enable mysql db for sql agent

[cheehook]

Description

Add mysql connector to allow mysql db connection from SQL agent
Before this, sql agent is only enabled and validated with sqlitedb.
This change is needed to support GenAIStudio v1.3 release.

Issues

n/a

Type of change

List the type of change like below. Please delete options that are not relevant.

• New feature (non-breaking change which adds new functionality)

Dependencies

• mysql-connector-python

Tests

Tested by pointing db_path and db_url to AWS RDS MySQL database when running the docker container

Logs in SQL agent container before adding mysql connector
[2025-03-17 05:41:30,596] [ INFO] - comps-react-agent - ========initiating agent============ [2025-03-17 05:41:30,596] [ INFO] - comps-react-agent - args: Namespace(stream=False, port=9096, agent_name='OPEA_Default_Agent', strategy='sql_agent_llama', role_description='LLM enhanced agent', tools=None, recursion_limit=12, require_human_feedback=False, debug=False, model='gpt-4o-mini-2024-07-18', llm_engine='openai', llm_endpoint_url='', max_new_tokens=4096, top_k=10, top_p=0.95, temperature=0.1, repetition_penalty=1.03, return_full_text=False, custom_prompt=None, with_memory=False, memory_type='checkpointer', timeout=60, db_path='mysql+mysqlconnector://admin:password@some_db_instance.us-east-1.rds.amazonaws.com.com:3306/dbname', db_name='dbname', use_hints=False, hints_file=None) Traceback (most recent call last): File "/home/user/comps/agent/src/agent.py", line 43, in <module> agent_inst = instantiate_agent(args) env_config: ['--port', '9096', '--strategy', 'sql_agent_llama', '--llm_endpoint_url', '', '--llm_engine', 'openai', '--model', 'gpt-4o-mini-2024-07-18', '--recursion_limit', '12', '--stream', 'false', '--max_new_tokens', '4096', '--temperature', '0.1', '--with_memory', 'false', '--db_path', 'mysql+mysqlconnector://admin:password@some_db_instance.us-east-1.rds.amazonaws.com.com:3306/dbname', '--db_name', 'dbname', '--use_hints', 'false'] ==========sys_args==========: Namespace(stream=False, port=9096, agent_name='OPEA_Default_Agent', strategy='sql_agent_llama', role_description='LLM enhanced agent', tools=None, recursion_limit=12, require_human_feedback=False, debug=False, model='gpt-4o-mini-2024-07-18', llm_engine='openai', llm_endpoint_url='', max_new_tokens=4096, top_k=10, top_p=0.95, temperature=0.1, repetition_penalty=1.03, return_full_text=False, custom_prompt=None, with_memory=False, memory_type='checkpointer', timeout=60, db_path='mysql+mysqlconnector://admin:password@some_db_instance.us-east-1.rds.amazonaws.com.com:3306/dbname', db_name='dbname', use_hints=False, hints_file=None) Initializing SQL Agent Llama Registered tools: [] @@@@ user defined tools: [] @@@@ Tools: ^^^^^^^^^^^^^^^^^^^^^^^ File "/home/user/comps/agent/src/integrations/agent.py", line 44, in instantiate_agent return SQLAgentLlama(args, with_memory, custom_prompt=custom_prompt) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/user/comps/agent/src/integrations/strategy/sqlagent/planner.py", line 130, in __init__ sql_tool = get_sql_query_tool(args.db_path) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/home/user/comps/agent/src/integrations/strategy/sqlagent/sql_tools.py", line 22, in get_sql_query_tool db = connect_to_db(db_path) ^^^^^^^^^^^^^^^^^^^^^^ File "/home/user/comps/agent/src/integrations/strategy/sqlagent/sql_tools.py", line 9, in connect_to_db db = SQLDatabase.from_uri(db_path) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.11/site-packages/langchain_community/utilities/sql_database.py", line 139, in from_uri return cls(create_engine(database_uri, **_engine_args), **kwargs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "<string>", line 2, in create_engine File "/usr/local/lib/python3.11/site-packages/sqlalchemy/util/deprecations.py", line 281, in warned return fn(*args, **kwargs) # type: ignore[no-any-return] ^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.11/site-packages/sqlalchemy/engine/create.py", line 602, in create_engine dbapi = dbapi_meth(**dbapi_args) ^^^^^^^^^^^^^^^^^^^^^^^^ File "/usr/local/lib/python3.11/site-packages/sqlalchemy/dialects/mysql/mysqlconnector.py", line 83, in import_dbapi from mysql import connector ModuleNotFoundError: No module named 'mysql'

