feat(rnd): CLI to Run and Stop the Server

[Swiftyos]

User description

This is the start of the cli.

Atm it starts and stops the server, storing the running processes pid on .config/agpt

---

PR Type

Enhancement, Tests

---

Description

• Added CLI commands to run and stop the server, including storing and retrieving the server process PID.
• Introduced new block classes and helper functions for agent execution.
• Implemented database connection management and base model class.
• Added execution management classes and functions.
• Enhanced executor with node execution and lifecycle management.
• Enhanced server with async agent execution and integrated database connection management.
• Added async tests for agent execution and graph creation.
• Updated dependencies for testing and code quality.
• Updated Prisma schema to reflect new data models and relationships.

---

Changes walkthrough 📝

	Relevant files
Enhancement	10 files app.py Add background process functionality to server application rnd/autogpt_server/autogpt_server/app.py Added background_process function to run the server in the background. Modified main function to include the new background process. +10/-1    cli.py Implement CLI for running and stopping the server                rnd/autogpt_server/autogpt_server/cli.py Introduced CLI commands to run and stop the server. Added functionality to store and retrieve server process PID. +80/-0    data.py Remove deprecated ExecutionQueue class                                      rnd/autogpt_server/autogpt_server/data.py Removed the ExecutionQueue class and related code. +0/-36    block.py Add block classes and helper functions for agent execution rnd/autogpt_server/autogpt_server/data/block.py Added new block classes for agent execution. Implemented helper functions for block initialization and retrieval. +130/-0  db.py Introduce database connection management and base model    rnd/autogpt_server/autogpt_server/data/db.py Introduced database connection and disconnection functions. Added BaseDbModel class for database models. +28/-0    execution.py Add execution management classes and functions                      rnd/autogpt_server/autogpt_server/data/execution.py Added Execution class and ExecutionQueue class. Implemented functions for managing execution lifecycle. +97/-0    graph.py Add graph and node management classes and functions            rnd/autogpt_server/autogpt_server/data/graph.py Added Node, Edge, and Graph classes. Implemented functions for graph and node management. +166/-0  executor.py Enhance executor with node execution and lifecycle management rnd/autogpt_server/autogpt_server/executor/executor.py Added functions for executing nodes and managing execution lifecycle. Modified executor to use ProcessPoolExecutor. +110/-19 server.py Enhance server with async agent execution and DB management rnd/autogpt_server/autogpt_server/server/server.py Added async support for agent execution. Integrated database connection management. +42/-11  schema.prisma Update Prisma schema for new data models                                  rnd/autogpt_server/schema.prisma Updated schema to reflect new data models and relationships. +27/-47
Tests	1 files test_app.py Add async tests for agent execution and graph creation      rnd/autogpt_server/test/test_app.py Added tests for agent execution and graph creation. Implemented async test functions. +100/-22
Dependencies	1 files pyproject.toml Update dependencies for testing and code quality                  rnd/autogpt_server/pyproject.toml Added new dependencies for testing and code quality. +3/-0

---

💡 PR-Agent usage:
Comment /help on the PR to get a list of all available PR-Agent tools and their descriptions

[netlify]

✅ Deploy Preview for auto-gpt-docs canceled.

Name	Link
🔨 Latest commit	5977668
🔍 Latest deploy log	https://app.netlify.com/sites/auto-gpt-docs/deploys/66681297a43760000823ffb1

[codiumai-pr-agent-pro]

PR Description updated to latest commit (1b04825)

• Copy walkthrough table to "Files Changed" Tab

[codiumai-pr-agent-pro]

