DictDB: A Lightweight In-Memory Database for Python

Why DictDB?

When prototyping or writing tests, you need a data store as agile as your code. SQLite feels heavy; plain dictionaries lack structure and querying power. DictDB bridges that gap: an in-memory, dictionary-backed database with SQL-like CRUD, optional schemas, indexing, and persistence. No server required.

Features

• Zero Configuration: Import and use. No server, no setup.
• Query DSL: Fluent interface with Python operators (table.age >= 18).
• Indexing: Hash indexes for O(1) lookups, sorted indexes for range queries.
• Optional Schemas: Type validation when you need it.
• Persistence: Save/load via JSON or Pickle (sync and async).
• Automatic Backups: Periodic and incremental backup support.
• Thread Safety: Reader-writer locks for concurrent access.
• Minimal Dependencies: Only sortedcontainers required.

Installation

pip install dctdb

Note: The PyPI package is dctdb, but the import is dictdb.

Requirements: Python 3.13+

Quick Example

from dictdb import DictDB, Condition

db = DictDB()
db.create_table("users", primary_key="id")
users = db.get_table("users")

# Insert
users.insert({"name": "Alice", "age": 30, "role": "admin"})
users.insert({"name": "Bob", "age": 25, "role": "user"})

# Select with conditions
admins = users.select(where=Condition(users.role == "admin"))

# Update
users.update({"age": 31}, where=Condition(users.name == "Alice"))

# Delete
users.delete(where=Condition(users.name == "Bob"))

# Persist
db.save("data.json", file_format="json")

Feature Comparison

Resources

• Documentation: https://mhbxyz.github.io/dictdb/
• GitHub: https://github.com/mhbxyz/dictdb
• License: Apache License 2.0