Logs in SQL agent container after adding mysql connector

[2025-03-19 05:04:13,792] [    INFO] - comps-react-agent - ========initiating agent============
[2025-03-19 05:04:13,792] [    INFO] - comps-react-agent - args: Namespace(stream=False, port=9096, agent_name='OPEA_Default_Agent', strategy='sql_agent', role_description='LLM enhanced agent', tools=None, recursion_limit=12, require_human_feedback=False, debug=False, model='gpt-4o-mini-2024-07-18', llm_engine='openai', llm_endpoint_url='', max_new_tokens=4096, top_k=10, top_p=0.95, temperature=0.1, repetition_penalty=1.03, return_full_text=False, custom_prompt=None, with_memory=False, memory_type='checkpointer', timeout=60, db_path='mysql+mysqlconnector://admin:password@some_db_instance.us-east-1.rds.amazonaws.com.com:3306/dbname', db_name='dbname', use_hints=False, hints_file=None)
/home/user/comps/agent/src/integrations/strategy/sqlagent/sql_tools.py:29: LangChainDeprecationWarning: The class `QuerySQLDataBaseTool` was deprecated in LangChain 0.3.12 and will be removed in 1.0. An updated version of the class exists in the :class:`~langchain-community package and should be used instead. To use it run `pip install -U :class:`~langchain-community` and import as `from :class:`~langchain_community.tools import QuerySQLDatabaseTool``.
  db_query_tool = QuerySQLDataBaseTool(db=db, name="sql_db_query", description=query_sql_database_tool_description)
[2025-03-19 05:04:25,412] [    INFO] - Base service - CORS is enabled.
[2025-03-19 05:04:25,412] [    INFO] - Base service - Setting up HTTP server
[2025-03-19 05:04:25,413] [    INFO] - Base service - Uvicorn server setup on port 9096
INFO:     Waiting for application startup.
INFO:     Application startup complete.
INFO:     Uvicorn running on http://0.0.0.0:9096 (Press CTRL+C to quit)
[2025-03-19 05:04:25,416] [    INFO] - Base service - HTTP server setup successful
[2025-03-19 05:24:42,744] [    INFO] - comps-react-agent - -----------NOT STREAMING-------------
env_config:  ['--port', '9096', '--strategy', 'sql_agent', '--llm_endpoint_url', '', '--llm_engine', 'openai', '--model', 'gpt-4o-mini-2024-07-18', '--recursion_limit', '12', '--stream', 'false', '--max_new_tokens', '4096', '--temperature', '0.1', '--with_memory', 'false', '--db_path', 'mysql+mysqlconnector://admin:password@some_db_instance.us-east-1.rds.amazonaws.com.com:3306/dbname', '--db_name', 'dbname', '--use_hints', 'false']
==========sys_args==========:
 Namespace(stream=False, port=9096, agent_name='OPEA_Default_Agent', strategy='sql_agent', role_description='LLM enhanced agent', tools=None, recursion_limit=12, require_human_feedback=False, debug=False, model='gpt-4o-mini-2024-07-18', llm_engine='openai', llm_endpoint_url='', max_new_tokens=4096, top_k=10, top_p=0.95, temperature=0.1, repetition_penalty=1.03, return_full_text=False, custom_prompt=None, with_memory=False, memory_type='checkpointer', timeout=60, db_path='mysql+mysqlconnector://admin:password@some_db_instance.us-east-1.rds.amazonaws.com.com:3306/dbname', db_name='dbname', use_hints=False, hints_file=None)
Initializing SQL Agent
Registered tools:  []
SQL Query Tool Created:  description='Input to this tool is a detailed and correct SQL query, output is a result from the database. If the query is not correct, an error message will be returned. If an error is returned, rewrite the query, check the query, and try again. ' db=<langchain_community.utilities.sql_database.SQLDatabase object at 0x7f1f09698890>
@@@@ ALL Tools:  [QuerySQLDataBaseTool(description='Input to this tool is a detailed and correct SQL query, output is a result from the database. If the query is not correct, an error message will be returned. If an error is returned, rewrite the query, check the query, and try again. ', db=<langchain_community.utilities.sql_database.SQLDatabase object at 0x7f1f09698890>)]
@@@ Initial State:  {'messages': [HumanMessage(content='How many projects are there in the Easydata platform?', additional_kwargs={}, response_metadata={})], 'is_last_step': False, 'hint': ''}
================================ Human Message =================================

How many projects are there in the Easydata platform?
----------Call Agent Node----------
================================== Ai Message ==================================

To determine how many projects are in the Easydata platform, I will query the `project` table, which contains information about the projects. 

Let's execute a SQL query to count the total number of projects in the `project` table. 

Here is the SQL query I will use:

sql
SELECT COUNT(*) AS total_projects FROM project;


Now, I will execute this query.
Tool Calls:
  sql_db_query (call_qObpvz37SoGmG9v72O2mmaDt)
 Call ID: call_qObpvz37SoGmG9v72O2mmaDt
  Args:
    query: SELECT COUNT(*) AS total_projects FROM project;
@@@@ Called Tool:  sql_db_query
@@@@ Going to Query Fixer
================================= Tool Message =================================
Name: sql_db_query

[(3722,)]
----------Call Query Fixer Node----------
@@@@ Executed SQL Query:  SELECT COUNT(*) AS total_projects FROM project;
@@@@ Execution Result:  [(3722,)]