PyFabricConnect

Unified Python connectivity toolkit for Microsoft Fabric Warehouses, Lakehouses, and SQL Analytics Endpoints.

PyFabricConnect simplifies secure connectivity to Microsoft Fabric using modern authentication methods and multiple connector backends including JDBC, ODBC, and SQLAlchemy.

---

Features

• JDBC connectivity for Spark workloads
• ODBC connectivity using pyodbc
• SQLAlchemy integration
• Service Principal authentication
• OAuth Access Token authentication
• Pandas DataFrame support
• Spark DataFrame support
• Unified authentication abstraction
• Enterprise-ready architecture

---

Supported Technologies

Technology	Supported
Microsoft Fabric Warehouse	✅
Lakehouse SQL Endpoint	✅
Spark JDBC	✅
pyodbc	✅
SQLAlchemy	✅
Pandas	✅
PySpark	✅

---

Installation

Base Installation

pip install pyfabricconnect

Spark Support

Microsoft Fabric environments already include PySpark and an active Spark runtime.

For local Spark environments, install Spark support with:

pip install pyfabricconnect[spark]

Development Environment

pip install pyfabricconnect[dev]

---

Quick Start

Service Principal Authentication

from pyfabricconnect import FabricClient
from pyfabricconnect.auth import ServicePrincipalAuth

auth = ServicePrincipalAuth(
    tenant_id="YOUR_TENANT_ID",
    client_id="YOUR_CLIENT_ID",
    client_secret="YOUR_CLIENT_SECRET"
)

client = FabricClient(
    server="YOUR_SERVER",
    database="WH_PYFABRICCONNECT",
    auth=auth
)

df = client.query_pandas("""
SELECT * FROM dbo.qlbacan
""")

print(df.head())

---

Spark Example

The query_spark() method requires an active SparkSession.

from pyspark.sql import SparkSession

from pyfabricconnect import FabricClient
from pyfabricconnect.auth import ServicePrincipalAuth

spark = SparkSession.builder.getOrCreate()

auth = ServicePrincipalAuth(
    tenant_id="YOUR_TENANT_ID",
    client_id="YOUR_CLIENT_ID",
    client_secret="YOUR_CLIENT_SECRET"
)

client = FabricClient(
    server="YOUR_SERVER",
    database="WH_PYFABRICCONNECT",
    auth=auth,
    spark=spark
)

spark_df = client.query_spark("""
SELECT * FROM dbo.qlbacan
""")

spark_df.show()

---

SQLAlchemy Example

engine = client.engine()

---

Supported Environments

PyFabricConnect is designed to work with:

• Microsoft Fabric Notebooks
• Azure Synapse Analytics
• Databricks
• Local PySpark environments
• Python applications
• Data engineering pipelines

---

Project Structure

pyfabricconnect/
│
├── pyfabricconnect/
│   ├── auth/
│   ├── connectors/
│   ├── core/
│   └── utils/
│
├── tests/
├── examples/
├── README.md
├── LICENSE
├── pyproject.toml
└── .gitignore

---

Project Status

PyFabricConnect is currently under active development.

---

Roadmap

v0.1.0

• JDBC Connector
• ODBC Connector
• SQLAlchemy Connector
• Service Principal Authentication
• OAuth Token Authentication

Future Releases

• Managed Identity
• Token Refresh
• Retry Policies
• Structured Logging
• Async Support
• Connection Pooling

---

Contributing

Contributions, issues, and feature requests are welcome.

---

Support the Project

If PyFabricConnect helps your projects or organization, consider supporting development.

Your support helps maintain:

• Fabric integrations
• Authentication improvements
• New connectors
• Documentation
• Community support

Donations

• PayPal Donations

---

License

Apache License 2.0

Copyright 2026 Eduardo Osorio Venegas - EOSOVNGAS