PR Review 🔍

⏱️ Estimated effort to review [1-5]	4, because the PR introduces a significant amount of new functionality across multiple files, including CLI commands, server management, execution logic, and database interactions. The complexity is increased due to the integration of asynchronous operations and the handling of subprocesses and multiprocessing. Reviewing this PR requires a deep understanding of the system's architecture and the new features' impact on existing functionalities.
🧪 Relevant tests	Yes
⚡ Possible issues	Possible Bug: The stop command in cli.py reads the PID from a file and directly calls kill on it without checking if the PID is still valid or if it belongs to the server process. This could lead to unintended termination of unrelated processes.
	Error Handling: In executor.py, the function execute_node_sync catches exceptions broadly and logs them, but it does not re-raise them or handle them in a way that the calling function can react appropriately. This might lead to errors being swallowed silently during asynchronous execution.
	Resource Leak: In server.py, the server startup and shutdown events are connected to the database connect and disconnect functions. However, there is no error handling around these operations. If an exception occurs during connection or disconnection, it might leave the database connection in an inconsistent state.
🔒 Security concerns	No

[codiumai-pr-agent-pro]

PR Code Suggestions ✨

Category	Suggestion	Score
Possible issue	Use asyncio.new_event_loop() to avoid potential RuntimeError when no event loop is running The execute_node_sync function uses asyncio.get_event_loop() which can raise a RuntimeError if no event loop is running. It's safer to use asyncio.new_event_loop() and set it as the current event loop. rnd/autogpt_server/autogpt_server/executor/executor.py [97-98] -loop = asyncio.get_event_loop() +loop = asyncio.new_event_loop() +asyncio.set_event_loop(loop) return loop.run_until_complete(execute_node(data)) Apply this suggestion Suggestion importance[1-10]: 8 Why: The suggestion correctly identifies a potential issue with asyncio.get_event_loop() which can indeed raise a RuntimeError if no event loop is running in the current thread. Using asyncio.new_event_loop() and setting it as the current event loop is a safer approach, especially in a multi-threaded environment.	8
	Add exception handling to ensure database operations are successful before updating the queue The add_execution function should handle potential exceptions when interacting with the database to ensure that the queue is not updated if the database operation fails. rnd/autogpt_server/autogpt_server/data/execution.py [51-61] -await AgentNodeExecution.prisma().create( - data={ - "id": execution.id, - "executionId": execution.run_id, - "agentNodeId": execution.node_id, - "executionStatus": ExecutionStatus.QUEUED, - "inputData": json.dumps(execution.data), - "creationTime": datetime.now(), - } -) -return queue.add(execution) +try: + await AgentNodeExecution.prisma().create( + data={ + "id": execution.id, + "executionId": execution.run_id, + "agentNodeId": execution.node_id, + "executionStatus": ExecutionStatus.QUEUED, + "inputData": json.dumps(execution.data), + "creationTime": datetime.now(), + } + ) + return queue.add(execution) +except Exception as e: + # Handle the exception (e.g., log it) + raise e Apply this suggestion Suggestion importance[1-10]: 8 Why: Adding exception handling is crucial to prevent the queue from being updated when a database operation fails, which could lead to inconsistent states.	8
Possible bug	Use dict.get to safely access dictionary values and avoid potential KeyError The get_node_input function should handle the case where latest_executions might not contain all expected nodes, which could lead to a KeyError. rnd/autogpt_server/autogpt_server/data/graph.py [119-121] -name: latest_executions[node_id].outputData +name: latest_executions.get(node_id, {}).get('outputData', None) for name, node_id in node.input_nodes.items() -if node_id in latest_executions Apply this suggestion Suggestion importance[1-10]: 8 Why: This suggestion correctly addresses a potential KeyError by recommending the use of dict.get method, which is a safer way to access dictionary entries. This change would prevent runtime errors in scenarios where latest_executions does not contain all expected node IDs.	8

[github-actions]

This pull request has conflicts with the base branch, please resolve those so we can evaluate the pull request.

[github-actions]

Conflicts have been resolved! 🎉 A maintainer will review the pull request shortly.

[github-actions]

This pull request has conflicts with the base branch, please resolve those so we can evaluate the pull request.

The base branch was changed.

[github-actions]

Conflicts have been resolved! 🎉 A maintainer will review the pull request shortly.