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 #6381] fix(langgraph): merge CONF sections and preserve runtime in subgraphs #5005

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

Original Pull Request: https://github.com/langchain-ai/langgraph/pull/6381

State: closed
Merged: No

Description:

Fixes issue where runtime (store and context) is lost when manually invoking a subgraph with a config parameter.

Users need to pass custom configs when invoking subgraphs (e.g., to set thread_id for checkpointing). However, this caused the runtime configuration to be overwritten instead of merged, losing access to store and context.

The fix changes ensure_config() to merge CONF sections instead of replacing them.

Minimum reproducible example:

from dataclasses import dataclass
from langchain_core.runnables import RunnableConfig
from langgraph.graph.state import StateGraph
from langgraph.runtime import Runtime
from typing_extensions import TypedDict
from langgraph.store.memory import InMemoryStore

@dataclass
class Context:
    username: str

class State(TypedDict):
    foo: str

# Subgraph
def subgraph_node_1(state: State, runtime: Runtime[Context]):
    assert runtime.store is not None, "Store is required"
    assert runtime.context is not None, "Context is required"
    username = state['foo']
    runtime.store.put(('subgraph', '1'), 'foo', {'value': 'hi! ' + username})
    return {'foo': 'hi! ' + username}

subgraph_builder = StateGraph(State, context_schema=Context)
subgraph_builder.add_node(subgraph_node_1)
subgraph_builder.set_entry_point('subgraph_node_1')
subgraph = subgraph_builder.compile()

# Parent graph
def main_node(state: State, runtime: Runtime[Context]):
    last_foo = runtime.store.get(('subgraph', '1'), 'foo')
    if last_foo:
        last_foo = last_foo.value['value']
    else:
        last_foo = runtime.context.username
    return {'foo': 'hello ' + str(last_foo)}

def invoke_subgraph(state: State, runtime: Runtime[Context], config: RunnableConfig):
    new_config = {
        **config,
        'configurable': {
            **config.get('configurable', {}),
            'thread_id': '1'
        },
    }
    return subgraph.invoke(input=state, config=new_config)

store = InMemoryStore()
builder = StateGraph(State, context_schema=Context)
builder.add_node(main_node)
builder.add_node(invoke_subgraph)
builder.set_entry_point('main_node')
builder.add_edge('main_node', 'invoke_subgraph')
graph = builder.compile(store=store)

context = Context(username='Alice')
result = graph.invoke(input={'foo': 'world'}, context=context)
print(result)  # Before fix: AssertionError. After fix: {'foo': 'hi! hello Alice'}
**Original Pull Request:** https://github.com/langchain-ai/langgraph/pull/6381 **State:** closed **Merged:** No --- **Description:** Fixes issue where runtime (store and context) is lost when manually invoking a subgraph with a config parameter. Users need to pass custom configs when invoking subgraphs (e.g., to set `thread_id` for checkpointing). However, this caused the runtime configuration to be overwritten instead of merged, losing access to store and context. The fix changes `ensure_config()` to merge CONF sections instead of replacing them. **Minimum reproducible example:** ```python from dataclasses import dataclass from langchain_core.runnables import RunnableConfig from langgraph.graph.state import StateGraph from langgraph.runtime import Runtime from typing_extensions import TypedDict from langgraph.store.memory import InMemoryStore @dataclass class Context: username: str class State(TypedDict): foo: str # Subgraph def subgraph_node_1(state: State, runtime: Runtime[Context]): assert runtime.store is not None, "Store is required" assert runtime.context is not None, "Context is required" username = state['foo'] runtime.store.put(('subgraph', '1'), 'foo', {'value': 'hi! ' + username}) return {'foo': 'hi! ' + username} subgraph_builder = StateGraph(State, context_schema=Context) subgraph_builder.add_node(subgraph_node_1) subgraph_builder.set_entry_point('subgraph_node_1') subgraph = subgraph_builder.compile() # Parent graph def main_node(state: State, runtime: Runtime[Context]): last_foo = runtime.store.get(('subgraph', '1'), 'foo') if last_foo: last_foo = last_foo.value['value'] else: last_foo = runtime.context.username return {'foo': 'hello ' + str(last_foo)} def invoke_subgraph(state: State, runtime: Runtime[Context], config: RunnableConfig): new_config = { **config, 'configurable': { **config.get('configurable', {}), 'thread_id': '1' }, } return subgraph.invoke(input=state, config=new_config) store = InMemoryStore() builder = StateGraph(State, context_schema=Context) builder.add_node(main_node) builder.add_node(invoke_subgraph) builder.set_entry_point('main_node') builder.add_edge('main_node', 'invoke_subgraph') graph = builder.compile(store=store) context = Context(username='Alice') result = graph.invoke(input={'foo': 'world'}, context=context) print(result) # Before fix: AssertionError. After fix: {'foo': 'hi! hello Alice'} ```
yindo added the pull-request label 2026-02-20 17:51:05 -05:00
yindo closed this issue 2026-02-20 17:51:05 -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#5005
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?