Skip to content

Commit

Permalink
Initial commit
Browse files Browse the repository at this point in the history
  • Loading branch information
Justin Ramsey committed Dec 13, 2024
0 parents commit f1e3cf1
Show file tree
Hide file tree
Showing 8 changed files with 576 additions and 0 deletions.
5 changes: 5 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
@@ -0,0 +1,5 @@
**/build
**/.venv
**/dist
**/langgraph_checkpoint_dynamodb.egg-info
**/.idea
21 changes: 21 additions & 0 deletions LICENSE
Original file line number Diff line number Diff line change
@@ -0,0 +1,21 @@
MIT License

Copyright (c) 2024 researchwiseai

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
153 changes: 153 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
@@ -0,0 +1,153 @@
# langgraph-checkpoint-dynamodb

Implementation of a LangGraph CheckpointSaver that uses a AWS's DynamoDB

## Inspiration

Based on: https://github.com/researchwiseai/langgraphjs-checkpoint-dynamodb

## Required DynamoDB Tables

To be able to use this checkpointer, two DynamoDB table's are needed, one to store
checkpoints and the other to store writes. Below are some examples of how you
can create the required tables.

### Terraform

```hcl
# Variables for table names
variable "checkpoints_table_name" {
type = string
}
variable "writes_table_name" {
type = string
}
# Checkpoints Table
resource "aws_dynamodb_table" "checkpoints_table" {
name = var.checkpoints_table_name
billing_mode = "PAY_PER_REQUEST"
hash_key = "thread_id"
range_key = "checkpoint_id"
attribute {
name = "thread_id"
type = "S"
}
attribute {
name = "checkpoint_id"
type = "S"
}
}
# Writes Table
resource "aws_dynamodb_table" "writes_table" {
name = var.writes_table_name
billing_mode = "PAY_PER_REQUEST"
hash_key = "thread_id_checkpoint_id_checkpoint_ns"
range_key = "task_id_idx"
attribute {
name = "thread_id_checkpoint_id_checkpoint_ns"
type = "S"
}
attribute {
name = "task_id_idx"
type = "S"
}
}
```

### AWS CDK

```python
from aws_cdk import (
Stack,
aws_dynamodb as dynamodb,
)
from constructs import Construct

class DynamoDbStack(Stack):
def __init__(self, scope: Construct, id: str, **kwargs):
super().__init__(scope, id, **kwargs)

checkpoints_table_name = 'YourCheckpointsTableName'
writes_table_name = 'YourWritesTableName'

# Checkpoints Table
dynamodb.Table(
self,
'CheckpointsTable',
table_name=checkpoints_table_name,
billing_mode=dynamodb.BillingMode.PAY_PER_REQUEST,
partition_key=dynamodb.Attribute(
name='thread_id',
type=dynamodb.AttributeType.STRING,
),
sort_key=dynamodb.Attribute(
name='checkpoint_id',
type=dynamodb.AttributeType.STRING,
),
)

# Writes Table
dynamodb.Table(
self,
'WritesTable',
table_name=writes_table_name,
billing_mode=dynamodb.BillingMode.PAY_PER_REQUEST,
partition_key=dynamodb.Attribute(
name='thread_id_checkpoint_id_checkpoint_ns',
type=dynamodb.AttributeType.STRING,
),
sort_key=dynamodb.Attribute(
name='task_id_idx',
type=dynamodb.AttributeType.STRING,
),
)
```

## Using the Checkpoint Saver

### Default

To use the DynamoDB checkpoint saver, you only need to specify the names of
the checkpoints and writes tables. In this scenario the DynamoDB client will
be instantiated with the default configuration, great for running on AWS Lambda.

```python
from langgraph_checkpoint_dynamodb import DynamoDBSaver
...
checkpoints_table_name = 'YourCheckpointsTableName'
writes_table_name = 'YourWritesTableName'

memory = DynamoDBSaver(
checkpoints_table_name=checkpoints_table_name,
writes_table_name=writes_table_name,
)

graph = workflow.compile(checkpointer=memory)
```

### Providing Client Configuration

If you need to provide custom configuration to the DynamoDB client, you can
pass in an object with the configuration options. Below is an example of how
you can provide custom configuration.

```python
memory = DynamoDBSaver(
checkpoints_table_name=checkpoints_table_name,
writes_table_name=writes_table_name,
client_config={
'region': 'us-west-2',
'accessKeyId': 'your-access-key-id',
'secretAccessKey': 'your-secret-access-key',
}
)
```
Empty file.
Loading

0 comments on commit f1e3cf1

Please sign in to comment.