🚀 Quickstart

Here's a quick start guide for using the Minds SQL Core project:

🛠️ Set Up Configuration:

• Ensure you have the necessary environment variables set up in a .env file or your environment configuration. These variables include:
  1. API_KEY: Your API key for authentication with LLMs like OpenAI, Gemini, LLAMA, etc.
  2. DB_URL: The URL or connection string for your database.
  3. EXAMPLE_PATH: The path to any example JSON files or data you may want to use for bulk indexing.

🎬 Initialize a Class:

• To harness the power of MindSQL, begin by creating a Python class that extends three essential components: a database class, a vector store, and an LLm class. A basic MindSQL class could look like this:

  from mindsql.databases import DATABASE
  from mindsql.llms import LLM
  from mindsql.vectorstores import VECTORSTORE
  
  load_dotenv()
  
  api_key = os.getenv('API_KEY')
  db_url = os.getenv('DB_URL')
  example_path = os.getenv('EXAMPLE_PATH')
  
  class MyClass(DATABASE, LLM, VECTORSTORE):
      def __init__(self, config):
          DATABASE.__init__(self, config=config)
          LLM.__init__(self, config=config)
          VECTORSTORE.__init__(self, config=config)

• Once you have created the class you can instantiate it by passing the appropriate configs like this:

  config = {'api_key': api_key}
  
  minds = MyClass(config=config)

• Now, the minds object becomes the primary interface for interacting with the MindSQL package. Let's begin by establishing a connection with our database.

  conn = minds.create_connection(url=db_url)

  With this connection established, you are ready to utilize the functionalities of MindSQL for your project.

📑 Retrieve and Index DDL Statements:

• Use the get_all_ddls method to retrieve all DDL (Data Definition Language) statements from the database.

• Index the retrieved DDL statements into the vector store using the index_ddl method.

  ddls = minds.get_all_ddls(connection=conn, database='YOUR-DATABASE-NAME')
  
  for ind in ddls.index:
      minds.index_ddl(ddls["DDL"][ind])

⚡ Execute Database Queries:

• Use the ask_db method to execute database queries and retrieve results.

• Provide the question, database connection, and optional parameters like table names or visualization preferences.

• The ask_db method can optionally generate charts based on the query results if the visualize parameter is set to True.

  response = minds.ask_db(question="YOUR-QUESTION", connection=conn, visualize=True, table_names=['TABLE_NAME'])

📊 Visualization:

• Visualize the data using the returned chart object.

  chart = response["chart"]
  chart.show()

• Close the DB connection once you are done:

  conn.close()