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 #6634] feat(langgraph): support Pydantic state with aliased fields #5182

New Issue
Open
opened 2026-02-20 17:51:20 -05:00 by yindo · 0 comments
yindo commented 2026-02-20 17:51:20 -05:00
Owner
Copy Link
Copy Source

Original Pull Request: https://github.com/langchain-ai/langgraph/pull/6634

State: open
Merged: No

Summary

Adds support for Pydantic BaseModel state schemas with aliased fields (Field(alias=...)).

Problem

When using Pydantic models with aliased fields, users couldn't provide input using alias keys:

class State(BaseModel):
    foo: str = Field(alias='bar')

graph = StateGraph(State).compile()
graph.invoke({'bar': 'hello'})  # Failed - alias not recognized

The issue was that channels are named by field names (e.g., foo), but input came with alias keys (e.g., bar), so map_input couldn't match them.

Solution

  1. Extract alias-to-field-name mappings from Pydantic schemas via _get_pydantic_alias_map()
  2. Translate input alias keys to field names in the START node's _get_updates() closure
  3. Translate field names back to aliases when instantiating Pydantic models via _coerce_state() (since Pydantic expects aliases for aliased fields)

Changes

  • langgraph/graph/state.py

    • Added _get_pydantic_alias_map() helper function
    • Modified attach_node() to translate alias keys for START nodes
    • Modified _coerce_state() to translate field names to aliases for Pydantic instantiation

  • langgraph/pregel/_io.py

    • Extended map_input() with optional alias_map parameter

  • langgraph/pregel/main.py and langgraph/pregel/_loop.py

    • Added input_alias_map attribute for future extensibility

Test plan

  • test_pydantic_aliased_fields - Basic aliased field test
  • test_pydantic_multiple_aliased_fields - Multiple aliased fields (camelCase)
  • test_pydantic_mixed_alias_and_fieldname - Mix of aliased and non-aliased fields
  • test_pydantic_alias_with_reducer - Aliased fields with reducer annotations
  • test_pydantic_aliased_fields_multinode_with_checkpointer - Integration test with multi-node graph, checkpointing, and interrupt/resume
  • test_pydantic_aliased_fields_async - Async graph execution
  • test_get_pydantic_alias_map - Unit test for alias extraction
TEST="tests/test_pydantic.py tests/test_state.py::test_get_pydantic_alias_map -v" make test
# Result: 10 passed

TEST="tests/test_pregel.py -v -k 'not stress'" make test
# Result: 420 passed

Fixes #2555

**Original Pull Request:** https://github.com/langchain-ai/langgraph/pull/6634 **State:** open **Merged:** No --- ## Summary Adds support for Pydantic `BaseModel` state schemas with aliased fields (`Field(alias=...)`). ### Problem When using Pydantic models with aliased fields, users couldn't provide input using alias keys: ```python class State(BaseModel): foo: str = Field(alias='bar') graph = StateGraph(State).compile() graph.invoke({'bar': 'hello'}) # Failed - alias not recognized ``` The issue was that channels are named by field names (e.g., `foo`), but input came with alias keys (e.g., `bar`), so `map_input` couldn't match them. ### Solution 1. Extract alias-to-field-name mappings from Pydantic schemas via `_get_pydantic_alias_map()` 2. Translate input alias keys to field names in the START node's `_get_updates()` closure 3. Translate field names back to aliases when instantiating Pydantic models via `_coerce_state()` (since Pydantic expects aliases for aliased fields) ### Changes - **`langgraph/graph/state.py`** - Added `_get_pydantic_alias_map()` helper function - Modified `attach_node()` to translate alias keys for START nodes - Modified `_coerce_state()` to translate field names to aliases for Pydantic instantiation - **`langgraph/pregel/_io.py`** - Extended `map_input()` with optional `alias_map` parameter - **`langgraph/pregel/main.py`** and **`langgraph/pregel/_loop.py`** - Added `input_alias_map` attribute for future extensibility ## Test plan - [x] `test_pydantic_aliased_fields` - Basic aliased field test - [x] `test_pydantic_multiple_aliased_fields` - Multiple aliased fields (camelCase) - [x] `test_pydantic_mixed_alias_and_fieldname` - Mix of aliased and non-aliased fields - [x] `test_pydantic_alias_with_reducer` - Aliased fields with reducer annotations - [x] `test_pydantic_aliased_fields_multinode_with_checkpointer` - Integration test with multi-node graph, checkpointing, and interrupt/resume - [x] `test_pydantic_aliased_fields_async` - Async graph execution - [x] `test_get_pydantic_alias_map` - Unit test for alias extraction ```bash TEST="tests/test_pydantic.py tests/test_state.py::test_get_pydantic_alias_map -v" make test # Result: 10 passed TEST="tests/test_pregel.py -v -k 'not stress'" make test # Result: 420 passed ``` Fixes #2555
yindo added the pull-request label 2026-02-20 17:51:20 -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#5182
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?