Inline SQL

An engine for inline SQL in any Python program.

import pandas as pd
from inline_sql import sql


def head_data(count: int) -> pd.DataFrame:
    return sql << "SELECT * FROM 'cars.csv' LIMIT $count"


cars = head_data(50)

origin_counts = sql << """
    SELECT origin, COUNT(*) FROM $cars
    GROUP BY origin
    ORDER BY count DESC
"""
print(origin_counts)

most_common = origin_counts.origin[0]
print(sql << """
    SELECT AVG(horsepower) FROM $cars
    WHERE origin = $most_common
""")

Operates directly on an in-memory database: access local variables (pandas frames), CSV files, and interpolate values into queries.

Table of Contents

• Installation
• License

Installation

pip install inline-sql

Contributions

Contributions are appreciated, especially in the following areas:

• Additional query engines
• File format support (CSV, Parquet, Arrow)
• Support for dataframe libraries (e.g., Polars)
• Distributed computing (Ray, Dask)
• Documentation

License

inline-sql is distributed under the terms of the MIT license.