twat-cache 1.6.0

Flexible caching utilities for Python functions

twat-cache

(work in progress)

A flexible caching utility package for Python functions that provides a unified interface for caching function results using various backends (memory, disk, SQL).

Features

• Simple decorator interface for caching function results
• Multiple caching backends:
  • Memory-based LRU cache (always available)
  • SQL-based disk cache using diskcache (optional)
  • Efficient array caching using joblib (optional)
• Automatic cache directory management
• Type hints and modern Python features

Installation

pip install twat-cache

Usage

Basic usage with default memory caching:

from twat_cache import ucache

@ucache()
def expensive_computation(x):
    # Results will be cached automatically
    return x * x

result = expensive_computation(42)  # Computed
result = expensive_computation(42)  # Retrieved from cache

Using SQL-based disk cache:

@ucache(folder_name="my_cache", use_sql=True)
def process_data(data):
    # Results will be cached to disk using SQL backend
    return data.process()

Using joblib for efficient array caching:

import numpy as np
from twat_cache import ucache

@ucache(folder_name="array_cache")
def matrix_operation(arr):
    # Large array operations will be efficiently cached
    return np.dot(arr, arr.T)

Cache Location

The package automatically manages cache directories using the following strategy:

1. If platformdirs is available, uses the platform-specific user cache directory
2. Otherwise, falls back to ~/.cache

You can get the cache path programmatically:

from twat_cache import get_cache_path

cache_dir = get_cache_path("my_cache")

Dependencies

• Required: None (basic memory caching works without dependencies)
• Optional:
  • platformdirs: For platform-specific cache directories
  • diskcache: For SQL-based disk caching
  • joblib: For efficient array caching

License

MIT License