Annotated reducer functions in LangGraph state schemas. The issue was that default values defined in Pydantic fields were being ignored, causing state fields to start with empty values instead of the specified defaults.

The fix updates the get_field_default function to handle Pydantic models, updates the BinaryOperatorAggregate class to accept and use default values, and connects the schema context through the state management system.

Issue: Fix #5225

Dependencies: None - this is a bug fix that updates existing functionality without adding new dependencies.

Changes Made

Core Implementation:

  1. Updated get_field_default function (langgraph/_internal/_fields.py):

    • Added support for Pydantic BaseModel default value extraction
    • Handles both Field(default_factory=...) and Field(default=...)
    • Properly processes PydanticUndefinedType markers

  2. Updated BinaryOperatorAggregate class (langgraph/channels/binop.py):

    • Added default parameter to constructor
    • Modified initialization to use provided defaults instead of empty instances
    • Updated copy() and from_checkpoint() methods to preserve defaults

  3. Connected schema context (langgraph/graph/state.py):

    • Modified _is_field_binop, _get_channel, and _get_channels functions
    • Added schema context passing for proper default value resolution

Testing:

Created comprehensive test suite in tests/test_pydantic_defaults_with_reducers.py with 12 test cases:

  • 5 core functionality tests for Pydantic defaults with reducers
  • 7 backward compatibility tests ensuring existing behavior is preserved

Example Usage After Fix

from pydantic import BaseModel, Field
from typing import Annotated
import operator
from langgraph.graph import StateGraph, START, END

class State(BaseModel):
    messages: Annotated[list[str], operator.add] = Field(default_factory=lambda: ["initial"])
    count: Annotated[int, operator.add] = Field(default=0)

def my_node(state: State) -> dict:
    return {"messages": ["new"], "count": 1}

graph = StateGraph(State)
graph.add_node("process", my_node)
graph.add_edge(START, "process")
graph.add_edge("process", END)

result = graph.compile().invoke({})
# result["messages"] == ["initial", "new"]  ✅ Now works!
# result["count"] == 1  ✅ Default + reducer

Test Results

All tests pass:

$ python -m pytest tests/test_pydantic_defaults_with_reducers.py -v
=============================================== 12 passed, 12 warnings in 0.11s ================================================
**Original Pull Request:** https://github.com/langchain-ai/langgraph/pull/5789 **State:** closed **Merged:** No --- **Description:** This PR fixes a bug where Pydantic `Field(default=...)` and `Field(default_factory=...)` values were not being applied when used with `Annotated` reducer functions in LangGraph state schemas. The issue was that default values defined in Pydantic fields were being ignored, causing state fields to start with empty values instead of the specified defaults. The fix updates the `get_field_default` function to handle Pydantic models, updates the `BinaryOperatorAggregate` class to accept and use default values, and connects the schema context through the state management system. **Issue:** Fix #5225 **Dependencies:** None - this is a bug fix that updates existing functionality without adding new dependencies. ## Changes Made ### Core Implementation: 1. **Updated `get_field_default` function** (`langgraph/_internal/_fields.py`): - Added support for Pydantic BaseModel default value extraction - Handles both `Field(default_factory=...)` and `Field(default=...)` - Properly processes `PydanticUndefinedType` markers 2. **Updated `BinaryOperatorAggregate` class** (`langgraph/channels/binop.py`): - Added `default` parameter to constructor - Modified initialization to use provided defaults instead of empty instances - Updated `copy()` and `from_checkpoint()` methods to preserve defaults 3. **Connected schema context** (`langgraph/graph/state.py`): - Modified `_is_field_binop`, `_get_channel`, and `_get_channels` functions - Added schema context passing for proper default value resolution ### Testing: Created comprehensive test suite in `tests/test_pydantic_defaults_with_reducers.py` with 12 test cases: - 5 core functionality tests for Pydantic defaults with reducers - 7 backward compatibility tests ensuring existing behavior is preserved ## Example Usage After Fix ```python from pydantic import BaseModel, Field from typing import Annotated import operator from langgraph.graph import StateGraph, START, END class State(BaseModel): messages: Annotated[list[str], operator.add] = Field(default_factory=lambda: ["initial"]) count: Annotated[int, operator.add] = Field(default=0) def my_node(state: State) -> dict: return {"messages": ["new"], "count": 1} graph = StateGraph(State) graph.add_node("process", my_node) graph.add_edge(START, "process") graph.add_edge("process", END) result = graph.compile().invoke({}) # result["messages"] == ["initial", "new"] ✅ Now works! # result["count"] == 1 ✅ Default + reducer ``` ## Test Results All tests pass: ```bash $ python -m pytest tests/test_pydantic_defaults_with_reducers.py -v =============================================== 12 passed, 12 warnings in 0.11s ================================================ ```
yindo added the pull-request label 2026-02-20 17:50:31 -05:00
yindo closed this issue 2026-02-20 17:50:31 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
agents
awaiting author revision
bug
change
chat models
checkpointer
ci
code quality
code quality
dependencies
documentation
drawing
duplicate
enhancement
good first issue
help wanted
invalid
maintainer
meta
missing information
not maintained by langchain
open-swe-auto
open-swe-max
open-swe-max-auto
pending
performance
pull-request
Mirrored from GitHub Pull Request
 question
requires design
sdk
server
stale
studio
tests
type checking
usage question
v0.5
v1.1
No Label pull-request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langchain-ai/langgraph#4647
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?