Introduction to LangChain Framework

LangChain is an open-source framework designed to help developers build applications based on large language models (LLM). It makes the development process more efficient and convenient through the following core advantages:

1. Data Source Integration: LangChain can seamlessly integrate LLM with real-time databases, APIs, and other multi-source data, ensuring that the generated content is both accurate and contextual.
2. Componentized Process Design: By flexibly configuring task chains, it clearly connects preprocessing, model invocation, and post-processing steps, improving execution efficiency.
3. Simplified Large Model Access: Provides a simple interface, lowering the threshold for technical development using LLM, and quickly realizing complex NLP functions.
4. Highly Extensible Customization: With good scalability, it can meet the needs of different business scenarios, helping developers fully leverage the potential of LLM.

Basic Development Process

This section will demonstrate how to use LangChain and clickzetta-sqlalchemy together to implement a simple application that queries a Lakehouse and displays the results.

Environment Preparation

To interface LangChain with different data sources, you need to install clickzetta-sqlalchemy in your Python environment. The installation method is as follows:

pip install langchain clickzetta-sqlalchemy

Example Code

First, create a file named demo.py and edit the code as follows:

from langchain_community.utilities import SQLDatabase
import streamlit as st

# Get Lakehouse authentication information from Streamlit secret manager
username = st.secrets.lakehouse.username
password = st.secrets.lakehouse.password
account = st.secrets.lakehouse.account
endpoint = st.secrets.lakehouse.endpoint
workspace = st.secrets.lakehouse.workspace
schema = st.secrets.lakehouse.schema
virtualcluster = st.secrets.lakehouse.virtualcluster

# Create connection string
CONNECTION_STRING = (
    f"clickzetta://{username}:{password}"
    f"@{account}.{endpoint}/{workspace}?schema={schema}&virtualcluster={virtualcluster}"
)

# Create SQLDatabase instance from connection string
db = SQLDatabase.from_uri(CONNECTION_STRING, schema=schema)

Next, execute the query and return the query results:

# Execute query
result = db.run("SELECT * FROM Artist LIMIT 12;", fetch="cursor")
# Print result type
print(type(result))
# Display query results
pprint(list(result.mappings()))

To bind query parameters, please use the optional parameters parameter.

result = db.run("SELECT * FROM Artist WHERE Name LIKE :search;",parameters={"search": "p%"},fetch="cursor",)
pprint(list(result.mappings()))

Reference

langchain Official Documentation

SQLDatabase Development Guide