FractalDB - KV Store

A fast, lightweight key-value store designed for hobbyists, combining the simplicity of NoSQL with SQLite3's robust filtering and sorting capabilities.

Features

• NoSQL-like Experience – Work with database records as Python dictionaries.
• Simplified Access – Retrieve data without remembering column names or indices.
• SQL Power – Use SQL queries for filtering and sorting.
• Lightweight & Fast – Minimal overhead with SQLite’s reliability.

Installation

pip install git+https://github.com/jnsougata/fractal.git

Example Usage

from fractal import condition, DB, Schema, Field

if __name__ == "__main__":

    db = DB("example.db")

    users = db.collection(
        name="users",
        schema=Schema(Field("name", str), Field("age", int))
    )

    inserted = users.insert(
        {"name": "Alice", "age": 25},
        {"name": "Bob", "age": 30},
        {"name": "Charlie", "age": 35},
        {"name": "David", "age": 40},
    )

    c1 = condition("name").anyof("Alice", "Bob") and condition("age").between(20, 40)
    c2 = condition("age").between(20, 30) and condition("name").substring("li")

    users = users.query().where(c1 or c2)
    for user in users:
        print(user)

Sample Output:

{'key': '379b87afd3d74f49bc1c59b8185bb534', 'timestamp': '2025-04-08 16:25:13.012322', 'name': 'Alice', 'age': 25}
{'key': '15676bf56c524e5b87db4ee0a659efc7', 'timestamp': '2025-04-08 16:25:13.012322', 'name': 'Bob', 'age': 30}

Documentation

Half-baked docs here.

TODO

• Add more SQL features (e.g., JOIN, GROUP BY).
• Add more complex queries and filtering options.
• Implement view handling.
• Better error handling.
• Composite column support (no flattening).
• Hierarchical column search.
• More examples and documentation.
• Unit tests.
• Performance benchmarks.