feat(redis): enhance checkpoint handling and add example notebooks

  Core changes:
  - Fix Redis key parsing to handle additional components in _parse_redis_checkpoint_writes_key method
  - Modify error handling to accept keys with more than 6 components by slicing (parts[:6])
  - Update error message to indicate "at least 6 parts" needed rather than "expected 6"

  Added comprehensive test suites:
  - Basic key parsing tests (test_key_parsing.py)
  - Subgraph key parsing tests (test_subgraph_key_parsing.py)
  - Fix verification tests (test_fix_verification.py)
  - Semantic search key tests (test_semantic_search_keys.py)
  - Streaming tests (test_streaming_modes.py, test_streaming.py)

  Added new example notebooks:
  - Memory management examples (add-summary-conversation-history, delete-messages, etc.)
  - Human-in-the-loop examples (breakpoints, edit-graph-state, time-travel, etc.)
  - Subgraph management notebooks (subgraph-persistence, subgraphs-manage-state)
  - Agent creation examples with history management and HITL

Author: bsbodden

.gitignore

@@ -220,3 +220,4 @@ pip-selfcheck.json
 libs/redis/docs/.Trash*
 .python-version
 .idea/*
+examples/.Trash*

examples/Dockerfile.jupyter

@@ -5,12 +5,11 @@ RUN useradd -m jupyter
 
 WORKDIR /home/jupyter/workspace
 
-# Copy the library files (only copy the checkpoint-redis directory)
-COPY ./libs/checkpoint-redis /home/jupyter/workspace/libs/checkpoint-redis
+# Copy notebook files to the workspace
+COPY ./ /home/jupyter/workspace/examples/
 
-# Create necessary directories and set permissions
-RUN mkdir -p /home/jupyter/workspace/libs/checkpoint-redis/examples && \
-    chown -R jupyter:jupyter /home/jupyter/workspace
+# Set permissions
+RUN chown -R jupyter:jupyter /home/jupyter/workspace
 
 # Switch to non-root user
 USER jupyter
@@ -21,15 +20,17 @@ ENV PATH="/home/jupyter/venv/bin:$PATH"
 
 # Install dependencies
 RUN pip install --no-cache-dir --upgrade pip && \
-    pip install --no-cache-dir langgraph>=0.3.0 && \
-    pip install --no-cache-dir -e /home/jupyter/workspace/libs/checkpoint-redis && \
-    pip install --no-cache-dir jupyter redis>=5.2.1 redisvl>=0.5.1 langchain-openai langchain-anthropic python-ulid
+    pip install --no-cache-dir "httpx>=0.24.0,<1.0.0" && \
+    pip install --no-cache-dir "langgraph>=0.3.0" && \
+    pip install --no-cache-dir "langgraph-checkpoint-redis>=0.0.4" && \
+    pip install --no-cache-dir jupyter "redis>=5.2.1" "redisvl>=0.5.1" langchain-openai langchain-anthropic python-ulid
+# Note: Notebook-specific dependencies will be installed in the notebook cells as needed
 
 # Set the working directory to the examples folder
-WORKDIR /home/jupyter/workspace/libs/checkpoint-redis/examples
+WORKDIR /home/jupyter/workspace/examples
 
 # Expose Jupyter port
 EXPOSE 8888
 
 # Start Jupyter Notebook with checkpoints disabled
-CMD ["jupyter", "notebook", "--ip=0.0.0.0", "--port=8888", "--no-browser", "--NotebookApp.token=''", "--NotebookApp.password=''", "--NotebookApp.allow_root=True", "--NotebookApp.disable_check_xsrf=True", "--FileContentsManager.checkpoints_kwargs={'root_dir':'/tmp/checkpoints'}"]
+CMD ["jupyter", "notebook", "--ip=0.0.0.0", "--port=8888", "--no-browser", "--ServerApp.token=''", "--ServerApp.password=''", "--ServerApp.allow_root=True", "--NotebookApp.disable_check_xsrf=True", "--FileContentsManager.checkpoints_kwargs={'root_dir':'/tmp/checkpoints'}"]

examples/README.md

@@ -4,7 +4,7 @@ This directory contains Jupyter notebooks demonstrating the usage of Redis with
 
 ## Running Notebooks with Docker
 
-To run these notebooks using the local development version of the Redis checkpoint package:
+To run these notebooks using Docker (recommended for consistent environment):
 
 1. Ensure you have Docker and Docker Compose installed on your system.
 2. Navigate to this directory (`examples`) in your terminal.
@@ -15,21 +15,57 @@ To run these notebooks using the local development version of the Redis checkpoi
    ```
 
 4. Look for a URL in the console output that starts with `http://127.0.0.1:8888/tree`. Open this URL in your web browser to access Jupyter Notebook.
-5. You can now run the notebooks, which will use the local development version of the Redis checkpoint package.
+5. You can now run the notebooks with all dependencies pre-installed.
 
-Note: The first time you run this, it may take a few minutes to build the Docker image.
+Note: 
+- The first time you run this, it may take a few minutes to build the Docker image.
+- The Docker setup uses a simplified structure where the examples are self-contained, making it portable and independent of the repository structure.
 
 To stop the Docker containers, use Ctrl+C in the terminal where you ran `docker compose up`, then run:
 
 ```bash
 docker compose down
 ```
 
+## Running Notebooks Locally
+
+If you prefer to run these notebooks locally without Docker:
+
+1. Make sure you have Redis running locally or accessible from your machine.
+2. Install the required dependencies:
+
+   ```bash
+   pip install langgraph-checkpoint-redis
+   pip install langgraph>=0.3.0 
+   pip install jupyter redis>=5.2.1 redisvl>=0.5.1 
+   pip install langchain-openai langchain-anthropic 
+   pip install python-ulid "httpx>=0.24.0,<1.0.0"
+   
+   # Some notebooks may require additional packages, which will be installed 
+   # within the notebooks themselves when needed
+   ```
+
+3. Set the appropriate Redis connection string in the notebooks.
+4. Launch Jupyter Notebook:
+
+   ```bash
+   jupyter notebook
+   ```
+
+5. Navigate to the notebook you want to run and open it.
+
 ## Notebook Contents
 
-- `persistence_redis.ipynb`: Demonstrates the usage of `RedisSaver` and `AsyncRedisSaver` checkpoint savers with LangGraph.
+- `persistence-functional.ipynb`: Demonstrates the usage of `RedisSaver` and functional persistence patterns with LangGraph.
 - `create-react-agent-memory.ipynb`: Shows how to create an agent with persistent memory using Redis.
 - `cross-thread-persistence.ipynb`: Demonstrates cross-thread persistence capabilities with Redis.
-- `persistence-functional.ipynb`: Shows functional persistence patterns with Redis.
+- `cross-thread-persistence-functional.ipynb`: Shows functional cross-thread persistence patterns with Redis.
+- `create-react-agent-manage-message-history.ipynb`: Shows how to manage conversation history in a ReAct agent with Redis.
+- `subgraph-persistence.ipynb`: Demonstrates persistence with subgraphs using Redis.
+- `subgraphs-manage-state.ipynb`: Shows how to manage state in subgraphs with Redis.
+- `create-react-agent-hitl.ipynb`: Demonstrates human-in-the-loop (HITL) capabilities with Redis.
+- `human_in_the_loop/*.ipynb`: Demonstrates various human-in-the-loop interaction patterns with LangGraph and Redis.
+
+All notebooks have been updated to use the Redis implementation instead of memory implementation, showcasing the proper usage of Redis integration with LangGraph.
 
 These notebooks are designed to work both within this Docker environment (using local package builds) and standalone (using installed packages via pip).