• Client Library Documentation

• Product Documentation

TheCloud SQL for PostgreSQL for LlamaIndex package provides a first class experience for connecting toCloud SQL instances from the LlamaIndex ecosystem while providing the following benefits:

• Simplified & Secure Connections: easily and securely create shared connection pools to connect to Google Cloud databases utilizing IAM for authorization and database authentication without needing to manage SSL certificates, configure firewall rules, or enable authorized networks.

• Improved metadata handling: store metadata in columns instead of JSON, resulting in significant performance improvements.

• Clear separation: clearly separate table and extension creation, allowing for distinct permissions and streamlined workflows.

Quick Start

In order to use this library, you first need to go through the followingsteps:

1. Select or create a Cloud Platform project.

2. Enable billing for your project.

3. Enable the Cloud SQL Admin API.

4. Setup Authentication.

Installation

Install this library in avirtualenv using pip.virtualenv is a tool to create isolated Python environments. The basic problem it addresses isone of dependencies and versions, and indirectly permissions.

Withvirtualenv, it’spossible to install this library without needing system installpermissions, and without clashing with the installed systemdependencies.

Supported Python Versions

Python >= 3.9

Mac/Linux

pip install virtualenvvirtualenv <your-env>source <your-env>/bin/activate<your-env>/bin/pip install llama-index-cloud-sql-pg

Windows

pip install virtualenvvirtualenv <your-env><your-env>\Scripts\activate<your-env>\Scripts\pip.exe install llama-index-cloud-sql-pg

Example Usage

Code samples and snippets live in thesamples/ folder.

Vector Store Usage

Use a vector store to store embedded data and perform vector search.

import google.authfrom llama_index.core import Settingsfrom llama_index.embeddings.vertex import VertexTextEmbeddingfrom llama_index_cloud_sql_pg import PostgresEngine, PostgresVectorStorecredentials, project_id = google.auth.default()engine = await PostgresEngine.afrom_instance(   "project-id", "region", "my-instance", "my-database")Settings.embed_model = VertexTextEmbedding(   model_name="textembedding-gecko@003",   project="project-id",   credentials=credentials,)vector_store = await PostgresVectorStore.create(   engine=engine, table_name="vector_store")

Chat Store Usage

A chat store serves as a centralized interface to store your chat history.

from llama_index.core.memory import ChatMemoryBufferfrom llama_index_cloud_sql_pg import PostgresChatStore, PostgresEngineengine = await PostgresEngine.afrom_instance(   "project-id", "region", "my-instance", "my-database")chat_store = await PostgresChatStore.create(   engine=engine, table_name="chat_store")memory = ChatMemoryBuffer.from_defaults(   token_limit=3000,   chat_store=chat_store,   chat_store_key="user1",)

Document Reader Usage

A Reader ingest data from different data sources and data formats into a simple Document representation.

from llama_index.core.memory import ChatMemoryBufferfrom llama_index_cloud_sql_pg import PostgresReader, PostgresEngineengine = await PostgresEngine.afrom_instance(   "project-id", "region", "my-instance", "my-database")reader = await PostgresReader.create(   engine=engine, table_name="my-db-table")documents = reader.load_data()

Document Store Usage

Use a document store to make storage and maintenance of data easier.

from llama_index_cloud_sql_pg import PostgresEngine, PostgresDocumentStoreengine = await PostgresEngine.afrom_instance(   "project-id", "region", "my-instance", "my-database")doc_store = await PostgresDocumentStore.create(   engine=engine, table_name="doc_store")

Index Store Usage

Use an index store to keep track of indexes built on documents.

from llama_index_cloud_sql_pg import PostgresIndexStore, PostgresEngineengine = await PostgresEngine.from_instance(   "project-id", "region", "my-instance", "my-database")index_store = await PostgresIndexStore.create(   engine=engine, table_name="index_store")

Contributions

Contributions to this library are always welcome and highly encouraged.

SeeCONTRIBUTING for more information how to get started.

Please note that this project is released with a Contributor Code of Conduct. By participating inthis project you agree to abide by its terms. SeeCode of Conduct for moreinformation.

License

Apache 2.0 - SeeLICENSEfor more information.

Disclaimer

This is not an officially supported Google product.