Installation¶
Requirements¶
- Python 3.10 or later (3.10, 3.11, 3.12, 3.13)
- Works in Microsoft Fabric Notebooks out of the box (Fabric uses Python 3.10/3.11)
Core Install¶
Core dependencies (installed automatically): faker, numpy, pandas, click.
Optional Extras¶
Install additional capabilities as needed:
# Parquet output (pyarrow)
pip install sqllocks-spindle[parquet]
# Excel output (openpyxl)
pip install sqllocks-spindle[excel]
# Delta Lake / Fabric Lakehouse output (deltalake + pyarrow)
pip install sqllocks-spindle[fabric]
# Streaming sinks — Event Hub + Kafka (azure-eventhub, kafka-python)
pip install sqllocks-spindle[streaming]
# Inference: FidelityReport + EmpiricalStrategy (scipy)
pip install sqllocks-spindle[inference]
# Fabric Inference: LakehouseProfiler — profile Delta tables directly (scipy + deltalake + pyarrow)
pip install sqllocks-spindle[fabric-inference]
# Fabric SQL Database writer (pyodbc + azure-identity)
pip install sqllocks-spindle[fabric-sql]
# Everything
pip install sqllocks-spindle[all]
# Development (pytest)
pip install sqllocks-spindle[dev]
Fabric Notebooks¶
In a Microsoft Fabric Notebook, install with:
The core dependencies (faker, numpy, pandas) are pre-installed in the Fabric Spark runtime. For Delta output, the Fabric runtime includes PySpark Delta — use sqllocks-spindle[fabric] for the deltalake (delta-rs) writer if you need direct Delta table writes outside of Spark.
Verify Installation¶
from sqllocks_spindle import Spindle, RetailDomain
result = Spindle().generate(domain=RetailDomain(), scale="fabric_demo", seed=42)
print(result.summary())