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 #5258] Add checkpoint memgraph #4333

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

Original Pull Request: https://github.com/langchain-ai/langgraph/pull/5258

State: closed
Merged: No

feat: Add checkpoint-memgraph Package

Description

This pull request introduces the langgraph-checkpoint-memgraph package, providing a robust and feature-rich integration with Memgraph as a backend for both a key-value store and a checkpoint saver. This implementation allows LangGraph applications to persist and manage state within a Memgraph graph database.

The package includes two primary components:

  1. MemgraphStore: A synchronous, BaseStore-compliant client for general-purpose data storage in Memgraph.
  2. MemgraphSaver / AsyncMemgraphSaver: A BaseCheckpointSaver-compliant implementation for persisting LangGraph checkpoints.

Key Features

MemgraphStore

The MemgraphStore provides a comprehensive interface for interacting with Memgraph.

  • Unified Interface: Supports key-value storage, document search, and vector similarity search through a single class.
  • TTL (Time-To-Live) Support: Nodes can be set to expire automatically. A background sweeper thread (start_ttl_sweeper) can be enabled to periodically clean up expired data, ensuring efficient database management.
  • Vector Search: Integrates with embedding models to store and perform similarity searches on text data. Configuration is handled via the MemgraphIndexConfig.
  • Robust Connection Management: Can be instantiated directly with a memgraph.Driver instance or conveniently from a connection string using the from_conn_string class method.
  • Idempotent Setup: Includes a setup() method to create necessary database constraints and indexes, simplifying initial configuration.

MemgraphSaver & AsyncMemgraphSaver

The checkpoint saver provides a complete checkpointing solution built directly on Memgraph.

  • Full Checkpoint Lifecycle: Implements all methods required by the BaseCheckpointSaver interface, including get_tuple, put, list, and delete_thread.
  • Async and Sync Support: Provides both MemgraphSaver for synchronous operations and AsyncMemgraphSaver for asynchronous applications, ensuring seamless integration into any LangGraph workflow.
  • Transactional Integrity: Leverages Memgraph's transactional capabilities to ensure that checkpoint operations are atomic.

Implementation Details

  • The package provides two distinct, parallel implementations: MemgraphStore for generic key-value operations and MemgraphSaver for managing LangGraph checkpoints.
  • MemgraphStore persists data in (:StoreItem) nodes, offering a flat key-value structure with optional TTL and vector indexing.
  • MemgraphSaver implements a graph model where checkpoints are grouped by a shared thread_id property. It creates (:Checkpoint) nodes to store the state of each run, and the parent_checkpoint_id property on these nodes creates a sequential history. Associated data is stored in (:Blob) and (:Write) nodes, which are also linked by the thread_id.
  • Helper modules (_internal.py, _ainternal.py) provide shared connection and session management logic for both sync and async operations.
  • Checkpoints are serialized to JSON before being stored as a property on the :Checkpoint node, allowing for efficient storage and retrieval of complex state objects.
**Original Pull Request:** https://github.com/langchain-ai/langgraph/pull/5258 **State:** closed **Merged:** No --- ### feat: Add `checkpoint-memgraph` Package #### Description This pull request introduces the `langgraph-checkpoint-memgraph` package, providing a robust and feature-rich integration with Memgraph as a backend for both a key-value store and a checkpoint saver. This implementation allows LangGraph applications to persist and manage state within a Memgraph graph database. The package includes two primary components: 1. **`MemgraphStore`**: A synchronous, `BaseStore`-compliant client for general-purpose data storage in Memgraph. 2. **`MemgraphSaver` / `AsyncMemgraphSaver`**: A `BaseCheckpointSaver`-compliant implementation for persisting LangGraph checkpoints. ### Key Features #### `MemgraphStore` The `MemgraphStore` provides a comprehensive interface for interacting with Memgraph. - **Unified Interface**: Supports key-value storage, document search, and vector similarity search through a single class. - **TTL (Time-To-Live) Support**: Nodes can be set to expire automatically. A background sweeper thread (`start_ttl_sweeper`) can be enabled to periodically clean up expired data, ensuring efficient database management. - **Vector Search**: Integrates with embedding models to store and perform similarity searches on text data. Configuration is handled via the `MemgraphIndexConfig`. - **Robust Connection Management**: Can be instantiated directly with a `memgraph.Driver` instance or conveniently from a connection string using the `from_conn_string` class method. - **Idempotent Setup**: Includes a `setup()` method to create necessary database constraints and indexes, simplifying initial configuration. #### `MemgraphSaver` & `AsyncMemgraphSaver` The checkpoint saver provides a complete checkpointing solution built directly on Memgraph. - **Full Checkpoint Lifecycle**: Implements all methods required by the `BaseCheckpointSaver` interface, including `get_tuple`, `put`, `list`, and `delete_thread`. - **Async and Sync Support**: Provides both `MemgraphSaver` for synchronous operations and `AsyncMemgraphSaver` for asynchronous applications, ensuring seamless integration into any LangGraph workflow. - **Transactional Integrity**: Leverages Memgraph's transactional capabilities to ensure that checkpoint operations are atomic. ### Implementation Details - The package provides two distinct, parallel implementations: `MemgraphStore` for generic key-value operations and `MemgraphSaver` for managing LangGraph checkpoints. - **`MemgraphStore`** persists data in `(:StoreItem)` nodes, offering a flat key-value structure with optional TTL and vector indexing. - **`MemgraphSaver`** implements a graph model where checkpoints are grouped by a shared `thread_id` property. It creates `(:Checkpoint)` nodes to store the state of each run, and the `parent_checkpoint_id` property on these nodes creates a sequential history. Associated data is stored in `(:Blob)` and `(:Write)` nodes, which are also linked by the `thread_id`. - Helper modules (`_internal.py`, `_ainternal.py`) provide shared connection and session management logic for both sync and async operations. - Checkpoints are serialized to JSON before being stored as a property on the `:Checkpoint` node, allowing for efficient storage and retrieval of complex state objects.
yindo added the pull-request label 2026-02-20 17:50:02 -05:00
yindo closed this issue 2026-02-20 17:50:02 -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#4333
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?